WO2025083024A1

WO2025083024A1 - Improved formulations of autoantigen conjugates

Info

Publication number: WO2025083024A1
Application number: PCT/EP2024/079143
Authority: WO
Inventors: Naomi BENNE; Femke BROERE
Original assignee: Universiteit Utrecht Holding BV
Current assignee: Universiteit Utrecht Holding BV
Priority date: 2023-10-16
Filing date: 2024-10-16
Publication date: 2025-04-24
Anticipated expiration: 2026-04-16

Abstract

The present invention relates to conjugates of a glucocorticoid receptor agonist and an autoantigen. The conjugates are useful for treating or preventing auto-immune diseases such as rheumatoid arthritis, or inflammation. The conjugates can induce tolerogenic dendritic cells and show increased efficiency in vivo as compared to cell therapy.

Description

Improved formulations of autoantigen conjugates

Field of the invention

Background art

Rheumatoid arthritis (RA) is a common auto-immune disease characterized by an influx of pro- inflammatory immune cells into the synovium, leading to pain and disability. Current treatments achieve only temporary remission in a portion of RA patients and constraint to lifelong medication, with progressive loss of therapeutic efficiency and increase of costs. Known treatments include use of disease-modifying antirheumatic drugs (DMARDS), non-steroidal anti-inflammatory agents (NSAIDS), and corticosteroids, although these often bring about off-target side-effects as they are not antigen-specific and are generally immunosuppressive.

Regulatory T cells (Tregs), mainly CD4⁺ T cells, are able to restore immune tolerance by suppressing effector cells in an antigen-specific manner. In patients with autoimmune disease, it is thought that a subtle change in function or presence of Tregs is involved in the pathogenesis of the disease. However, several studies show that the suppressive capacities of Tregs in synovial fluid of RA patients are diminished, while in peripheral blood these capacities are maintained. Other studies indicate a decrease in Treg numbers, which could cause excessive inflammation in RA. Antigen specific Tregs can suppress this excessive inflammation by suppressing the immune cells that cause the pathological autoimmune response while leaving protective immunity intact.

As a tool to induce antigen specific Tregs, tolerogenic dendritic cells (tolDCs) can be used. TolDCs are dendritic cells (DCs) that are modulated to become immune tolerance-inducing. Whether a DC is immune stimulatory or immune tolerant mainly depends on the environmental cues the DCs receive, which determines if they become immunogenic or tolerogenic.

Jansen et al. (doi: 10.3389/fimmu.2019.02068) described the use of dexamethasone(Dex)- pulsed TolDCs loaded with the RA-specific auto-antigen human proteoglycan (hPG), to suppress arthritis. WO2021038283 describes a method of treatment for rheumatoid arthritis which requires isolation of monocytes from the subject, after which first dexamethasone and then autoantigens are added to the cells. The cells are then transferred back to the subject. Like other cell-based therapies, such cell-based therapy is restricted to specialized medical facilities, increasing therapy cost and decreasing accessibility. The need for patient-specific cells also increases patient burden. There is a need for improved treatment of RA. There is a need for medicaments with longer shelf life. There is a need for off-the-shelf medicaments that do not require cell transfer. Summary of the invention

The inventors have surprisingly found that improved induction of immune tolerance could be obtained when an autoantigenic peptide is conjugated to a glucocorticoid receptor agonist. For example, it was found that antigen-corticosteroid complexes can induce antigen-specific immune tolerance, and prevent the development and progression of for instance arthritis in mice in an antigen-specific way. It was found that a representative antigen-dexamethasone complex, hPG- Dex, induces a tolerogenic phenotype in immature dendritic cells of humans and mice.

The invention provides a conjugate comprising:

(i) a glucocorticoid receptor agonist, and

(ii) a peptide that is autoantigenic.

Preferably the glucocorticoid receptor agonist is

(i) a corticosteroid, preferably dexamethasone, triamcinolone, methylprednisolone, prednisolone, or prednisone, more preferably dexamethasone or prednisolone, most preferably dexamethasone, or

(ii) a synthetic nonsteroidal glucocorticoid receptor agonist, preferably dagrocorat, AZD-5423, fosdagrocorat, or mapracorat.

The peptide is preferably an autoantigen associated with an autoimmune disease, such as rheumatoid arthritis, spondyloarthritis, juvenile idiopathic arthritis, psoriatic arthritis, psoriasis, Sjogren disease, systemic lupus erythematosus, dermatomyositis, systemic sclerosis, idiopathic thrombocytopenic purpura, alopecia, vitiligo, type 1 diabetes, myasthenia gravis, multiple sclerosis, Graves’ disease, or autoimmune hepatitis, preferably associated with rheumatoid arthritis or multiple sclerosis, most preferably associated with rheumatoid arthritis. The peptide is preferably derived from a protein selected from human proteoglycan, insulin, insulin precursor, preproinsulin, proinsulin, melanin, topoisomerase 1 , topoisomerase 2, a glutamate decarboxylase such as glutamate decarboxylase 2 or glutamate decarboxylase 65, collagen such as type II collagen, citrullinated human proteoglycan, alpha-enolase, citrullinated alpha-enolase, cartilage intermediate-layer protein, citrullinated cartilage intermediate-layer protein, fibrinogen, citrullinated fibrinogen, vimentin, citrullinated vimentin, acetylcholine receptor, a myelin protein such as myelin oligodendrocyte glycoprotein, myelin proteolipid protein, or myelin basic protein, thyrotropin receptor, and smooth muscle, preferably from human proteoglycan. In preferred embodiments the peptide comprises a sequence that has from 6 to 60, preferably 10 to 40, more preferably from 12 to 20 contiguous amino acids from the sequence of a protein as identified above, wherein the contiguous sequence can have zero, one, two, or three amino acid substitutions. Preferably the peptide has a length of 6 to 70 amino acids, preferably of 10 to 40 amino acids, more preferably of 12 to 20 amino acids. Preferably the peptide comprises or consists of the sequence of any one of SEQ ID NOs: 3-43 and optionally 66-68, preferably SEQ ID NOs: 3-18, more preferably SEQ ID NOs: 3-11 , even more preferably SEQ ID NO: 3.

In some embodiments the glucocorticoid receptor agonist and the peptide are linked through a linker, wherein the linker is preferably a peptide linker comprising from 1 to about 12 amino acids, preferably from 2 to about 8 amino acids, more preferably from about 3 to about 6 amino acids, most preferably 4 or 5 amino acids, wherein preferably the amino acids are hydrophilic. In preferred embodiments the glucocorticoid receptor agonist is dexamethasone, and the peptide is an autoantigen associated with rheumatoid arthritis or with multiple sclerosis, preferably associated with rheumatoid arthritis.

Also provided is a composition comprising the conjugate as described above, wherein the conjugate is comprised in a nanoparticle. The nanoparticle is preferably a micelle, polymeric nanoparticle, polysaccharidic nanoparticle, liposome, lipoplex, unilamellar vesicle, multilamellar vesicle, or a cross-linked or hybrid variant thereof, more preferably it is a liposome. Preferably the nanoparticle comprises one or more phospholipids, preferably two phospholipids, wherein the phospholipids are preferably selected from 1 ,2-dilauroyl-sn-glycero-3-phosphate (DLPA), 1 ,2- dilauroyl-sn-glycero-3- phosphoethanolamine (DLPE), 1 ,2-dimyristoyl-sn-glycero-3-phosphate (DMPA), 1 ,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), 1 ,2-dimyristoyl-sn-glycero-3- phosphoethanolamine (DMPE), 1 ,2-dimyristoyl-sn-glycero-3-phosphoglycerol (DMPG), 1 ,2- dimyristoyl-sn-glycero-3-phosphoserine (DMPS), 1 ,2-dipalmitoyl-sn-glycero-3-phosphate (DPPA),

1 .2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), 1 ,2-dipalmitoyl-sn-glycero-3- phosphoethanolamine (DPPE), 1 ,2-dipalmitoyl-sn-glycero-3-phosphoglycerol (DPPG), 1 ,2- dipalmitoyl-sn-glycero-3-phosphoserine (DPPS), 1 ,2-distearoyl-sn-glycero-3-phosphate (DSPA),

1 .2-distearoyl-sn-glycero-3-phosphocholine (DSPC), 1 ,2-distearoyl-sn-glycero-3- phosphoethanolamine (DSPE), 1 ,2-distearoyl-sn-glycero-3-phosphoglycerol (DSPG), 1 ,2- distearoyl-sn-glycero-3-phosphoserine (DSPS), hydrogenated soy phosphatidylcholine (HSPC),

1 .2-Dioleoyl-sn-glycero-3-phosphate (DOPA), 1 ,2-Dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE), 1 ,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1 ,2-Dioleoyl-sn-glycero-3- phosphoserine (DOPS), 1 ,2-Dilauroyl-sn-glycero-3-phosphocholine (DLPC), or polymer- conjugated phospholipid such as PEGylated phospholipid or polyglycerol phospholipid.

Also provided is the conjugate as described above, or the composition as described above, for use as a medicament. The medicament is preferably for the treatment of an immune or autoimmune disease and/or an inflammation, preferably for the treatment of rheumatoid arthritis, systemic lupus erythematosus, thrombocytopenia, Idiopathic thrombocytopenic purpura (ITP), Systemic Sclerosis, Graves’ Disease, Hashimoto’s thyroiditis, vasculitis, inflammatory bowel disease, multiple sclerosis, hemolytic anemia, thyroiditis, stiff person syndrome, pemphigus vulgaris, myasthenia gravis, lupus nephritis, Crohn’s Disease, or Ulcerative Colitis, more preferably for the treatment of rheumatoid arthritis, type 1 diabetes, myasthenia gravis, multiple sclerosis, Graves’ disease, or autoimmune hepatitis, even more preferably for the treatment of rheumatoid arthritis or multiple sclerosis. In some highly preferred embodiments it is for the treatment of rheumatoid arthritis. In some highly preferred embodiments it is for the treatment of multiple sclerosis. Also provided is a method for treating an autoimmune disease and/or inflammation, the method comprising the step of administering a conjugate or a composition as described above to a subject in need thereof. Description of embodiments

The inventors have surprisingly found that improved induction of immune tolerance could be obtained when an autoantigenic peptide is conjugated to a glucocorticoid receptor agonist. For example, it was found that antigen-corticosteroid complexes can induce antigen-specific immune tolerance, and prevent the development and progression of for instance arthritis in mice in an antigen-specific way. It was found that a representative antigen-dexamethasone complex, hPG- Dex, induces a tolerogenic phenotype in immature dendritic cells of humans and mice. Accordingly the invention provides a conjugate comprising:

(i) a glucocorticoid receptor agonist, and

(ii) a peptide that is autoantigenic.

Such a conjugate is referred to herein as a conjugate according to the invention. The term conjugate has its ordinary meaning in the art, and refers to any entity wherein the two components are bound to each other. Preferably the glucocorticoid receptor agonist and the peptide are covalently bound to each other. In some embodiments the conjugate comprises two or more peptides. In some embodiments the conjugate comprises two or more glucocorticoid receptor agonists. Preferably the conjugate comprises a single peptide and a single glucocorticoid receptor agonist. In preferred embodiments the conjugate consists of a single peptide and a single glucocorticoid receptor agonist. In other preferred embodiments the conjugate consists of a single peptide, a linker, and a single glucocorticoid receptor agonist. The peptide is preferably linear.

In some embodiments the glucocorticoid receptor agonist and the peptide are linked through a linker. Linkers are commonly used and can exist to facilitate the chemical coupling of the glucocorticoid receptor agonist and the peptide (for instance a triazole linker), or to increase the sterical distance between the glucocorticoid receptor agonist and the peptide (for instance an alkane linker), or to impart further properties to the conjugate such as increased water-solubility (for instance using an oligoethylene glycol or polyethylene glycol linker), increased lipophilicity (for instance using an aliphatic linker), increased positive charge (for instance using an oligolysine or polyethylene imine linker), an increased negative charge (for instance using an oligoglutamate or oligoacrylic linker), or any other property known to a skilled person.

In preferred embodiments the linker is a peptide linker. This is advantageous because a peptide linker can be seen as a biodegradable linker. Preferably a peptide linker comprising from 1 to about 12 amino acids, more preferably from 2 to about 8 amino acids, even more preferably from about 3 to about 6 amino acids, most preferably 4 or 5 amino acids. In a peptide linker, preferably the amino acids are hydrophilic. A skilled person is well aware of the properties of amino acids. Examples of hydrophilic amino acids are lysine, serine, threonine, histidine, ornithine, and tyrosine. Further examples of hydrophilic amino acids are aspartic acid and glutamic acid. Hydrophilic peptide linkers can comprise glycine or alanine. Examples of suitable linkers are peptides represented by one of SEQ ID NOs: 54-65, preferably 54-62. In some embodiments the linker is represented by one of SEQ ID NOs: 60-62. In some embodiments the linker is represented by one of SEQ ID NOs: 57-59. In some embodiments the linker is represented by one of SEQ ID NOs: 63- 65. Preferably the linker is represented by one of SEQ ID NOs: 54-56, more preferably SEQ ID NO: 54. Many linkers are suitable - it was found that a variety of linkers did not show differing effects on encapsulation of conjugates in nanoparticles such as liposomes, or on physicochemical properties of the nanoparticles.

When a linker is present, it preferably connects the N-terminus or the C-terminus of the peptide to the glucocorticoid receptor agonist. In preferred embodiments, when a linker is present, it connects the N-terminus of the peptide to the glucocorticoid receptor agonist. Preferably the linker only connects the peptide to the glucocorticoid receptor agonist. Preferably the glucocorticoid receptor agonist is linked to the linker or to the peptide via a reactive group. An ester of an agonist is convenient because the ester can introduce a carboxylic acid to the agonist. An ester can for instance be a succinate ester, to introduce a carboxylic ester that also introduces a free carboxylic acid. Succinate and similar dicarboxylic acids are attractive esters to introduce free carboxylic acids to agonists. These can be conveniently used for conjugation to an N-terminus. The ester is biodegradable and does not remove the activity of the agonist. Examples of suitable dicarboxylic acids are HOOC-(CH2)2-6-COOH, wherein one or two -CH2- moieties may be further substituted, for instance by methyl or methoxy. Preferably there is one, more preferably there are no such further substitutions.

Particularly preferred conjugates are a conjugate wherein the glucocorticoid receptor agonist is dexamethasone, and wherein the peptide is an autoantigen associated with rheumatoid arthritis. In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone succinate, and the peptide is an autoantigen associated with rheumatoid arthritis. In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone, and the peptide is an autoantigen associated with rheumatoid arthritis selected from a peptide consisting of the sequence represented by any one of SEQ ID NOs: 3-18, more preferably SEQ ID NOs: 3-11 , even more preferably SEQ ID NO: 3. In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone succinate, and the peptide is an autoantigen associated with rheumatoid arthritis selected from a peptide consisting of the sequence represented by any one of SEQ ID NOs: 3-18, more preferably SEQ ID NOs: 3-11 , even more preferably SEQ ID NO: 3.

In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone succinate, and the peptide is an autoantigen associated with rheumatoid arthritis selected from a peptide consisting of the sequence represented by any one of SEQ ID NOs: 3-18, more preferably SEQ ID NOs: 3-11 , even more preferably SEQ ID NO: 3, and the dexamethasone succinate is conjugated to the N-terminus of the peptide.

In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone succinate, and the peptide is an autoantigen associated with rheumatoid arthritis selected from a peptide consisting of the sequence represented by any one of SEQ ID NOs: 3-18, more preferably SEQ ID NOs: 3-11 , even more preferably SEQ ID NO: 3, and the dexamethasone succinate is conjugated to the N-terminus of the peptide via a peptide linker.

In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone succinate, and the peptide is an autoantigen associated with rheumatoid arthritis selected from a peptide consisting of the sequence represented by any one of SEQ ID NOs: 3-18, more preferably SEQ ID NOs: 3-11 , even more preferably SEQ ID NO: 3, and the dexamethasone succinate is conjugated to the N-terminus of the peptide via a peptide linker having a length of 3, 4, or 5 amino acids, preferably 4 amino acids, wherein the linker preferably comprises 1 , 2, 3, 4, or 5 lysine residues, preferably 4 lysine residues.

Particularly preferred conjugates are a conjugate wherein the glucocorticoid receptor agonist is dexamethasone, and wherein the peptide is an autoantigen associated with multiple sclerosis. In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone succinate, and the peptide is an autoantigen associated with multiple sclerosis. In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone, and the peptide is an autoantigen associated with multiple sclerosis selected from a peptide consisting of the sequence represented by any one of SEQ ID NOs: 35-41 , more preferably SEQ ID NOs: 35, 38, or 40, even more preferably SEQ ID NO: 35. In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone succinate, and the peptide is an autoantigen associated with multiple sclerosis selected from a peptide consisting of the sequence represented by any one of SEQ ID NOs: 35-41 , more preferably SEQ ID NOs: 35, 38, or 40.

In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone succinate, and the peptide is an autoantigen associated with multiple sclerosis selected from a peptide consisting of the sequence represented by any one of SEQ ID NOs: 35-41 , more preferably SEQ ID NOs: 35, 38, or 40, even more preferably SEQ ID NO: 35, and the dexamethasone succinate is conjugated to the N-terminus of the peptide.

In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone succinate, and the peptide is an autoantigen associated with multiple sclerosis selected from a peptide consisting of the sequence represented by any one of SEQ ID NOs: 35-41 , more preferably SEQ ID NOs: 35, 38, or 40, even more preferably SEQ ID NO: 35, and the dexamethasone succinate is conjugated to the N-terminus of the peptide via a peptide linker.

In other preferred embodiments the glucocorticoid receptor agonist is dexamethasone succinate, and the peptide is an autoantigen associated with multiple sclerosis selected from a peptide consisting of the sequence represented by any one of SEQ ID NOs: 35-41 , more preferably SEQ ID NOs: 35, 38, or 40, even more preferably SEQ ID NO: 35, and the dexamethasone succinate is conjugated to the N-terminus of the peptide via a peptide linker having a length of 3, 4, or 5 amino acids, preferably 4 amino acids, wherein the linker preferably comprises 1 , 2, 3, 4, or 5 lysine residues, preferably 4 lysine residues.

Glucocorticoid receptor agonists

The glucocorticoid receptor is the receptor to which cortisol and other glucocorticoids bind. When an agonist binds to the glucocorticoid receptor, it can regulate gene transcription. An activated receptor can up-regulate the expression of anti-inflammatory proteins in the nucleus or can repress the expression of pro-inflammatory proteins in the cytosol.

Glucocorticoid receptor agonists are known in the art. The glucocorticoid receptor agonist can be a corticosteroid or a synthetic nonsteroidal glucocorticoid receptor agonist, preferably it is a corticosteroid. Derivatives of agonists, such as esters, can also be used. Suitable esters are acetates, propionates, furoates, succinates, or pivalates, preferably succinates. A succinate ester is attractive because it introduces a carboxylic acid moiety that can be used for conjugation to the autoantigenic peptide or to a linker.

Examples of suitable corticosteroids are cortisone, cortisone acetate, cortodoxone, desoxycortone, desoxycortone ester, hydrocortisone, hydrocortisone ester, prebediolone acetate, pregnenolone, pregnenolone acetate, pregnenolone succinate, chloroprednisone, cloprednol, difluprednate, fludrocortisone, flugestone acetate, fluocinolone, fluoromethoIone, fluoromethoIone acetate, fluperolone, fluperolone acetate, fluprednisolone, fluprednisolone ester, loteprednol, medrysone, methylprednisolone, methylprednisolone ester, prednicarbate, prednisolone, prednisone, tixocortol, tixocortol pivalate, alclometasone, beclomethasone, beclomethasone ester, betamethasone, betamethasone ester, clobetasol, clobetasol propionate, clobetasone, clocortolone, clocortolone ester, cortivazol, desoximetasone, dexamethasone, dexamethasone ester, diflorasone, diflucortolone, diflucortolone valerate, fluclorolone, flumetasone, fluocortin, fluocortolone, fluocortolone ester, fluprednidene acetate, fluticasone, fluticasone furoate, fluticasone propionate, halometasone, meprednisone, mometasone, mometasone furoate, paramethasone, prednylidene, rimexolone, triamcinolone, ulobetasol, amcinonide, budesonide, ciclesonide, deflazacort, desonide, fluclorolone acetonide, fludroxycortide, flunisolide, fluocinolone acetonide, fluocinonide, formocortal, halcinonide, triamcinolone acetonide, and triamcinolone acetonide ester.

Preferred corticosteroids are alclometasone, beclomethasone, beclomethasone ester, betamethasone, betamethasone ester, clobetasol, clobetasol propionate, clobetasone, clocortolone, clocortolone ester, cortivazol, desoximetasone, dexamethasone, dexamethasone ester, diflorasone, diflucortolone, diflucortolone valerate, fluclorolone, flumetasone, fluocortin, fluocortolone, fluocortolone ester, fluprednidene acetate, fluticasone, fluticasone furoate, fluticasone propionate, halometasone, meprednisone, mometasone, mometasone furoate, paramethasone, prednylidene, rimexolone, triamcinolone, and ulobetasol, more preferably dexamethasone, triamcinolone, methylprednisolone, prednisolone, or prednisone, even more preferably dexamethasone or prednisolone, most preferably dexamethasone such as dexamethasone succinate, preferably 4-{[(11 p,16a)-9-Fluoro-11 ,17-dihydroxy-16-methyl-3,20- dioxopregna-1 ,4-dien-21-yl]oxy}-4-oxobutanoic acid. In some embodiments the corticosteroid is prednisolone. In some embodiments the corticosteroid is dexamethasone.

Examples of suitable synthetic nonsteroidal glucocorticoid receptor agonists are dagrocorat (also known as PF-251802), AZD-5423 (CAS: 1034148-04-3), GSK-9027 (CAS: 1229096-88-1), fosdagrocorat (also known as PF-4171327), or mapracorat (also known as ZK-245186). Preferred nonsteroidal agonists are dagrocorat, AZD-5423, fosdagrocorat, and mapracorat.

Autoantigenic peptide

The conjugate according to the invention comprises a peptide that is autoantigenic. Autoimmunity is the system of immune responses of an organism against its own healthy cells, tissues, or other normal body constituents. Conditions resulting from this type of immune response are known as autoimmune diseases, which can lead to for instance tissue damage, inflammation, and pain. Autoimmunity means presence of antibodies or T cells that react with self-protein that is present in all individuals, even in normal health state. Many autoimmune diseases are well known, and their mechanism understood. Thus, many autoantigens are known, including autoantigenic peptides. A skilled person is aware of autoimmune diseases and their associated peptides that are autoantigenic. An autoantigenic peptide can be referred to as an autoantigen.

In preferred embodiments, the peptide is an autoantigen associated with an autoimmune disease, such as rheumatoid arthritis, spondyloarthritis, juvenile idiopathic arthritis, psoriatic arthritis, psoriasis, Sjogren disease, systemic lupus erythematosus, dermatomyositis, systemic sclerosis, idiopathic thrombocytopenic purpura, alopecia, vitiligo, type 1 diabetes, myasthenia gravis, multiple sclerosis, Graves’ disease, or autoimmune hepatitis, preferably associated with rheumatoid arthritis or multiple sclerosis, more preferably associated with rheumatoid arthritis. It should be understood that the peptide is not necessarily autologous, it can also be a peptide that induces an autoantigenic response. This is for instance the case for celiac disease, where the ingestion of wheat gluten, and of homologous proteins of barley and rye, induces pronounced T cell-mediated inflammatory reactions. Thus in some embodiments the autoantigenic peptide is autologous, or the peptide is heterologous and induces an autoantigenic response. In other embodiments the autoantigenic peptide is autologous. In other embodiments the peptide is heterologous and induces an autoantigenic response.

The peptide is preferably a peptide consisting of naturally occurring proteinogenic amino acids. It can be modified as occurring in an organism, for instance it can be citrullinated. Preferably it is not otherwise modified. Preferably it is not modified. The termini of the peptide are preferably free termini, or are amidated, or acetylated, or methylated. Notably it is preferred that one of the termini is used as a reactive handle for the conjugation of the glucocorticoid receptor agonist.

The peptide can conveniently be derived from a protein known to be associated with an autoimmune disease. For example, the peptide can be derived from a protein selected from human proteoglycan, insulin, insulin precursor, preproinsulin, proinsulin, melanin, topoisomerase 1 , topoisomerase 2, a glutamate decarboxylase such as glutamate decarboxylase 2 or glutamate decarboxylase 65, collagen such as type II collagen, citrullinated human proteoglycan, alphaenolase, citrullinated alpha-enolase, cartilage intermediate-layer protein, citrullinated cartilage intermediate-layer protein, fibrinogen, citrullinated fibrinogen, vimentin, citrullinated vimentin, acetylcholine receptor, a myelin protein such as myelin oligodendrocyte glycoprotein, myelin proteolipid protein, or myelin basic protein, thyrotropin receptor, and smooth muscle. Preferably the peptide is derived from a protein selected from human proteoglycan, insulin, insulin precursor, preproinsulin, proinsulin, acetylcholine receptor, a myelin protein such as myelin oligodendrocyte glycoprotein, myelin proteolipid protein, or myelin basic protein. More preferably the peptide is derived from a protein selected from human proteoglycan, insulin, acetylcholine receptor, or myelin oligodendrocyte glycoprotein. Most preferably the peptide is derived from human proteoglycan. In other highly preferred embodiments the peptide is derived from myelin oligodendrocyte glycoprotein.

A peptide derived from a protein is preferably a peptide that comprises or consists of a contiguous amino acid sequence from the source protein. Preferably the peptide comprises or consists of a sequence that has from 6 to 70, preferably from 7 to 60, more preferably 10 to 40, still more preferably from 12 to 20 contiguous amino acids from the sequence of a protein known to be associated with an autoimmune disease, wherein the contiguous sequence can have zero, one, two, or three amino acid substitutions. In some embodiments the peptide comprises or consists of a sequence that has from 8 to 50, preferably 9 to 35, more preferably from 11 to 30 contiguous amino acids, with optional substitutions as described herein.

In some embodiments the contiguous sequence can have one, two, or three amino acid substitutions. In some embodiments the contiguous sequence can have zero, one, or two amino acid substitutions. In some embodiments the contiguous sequence can have zero or one amino acid substitutions. In some embodiments the contiguous sequence can have one amino acid substitutions. In some embodiments the contiguous sequence does not have amino acid substitutions. In preferred embodiments the peptide consists of the contiguous sequence.

In preferred embodiments, the peptide comprises or consists of the sequence of any one of SEQ ID NOs: 3-43. In other preferred embodiments, the peptide comprises or consists of the sequence of any one of SEQ ID NOs: 3-43 or 66-68, preferably SEQ ID NOs: 3-18, more preferably SEQ ID NOs: 3-11 , even more preferably SEQ ID NO: 3. In other preferred embodiments the peptide comprises or consists of the sequence of any one of SEQ ID NOs: 3-11 , 19, 20, 22, 35, 38, 40, 33, or 34.

In some embodiments the peptide is derived from human proteoglycan, type II collagen, citrullinated human proteoglycan, citrullinated alpha-enolase, citrullinated cartilage intermediatelayer protein, citrullinated fibrinogen, or citrullinated vimentin. Such peptides are associated with Rheumatoid arthritis. Such peptides preferably comprise or consist of a sequence represented by any one of SEQ ID NOs: 3-18, more preferably 3-11 , even more preferably 3.

In some embodiments the peptide is derived from insulin, proinsulin, insulin precursor, preproinsulin, glutamate decarboxylase 2, or glutamate decarboxylase 65. Such peptides are associated with Type 1 diabetes. Such peptides preferably comprise or consist of a sequence represented by any one of SEQ ID NOs: 19-28, more preferably SEQ ID NO: 19.

In some embodiments the peptide is derived from acetylcholine receptor. Such peptides are associated with myasthenia gravis. Such peptides preferably comprise or consist of a sequence represented by any one of SEQ ID NOs: 33-34, more preferably SEQ ID NO: 33.

In some embodiments the peptide is derived from myelin oligodendrocyte glycoprotein, myelin basic protein, or myelin proteolipid protein. Such peptides are associated with multiple sclerosis. Such peptides preferably comprise or consist of a sequence represented by any one of SEQ ID NOs: 35-41 . More preferably it is associated with myelin oligodendrocyte glycoprotein, such as SEQ ID NOs: 35, 38, or 40, even more preferably SEQ ID NO: 35. In some embodiments the peptide is derived from thyrotropin receptor. Such peptides are associated with Graves disease. Such peptides preferably comprise or consist of a sequence represented by SEQ ID NOs: 42.

In some embodiments the peptide is derived from smooth muscle. Such peptides are associated with autoimmune hepatitis. Such peptides preferably comprise or consist of a sequence represented by SEQ ID NOs: 43.

In some embodiments the peptide is derived from a gliadin, such as a-gliadin, y-gliadin, or w-gliadin. Such peptides are associated with celiac disease. Such peptides preferably comprise or consist of a sequence represented by SEQ ID NOs: 66-68. In these SEQ ID NOs it is preferred that 1 , 2, or 3 Q residues are deamidated by tissue transglutaminase (TG2). Further peptides associated with celiac disease are known in the art, see for instance J Immunol (2009) 182 (7): 4158-4166.

Examples of suitable proteins known to be associated with an autoimmune disease, as well as suitable peptides, are shown in the following table.

The peptide preferably has a length of about 6 to about 70 amino acids, more preferably of about 10 to about 40 amino acids, more preferably of about 12 to about 20 amino acids. The peptide can have a length of about 7 to about 35, about 8 to about 30, about 9 to about 25, about 10 to about 23, about 11 to about 22, about 13 to about 21 , about 14 to about 19, about 15 to about 18, or about 16 to about 17 amino acids. Compositions

The invention also provides a composition comprising a conjugate according to the invention and a pharmaceutically acceptable excipient. Such a composition is referred to herein as a composition according to the invention. Preferably, such a composition is formulated as a pharmaceutical composition. A preferred excipient is water, preferably purified water, more preferably ultrapure water. In other embodiments the water is part of a pharmacologically acceptable buffer such as saline, buffered saline, or more preferably phosphate buffered saline. A preferred buffer is 10 mM 4-(2-hydroxyethyl)-1 -piperazineethanesulfonic acid (HEPES) at pH 7-8 such as pH 12.

A composition preferably has a physiologically acceptable pH, more preferably in the range of 6 to 8, or 7 to 7.8. Further preferred excipients are adjuvants, binders, desiccants, or diluents. Further preferred compositions additionally comprise additional medicaments for treating conditions as described elsewhere herein, or for treating pain or inflammation. Preferred additional medicaments in this regards are immunotherapeutic agents or steroids. Compositions according to the invention preferably contain a therapeutically effective amount of the compound according to the invention.

It was found that the compositions are advantageous when the conjugate is comprised in a nanoparticle. Suitable compositions are those wherein the nanoparticle is a micelle, polymeric nanoparticle, polysaccharidic nanoparticle, liposome, lipoplex, unilamellar vesicle, multilamellar vesicle, or a cross-linked or hybrid variant thereof, preferably it is a liposome.

Micelles are preferably lipid micelles or polymer micelles, most preferably lipid micelles. Liposomes can be unilamellar or multilamellar, preferably unilamellar. Examples of suitable polymeric nanoparticles are beads or polymersomes, for instance polymeric nanoparticles based on PLGA, PLGA-N-trimethyl chitosan, PLA, PLA-PEMA, PLGA-PEG, PLGA-PEMA, or PLA-PEG, or optionally on mixtures thereof. Examples of suitable polysaccharidic nanoparticles are dextran nanoparticles, maltodextrin nanoparticles, and hyaluronic acid nanoparticles. An exemplary hybrid particle is a lipid-polymer hybrid nanoparticle.

The nanoparticles preferably comprise lipids, more preferably phospholipids. In preferred embodiments the nanoparticles comprise one or more phospholipids, more preferably two or more phospholipids. Good results were achieved when the conjugates are comprised in liposomes.

Phospholipids preferably contain a diglyceride, a phosphate group and a simple organic molecule such as choline. In particular, phospholipids include phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidic acid (PA), phosphatidylinositol (PI), phosphatidylserine (PS), sphingomyelin, plasmalogens, and phosphatidylcholine lipid derivatives where the two hydrocarbon chains are typically between about 12-26, preferably 14-22 carbon atoms in length, and can have varying degrees of unsaturation.

The phospholipid may comprise a net negative electrical charge or a net positive electrical charge or may be neutral. In a preferred embodiment, one or more phospholipids are neutral phospholipids. A neutral phospholipid is herein understood as a phospholipid that has no net electrical charge. In a more preferred embodiment of the invention the nanoparticle is a liposome that has a net negative charge. Most preferably a nanoparticle is a liposome and comprises a neutral phospholipid and a negatively charged phospholipid. Preferably the molar ratio of neutral phospholipid to negatively charged phospholipid is in the range of 10:1 to 1 :2, preferably in the range of 8:1 to 1 :1 , more preferably in the range of 6:1 to 2:1 , still more preferably in the range of 5:1 to 3:1 , such as 4:1 .

Preferably the nanoparticle comprises one or more phospholipids, preferably two phospholipids, wherein the phospholipids are preferably selected from 1 ,2-dilauroyl-sn-glycero-3- phosphate (DLPA), 1 ,2-dilauroyl-sn-glycero-3- phosphoethanolamine (DLPE), 1 ,2-dimyristoyl-sn- glycero-3-phosphate (DMPA), 1 ,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), 1 ,2- dimyristoyl-sn-glycero-3-phosphoethanolamine (DMPE), 1 ,2-dimyristoyl-sn-glycero-3- phosphoglycerol (DMPG), 1 ,2-dimyristoyl-sn-glycero-3-phosphoserine (DMPS), 1 ,2-dipalmitoyl-sn- glycero-3-phosphate (DPPA), 1 ,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), 1 ,2- dipalmitoyl-sn-glycero-3-phosphoethanolamine (DPPE), 1 ,2-dipalmitoyl-sn-glycero-3- phosphoglycerol (DPPG), 1 ,2-dipalmitoyl-sn-glycero-3-phosphoserine (DPPS), 1 ,2-distearoyl-sn- glycero-3-phosphate (DSPA), 1 ,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), 1 ,2-distearoyl- sn-glycero-3-phosphoethanolamine (DSPE), 1 ,2-distearoyl-sn-glycero-3-phosphoglycerol (DSPG), 1 ,2-distearoyl-sn-glycero-3-phosphoserine (DSPS), hydrogenated soy phosphatidylcholine (HSPC), 1 ,2-dioleoyl-sn-glycero-3-phosphate (DOPA), 1 ,2-dioleoyl-sn-glycero-3- phosphoethanolamine (DOPE), 1 ,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1 ,2-dioleoyl- sn-glycero-3-phosphoserine (DOPS), 1 ,2-dilauroyl-sn-glycero-3-phosphocholine (DLPC), or polymer-conjugated phospholipid such as PEGylated phospholipid or polyglycerol phospholipid. Particularly preferred phospholipids are DSPC and DSPG.

Preferred nanoparticles further comprise a sterol such as cholesterol. In some embodiments the nanoparticle, preferably a liposome, comprises phospholipid and a sterol, preferably cholesterol. Preferably the molar ratio of phospholipid to sterol is in the range of 1 :2 to 5:1 , preferably in the range of 1 :1 to 4:1 , more preferably in the range of 3:2 to 3:1 , still more preferably in the range of 2:1 to 3:1 , such as 5:2.

When a neutral phospholipid, a negatively charged phospholipid, and a sterol are comprised in a nanoparticle, their respective molar ratio is preferably in the range of 2:1 :1 to 8:1 :4, more preferably in the range of 6:2:3 to 6:1 :3, most preferably in the range of 6:2:3 to 5:1 :3, such as 4:1 :2. Highly preferred nanoparticles are liposomes, comprising or consisting of 1 ,2-distearoyl-sn-glycero- 3-phosphocholine and 1 ,2-distearoyl-sn-glycero-3-phosphoglycerol and cholesterol in a 4:1 :2 molar ratio.

The nanoparticles, preferably the liposomes, preferably have a size of 10-1000 nm, more preferably of 50-500 nm, more preferably of 70-400 nm, more preferably of 90-350 nm, still more preferably of 100-300 nm, more preferably of 110-290 nm, more preferably of 125-280 nm, more preferably of 150-270 nm, more preferably of 170-260 nm, most preferably of 180-220 nm. The size of a nanoparticle is preferably determined using microscopy or light scattering, more preferably light scattering, for instance such as described in the examples. The nanoparticles, preferably the liposomes, preferably have a polydispersity of about 0 to 0.15, more preferably of about 0.02 to about 0.14, more preferably of about 0.03 to about 0.13, particularly preferably of about 0.04 to about 0.12. In some embodiments the polydispersity is about 0.04 to about 0.11. In some embodiments the polydispersity is about 0.05 to about 0.11 . In some embodiments the polydispersity is about 0.03 to about 0.07.

The nanoparticles, preferably the liposomes, preferably have a zeta-potential of from -70 to -30 mV, more preferably of from -60 to -40 mV, more preferably of from -58 to -45 mV, more preferably of from -56 to -48 mV, even more preferably of from -55 to -50 mV, most preferably of from -52 to -55 mV. In some embodiments the nanoparticles, preferably the liposomes, have a zetapotential of from -55 to -60 mV. In some embodiments the nanoparticles, preferably the liposomes, have a zeta-potential of from -50 to -60 mV.

In embodiments wherein the conjugate is enclosed in a nanoparticle, a water soluble polymer can be conjugated to the nanoparticle. Preferably, the water soluble polymer is conjugated to a lipid such as a phospholipid or cholesterol.

In an embodiment of the invention, the water soluble polymer is at least one of: i) a polyalkylether, preferably the polyalkylether is linear polyethylene glycol (PEG), star PEG or multiarm branched PEG; ii) a homopolymer that is a PEG substitute or a PEG alternative, preferably the homopolymer is selected from the group consisting of polymethylethyleneglycol (PMEG), polyhydroxypropyleneglycol (PHPG), polypropyleneglycol (PPG), polymethylpropyleneglycol (PMPG), polyhydroxypropyleneoxide (PHPO), poly-oxazoline (POZ) and hydroxyethyl starch (HES); iii) a heteropolymer of small alkoxy monomers, preferably the heteropolymer comprises polyethyleneglycol /polypropyleneglycol (PEG/PPG). Preferably, the water soluble polymer has a molecular weight of at least about 120 Daltons and a polymerization number of at least 6 or of about 6 - 210. Several lipid-PEG conjugates are commercially available.

In a particular embodiment of the invention, the conjugate of the biocompatible lipid and the water soluble polymer is a conjugate of a polymer as defined above to a phospholipid as defined above, or to vitamin E or a derivative of vitamin E. Preferably, the conjugate is a phospholipid-PEG conjugate such as 1 ,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-polyethylene glycol (DSPE-PEG). More preferably, the conjugate is 1 ,2-distearoyl-sn-glycero-3-phosphoethanolamine- N-[methoxy(polyethylene glycol)-2000 (DSPE-mPEG2000) or d-alpha tocopheryl-N- [methoxy(polyethylene glycol)-1000 (TPEG1000).

Preferably, the surface of the nanoparticle is at least partly covered by the water soluble polymer. More preferably, the water soluble polymer covers the surface of the nanoparticle for at least about 10, 20, 30, 40, 50, 60, 70, 80, 90, 99 or 100%. In a further embodiment, the nanoparticle has a surface consisting of a water soluble polymer. Covering of a nanoparticle is preferably covering the solvent-accessible surface of a nanoparticle.

It was found that the conjugates can be encapsulated in nanoparticles, particularly in liposomes, with a good degree of encapsulation efficiency. Notably the encapsulation efficiency of conjugates is within a narrower bandwidth as compared to encapsulation efficiency of unconjugated peptides (see Fig. 13). Accordingly, provided is a method for encapsulating a peptide in a nanoparticle, the method comprising the step of forming a conjugate of the peptide wherein the conjugate is as described herein, followed by the step of forming the nanoparticle. Preferably in this method the nanoparticle is a liposome, more preferably it is a liposome that comprises a neutral phospholipid and a negatively charged phospholipid, even more preferably it is a liposome as described above. The method is preferably for encapsulating at least 30%, more preferably at least 35%, even more preferably at least 40% of the peptide in the nanoparticles. The peptide is preferably as described elsewhere herein, more preferably it has a net charge at pH 7 of at most 8, preferably at most 7, more preferably at most 4, still more preferably at most 3.

Uses and methods

The conjugate according to the invention, or the composition according the invention, are suitable for use as a medicament. The medicament can be for the treatment of an immune or an autoimmune disease and/or inflammation, preferably for the treatment of an autoimmune disease and/or inflammation, more preferably an autoimmune disease. Suitable examples of an immune or autoimmune disease and/or inflammation are rheumatoid arthritis, neuropathy, rhinitis, systemic lupus erythematosus, thrombocytopenia, Idiopathic thrombocytopenic purpura (ITP), anaphylaxis, transplantation rejection, Graft Versus Host Disease, Systemic Sclerosis, Atopic Dermatitis, Graves’ Disease, Hashimoto’s thyroiditis, vasculitis, Omen’s syndrome, chronic renal failure, inflammatory bowel disease, Crohn’s Disease, Ulcerative Colitis, Celiac’s Disease, diabetes mellitus, acute infectious mononucleosis, HIV, herpes virus associated diseases, multiple sclerosis, hemolytic anemia, thyroiditis, stiff person syndrome, pemphigus vulgaris, myasthenia gravis, or lupus nephritis, more preferably for the treatment of rheumatoid arthritis, type 1 diabetes, myasthenia gravis, multiple sclerosis, Graves’ disease, or autoimmune hepatitis, even more preferably for the treatment of rheumatoid arthritis, type 1 diabetes, myasthenia gravis, or multiple sclerosis. Good results were achieved for the treatment of rheumatoid arthritis. Good results were achieved for the treatment of multiple sclerosis.

Suitable examples of an autoimmune disease and/or inflammation are rheumatoid arthritis, systemic lupus erythematosus, thrombocytopenia, Idiopathic thrombocytopenic purpura (ITP), Systemic Sclerosis, Graves’ Disease, Hashimoto’s thyroiditis, vasculitis, inflammatory bowel disease, multiple sclerosis, hemolytic anemia, thyroiditis, stiff person syndrome, pemphigus vulgaris, myasthenia gravis, lupus nephritis, Crohn’s Disease, or Ulcerative Colitis, preferably rheumatoid arthritis, systemic lupus erythematosus, thrombocytopenia, Idiopathic thrombocytopenic purpura (ITP), Systemic Sclerosis, Graves’ Disease, Hashimoto’s thyroiditis, vasculitis, inflammatory bowel disease, multiple sclerosis, hemolytic anemia, thyroiditis, stiff person syndrome, pemphigus vulgaris, myasthenia gravis, lupus nephritis, more preferably rheumatoid arthritis, type 1 diabetes, myasthenia gravis, multiple sclerosis, Graves’ disease, or autoimmune hepatitis, even more preferably for the treatment of rheumatoid arthritis, type 1 diabetes, myasthenia gravis, or multiple sclerosis. Good results were achieved for the treatment of rheumatoid arthritis. Good results were achieved for the treatment of multiple sclerosis. The invention provides long-lasting systemic effects. A preferred use is for lowering the proportion of CD1 1 c⁺CD86⁺ dendritic cells in a subject, preferably as compared to PBS treatment. A preferred use is for increasing the amount of CD4⁺CD25⁺Foxp3⁺ cells, or for increasing the amount of CD4⁺PD-1⁺ T cells, or for increasing the amount of both CD4⁺CD25⁺Foxp3⁺ cells and the amount of CD4⁺PD-1⁺ T cells. This increase is preferably in spleens. A preferred use is for lowering the amount of I gG 1 or lgG2 or of both I gG 1 and lgG2 that are specific for the autoantigenic peptide comprised in the conjugate. Particularly the compositions wherein the conjugate is comprised in a liposome are suitable for use as a medicament for lowering the amounts of IgG 1 that is specific for the autoantigenic peptide comprised in the conjugate. A preferred use is such lowering of lgG1 without lowering lgG2. A further preferred use is for increasing expression of IL10. A further preferred use is for increasing gene expression of IDO. A further preferred use is for increasing protein expression of LAP. Particularly preferred is the use for increasing expression of IL10, increasing gene expression of IDO, and increasing protein expression of LAP. A further preferred use is for inducing antigen-specific CD49b⁺LAG-3⁺ Tr1 cells. A further preferred use is for inducing phenotypical tolDCs with a cytokine secretion and gene expression profile that can promote tolerance in T cells. A preferred use is for inducing tolerogenic DCs. A further preferred use is for inducing antigen-specific Tregs. The above uses offer additional therapeutic strategies for treating or preventing or ameliorating autoimmune diseases, possibly when combined with

The invention also provides a method for treating an autoimmune disease and/or inflammation, the method comprising the step of administering a conjugate or a composition according to the invention to a subject in need thereof. Preferably an effective dose is administered. The use can be for treatment, for reduction of symptoms, for prevention, or for amelioration of the disease and/or inflammation. Preferably at least one symptom is treated, reduced, prevented, or ameliorated. Symptoms can be pain or swelling.

In general, the conjugates and compositions according to the invention may be administered orally or via a parenteral route, usually injection or infusion. A “parenteral administration route” means modes of administration other than enteral and topical administration, usually by injection, and includes intravenous, intramuscular, intraarterial, intrathecal, intracapsular, intraorbital, intracardiac, intradermal, intraperitoneal, transtracheal, subcutaneous, subcuticular, intraarticluare, subcapsular, subarachnoid, intraspinal and intrasternal injection and infusion.

A subject is preferably a mammal, such as a primate, a cow, a horse, a camel, a dog, a cat, or a mouse. Primates are particularly preferred subjects, such as humans or non-human primates. Most preferably a subject is human. The subject is preferably a subject in need of treatment, such as a subject suffering from the autoimmune disease or inflammation, or a subject susceptible to developing the autoimmune disease or inflammation. Need of treatment can be need to cure or alleviate symptoms of an autoimmune disease or of inflammation, but it can also be prophylactic treatment. In preferred embodiments, treatment is primary prophylactic treatment for the prevention of the onset of symptoms of the autoimmune disease. The inventors found that the invention is also particularly suited for secondary prophylaxis, so in other preferred embodiments, treatment is secondary prophylactic treatment for the prevention of recurrence of symptoms after earlier treatment. In some embodiments, a subject is a young subject, preferably a juvenile subject, more preferably a newborn subject. In other preferred embodiments, a subject is elderly. An elderly subject is preferably over 50 years of age, more preferably over 60, even more preferably over 65, more preferably still over 70, most preferably over 75. Alternately the subject can be over 30, preferably over 35, more preferably over 40, most preferably over 45 years of age,

General definitions

Each embodiment as identified herein may be combined together unless otherwise indicated. All patent and literature references cited in the present specification are hereby incorporated by reference in their entirety.

In the context of this invention, a decrease or increase of a parameter to be assessed means a change of at least 5% of the value corresponding to that parameter. More preferably, a decrease or increase of the value means a change of at least 10%, even more preferably at least 20%, at least 30%, at least 40%, at least 50%, at least 70%, at least 90%, or 100%. In this latter case, it can be the case that there is no longer a detectable value associated with the parameter.

The use of a compound or composition as a medicament as described in this document can also be interpreted as the use of said compound or composition in the manufacture of a medicament. Similarly, whenever a compound or composition is used for as a medicament, it can also be used for the manufacture of a medicament, or in a method.

In this document and in its claims, the verb “to comprise” and its conjugations is used in its non-limiting sense to mean that items following the word are included, but items not specifically mentioned are not excluded. In addition, reference to an element by the indefinite article “a” or “an” does not exclude the possibility that more than one of the element is present, unless the context clearly requires that there be one and only one of the elements. The indefinite article “a” or “an” thus usually means “at least one”. The word “about” or “approximately” when used in association with a numerical value (e.g. about 10) preferably means that the value may be the given value (of 10) more or less 1 % of the value.

In the context of this invention, a cell or a sample can be a cell or a sample from a sample obtained from a subject. Such an obtained sample can be a sample that has been previously obtained from a subject. Such a sample can be obtained from a human subject. Such a sample can be obtained from a non-human subject.

Legends to the figures

Fig. 1 - hPG-Dex, free or encapsulated in liposomes, induces a tolerogenic phenotype in Balb/c BMDCs and human moDCs. Immature BMDCs or moDCs were stimulated overnight with LPS and hPG, hPG-Dex, hPG Liposomes (Lip), or hPG-Dex Lip. Relative expression of (A) IL1B and (B) IDO, normalized based on HPRT expression and relative to the hPG group, measured by qPCR. (C) IL-10 and (D) IL-12p70 concentration was measured in supernatants of BMDCs (pg/mL) using ELISA. I %TLR2⁺, (F) %LAP⁺, (G) %CD86⁺, and (H) %CD40⁺ in BMDCs, and (I) %CD86⁺ and (J) %CD40⁺ in moDCs as measured by flow cytometry in the live CD1 1 c⁺ population. Means (+ SD), ** p < 0.01 , *** p < 0.001 , ““ p < 0.0001 as determined by two-way ANOVA and Tukey’s multiple comparisons test.

Fig. 2 - Dexamethasone increases the uptake of liposomes by BMDCs. (A) Liposomes were prepared as described above, with the addition of 0.02 mol% 1 ,1 -dioctadecyl-3, 3,3,3- tetramethylindodicarbocyanine. Immature BMDCs were stimulated overnight with LPS and fluorescently labelled hPG Liposomes (Lip) or hPG-Dex Lip. The % of CD11 c⁺ that were positive forthe fluorescent label in the liposomes was measured using flow cytometry. (B) Immature BMDCs were stimulated overnight with LPS and hPG, hPG-Dex, hPG Lip, or hPG-Dex Lip. %MerTK of live CD11 c⁺ cells were measured using flow cytometry. Means + SD **** p < 0.0001 as determined by t-test or one-way ANOVA and Tukey’s multiple comparisons test.

Fig. 3 - Antigen-specific T cell skewing by hPG-Dex liposomes in vitro and in vivo. Immature BMDCs were stimulated overnight with hPG + LPS, hPG-Dex + LPS, hPG Liposomes (Lip) + LPS, or hPG- Dex Lip + LPS. Cells were washed and CFSE-labelled CD4⁺ T cells from hPG-TCR mice were coincubated for 3 days with the BMDCs. (A) %CD25⁺Foxp3⁺ (B) %CD49b⁺LAG-3⁺, and (C) %Tbet⁺ cells of CFSE CD4⁺ T cells, measured by flow cytometry. Thyl .T Balb/c mice were injected intramuscularly with hPG protein, followed intravenously with 500,000 Thy1.1 ⁺ hPG-TCR CD4⁺ T cells. 1 day later, mice were injected intravenously with 1 nmol hPG or 1 nmol hPG-Dex encapsulated in liposomes. 3 days later, mice were sacrificed, and spleens were isolated for flow cytometry. (D) %CD25⁺Foxp3⁺ and I %CD49b⁺LAG-3⁺ cells of Thy1 .1⁺CD4⁺ T cells. Means (+ SD), * p < 0.05, ** p < 0.01 , “* p < 0.001 , ““ p < 0.0001 , as determined by one-way ANOVA and Tukey’s multiple comparisons test (in vitro) or Bonferroni’s multiple comparisons test (in vivo).

Fig. 4 - Free and encapsulated OVA323K4-Dex induces antigen-specific CD25⁺Foxp3+ Tregs in vivo. CD45.1⁺ BI6 mice were injected intravenously with 500,000 OT-II CD45.2⁺CD4⁺ T cells. 1 day later, mice were injected subcutaneously with 85 nmol OVA323, 85 nmol OVA323-Dex, or 1 nmol OVA323- Dex encapsulated in liposomes. 7 days later, mice were sacrificed and spleens were isolated for flow cytometry. Cells were stained for Viakrome, CD45.2, CD4, CD25, and Foxp3 and measured with flow cytometry. Means + SD, * p < 0.05, “ p < 0.01 p < 0.001 , as determined by one-way ANOVA and Bonferroni’s multiple comparisons tests.

Fig. 5 - hPG-Dex Liposomes inhibit the development of arthritis in mice. Female Balb/c mice were injected i.p. on days 0 and 21 with a mixture of 2 mg DDA and 250 pg human proteoglycan to induce arthritis. (A) In the preventative model mice were treated on day 17 via intravenous injection of PBS, hPG-Dex tolDCs, or hPG-Dex liposomes. (B) In the curative model, mice were enrolled after arthritis was established, and treated on days 0 and 7 via intravenous injection of PBS, hPG-Dex liposomes, or OVA₃23-Dex liposomes. Means ± SEM, * p < 0.05, ** p < 0.01 , *“ p < 0.001 , ““ p < 0.0001 compared to PBS group, and t P < 0.05, ft P < 0.01 compared to hPG-Dex tolDCs, and ## p < 0.01 compared to OVA323-Dex liposomes as determined by two-way ANOVA and Bonferroni’s multiple comparisons test. (C) Anti-hPG lgG1 and (D) lgG2a antibodies were measured in the serum of mice 25 days after the first injection of the treatment. OD values per plate were normalized based on a calibration curve. Means + SD, * p < 0.05, **** p < 0.0001 , as determined by one-way ANOVA and Bonferroni’s multiple comparisons tests.

Fig. 6 - hPG-Dex liposomes enhance tolerogenic responses in arthritic mice. Female Balb/c mice were injected i.p. on days 0 and 21 with a mixture of 2 mg DDA and 250 pg human proteoglycan to induce arthritis. Mice were enrolled after arthritis was established and treated on days 0 and 7 via intravenous injection of PBS, hPG-Dex liposomes, or OVA323-Dex liposomes. Mice were sacrificed on day 25 and organs were isolated for analysis. (A) CD11 c⁺CD86⁺ DCs, (B) CD1 1 c⁺PD-L1⁺ DCs, (C) CD4⁺PD-1⁺ T cells, (D) CD4⁺CD25⁺Foxp3⁺ Tregs, I CD4⁺RORyT⁺ Th17, and (F) CD4⁺Tbet⁺ Th1 cells, % of all live cells in the spleen, measured by flow cytometry. (G) MPO, (H) IL1B, and (I) IL10 expression in the paws of mice, normalized based on HPRT expression using the Pfaffl method, measured by qPCR. Means + SD, * p < 0.05, ** p < 0.01 , *“* p < 0.0001 as determined by oneway ANOVA and Bonferroni’s multiple comparisons test.

Fig. 7 - Encapsulation efficiency of hPG (left), dexamethasone (middle), or of hPG-Dex conjugate (right) in liposomes. It was found that the linking of Dex to hPG increases the encapsulation of both components.

Fig. 8 - Variations in conjugate design do not hinder encapsulation efficiency (DSPC:DSPG:CHOL (4:1 :2 molar ratio) liposomes). (A) Encapsulation of three different OVA323-Dex conjugates (using linkers KKKK (SEQ ID NO: 54) or EEEE (SEQ ID NO: 57) or SSSS (SEQ ID NO: 63)). Encapsulation efficiency remains comparably high for all options. (B) using the same three conjugates, it was found that the size of the liposomes remains consistent around about 200 nm. (C) using the same three conjugates, it was found that the polydispersity of the liposomes remains consistent around about 0.04-0.07. (D) using the same three conjugates, it was found that the zeta-potential of the liposomes remains consistent around about -55 to -60 mV.

Fig. 9 - Dexamethasone and Prednisolone conjugated to OVA323 increase antigen-specific Tregs in vitro. Immature BMDCs were stimulated overnight with LPS and free OVA323 + dexamethasone, free OVA323 + prednisolone, or conjugated dexamethasone or prednisolone. After incubation, cells were washed and CD4⁺ T cells from OT-II mice were co-incubated for 3 days with the BMDCs. %CD25⁺Foxp3⁺ cells were measured by flow cytometry. Means (+ SD), “ p < 0.01 , *** p < 0.001 , **** p < 0.0001 as determined by one-way ANOVA and Tukey’s multiple comparisons test.

Fig. 10 - Conjugation of an antigen (here OVA323) to a glucocorticoid receptor agonist (here Dex, using either an E4 or K4 linker) increases antigen-specific Tregs in vivo compared to free OVA323. Encapsulation of conjugates allows for 200-fold dose reduction. Conjugation leads to increased effect with either linker. Means + SD, * p < 0.05, “* p < 0.001 , ““ p < 0.0001 , as determined by one-way ANOVA and Bonferroni’s multiple comparisons tests.

Fig. 11 - Variations in choice of autoantigenic peptide do not hinder encapsulation efficiency in DSPC:DSPG:CHOL (4:1 :2 molar ratio) liposomes. (A) Encapsulation of four different Dex conjugates (using N-terminal linkers SEQ ID NO: 54), hPG = human proteoglycan (SEQ ID NO: 3), MOG = myelin oligodendrocyte glycoprotein (SEQ ID NO: 35), Ins = insulin (SEQ ID NO: 19), AChR = acetylcholine receptor (SEQ ID NO: 33). Encapsulation efficiency remains comparably high for all options. (B) using the same four conjugates, it was found that the size of the liposomes remains consistent around about 200 nm. (C) using the same four conjugates, it was found that the polydispersity of the liposomes remains consistent around about 0.05-0.11 . (D) using the same four conjugates, it was found that the zeta-potential of the liposomes remains consistent around about - 50 to -60 mV.

Fig. 12 - Liposome stability over time. Liposomes were stored in 10 mM phosphate buffer (pH 7.4) at 4°C for up to 40 months. Periodical DLS measurements reveal no changes in liposome size. Tested conjugates were OVA-Dex and hPG-Dex, both with a linker (SEQ ID NO: 54) linker. Peptides with linker were also tested without Dex.

Fig. 13 - Conjugates have predictable encapsulation efficiency (EE). (A) EE of a range of peptides in DSPC:DSPG:CHOL liposomes reveal a significant positive correlation between peptide charge and EE. (B) EE of a range of peptide-Dex conjugates in the same liposomes is not correlated to net charge of the peptide.

Fig. 14 -MOG-Dex conjugates prevent the development of EAE (as a model for multiple sclerosis) in mice. Mice were injected with dexamethasone-conjugated MOG in liposomes, or with controls (empty liposome, saline, or use of an OVA323 conjugate instead). 4 days later, EAE was induced. 2 days after EAE induction, mice were injected again with MOG-Dex or controls. (A) Mice were scored daily by following a five-point standardized rating of clinical symptoms: 0, no signs; 1 , loss of tail tonus; 2, flaccid tail; 3, hind limb paresis; 4, hind limb paralysis; 5, death. Notably only MOG-Dex shows near-complete prevention of EAE, while the three controls show similarly worse outcomes (B) Mice were weighed daily. Notably only MOG-Dex shows a healthy weight profile, while the three controls show similarly worse outcomes. The legend for (B) is the same as shown for (A).

Example 1. Material & Methods

Synthesis and characterization of Dex-peptide conjugates

Preloaded Fmoc-Lys(Boc)-Wang resin, Fmoc-Arg(Pbf)-Wang resin, 9-fluorenylmethyloxycarbonyl (Fmoc)-protected amino acids, and trifluoroacetic acid (TFA) were purchased from Novabiochem GmbH (Hohenbrunn, Germany). Peptide grade dimethylformamide (DMF), dichloromethane (DCM), piperidine, N,N’-diisopropyl carbodiimide (DIC), and high-performance liquid chromatography (HPLC) grade acetonitrile were purchased from Biosolve BV (Valkenswaard, Netherlands). Ethyl cyanohydroxyiminoacetate (Oxyma pure) was purchased from Manchester Organics Ltd (Cheshire, UK). Triisopropylsilane (TIPS), BioUltra grade ammonium bicarbonate, succinyl anhydride, 4-dimethylaminopyridine (DMAP), and pyridine were purchased from Sigma- Aldrich Chemie BV (Zwijndrecht, Netherlands). Dex was purchased from Acros Organics BV (The Hague, Netherlands).

Dex-peptide conjugates were synthesized using a synthetic approach that was described previously. In brief, the peptide epitope sequences were synthesized by microwave-assisted Fmoc- based chemistry using an H12 liberty blue peptide synthesizer (OEM Corporation, US). Dex succinate was coupled to the N-terminus of the peptide as with other Fmoc-protected amino acids. TFA/water/TIPS (95/2.5/2.5) was used to cleave the peptide off the resin and remove the side chain protecting groups. Peptides were purified by Prep-HPLC using Reprosil-Pur C18 column (10 pm, 250 x 22 mm). Mass spectrometry (MS) analysis was performed using a Bruker microTOF-Q instrument in positive mode to confirm the identity of the synthetic products. The epitope was derived from the hPG and Ovalbumin (OVA) antigens with the sequence ATEGRVRVNSAYQDK (SEQ ID NO: 3) and ISQAVHAAHAEINEAGR (SEQ ID NO: 2), respectively. A lysine tetramer linker (SEQ ID NO: 54) was added to the N-terminus of the sequences, linking the peptide to the dexamethasone. The Dex-peptide conjugates were cleaved and purified as described above for the peptides. Where not indicated otherwise, linkers, when present, are at the N-terminus.

Liposome preparation and characterization

Liposomes were prepared using an established thin film dehydration-rehydration method. The phospholipids 1 ,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) and 1 ,2-distearoyl-sn-glycero- 3-phosphoglycerol (DSPG), were purchased from Avanti Polar Lipids, Birmingham, AL, USA. Cholesterol (CHOL) was purchased from Sigma-Aldrich. Briefly, 180 mg total of dry powder DSPC:DSPG:CHOL in a 4:1 :2 molar ratio was weighed and transferred to a dry 100 mL roundbottom flask. The lipids were dissolved in 8 mL chloroform and 8 mL methanol. The solvents were evaporated under a vacuum in a rotary evaporator for 1 h at 40 °C, followed by an N2 stream for 30 min at RT. The resulting lipid film was rehydrated with 2000 pg of hPG, hPG-Dex, or OVA323-Dex dissolved in 10 mM 4-(2-hydroxyethyl)-1 -piperazineethanesulfonic acid (HEPES, pH 7.2) buffer to a total volume of 4 mL and homogenized by rotation in a water bath at 40°C for 1 h. For empty liposomes, liposomes were rehydrated with 4 mL of 10mM HEPES buffer. The resulting suspension was sized by high-pressure extrusion (LIPEX Extruder, Northern Lipids Inc., Burnaby, BC, Canada) on a heating plate set at 60°C by passing the dispersion four times through stacked 400 nm and 200 nm pore-size membranes (Whatman® NucleoporeTM, GE Healthcare, Amersham, UK). To separate non-encapsulated cargo from the liposomes, liposomes were ultracentrifuged (Type 70.1 Ti rotor) for 35 min at 55,000 rpm at 4 °C. This was repeated three times. Liposomes were stored at 4 °C and their stability was measured periodically. Liposomes were used within 2 months for in vitro experiments and within 2 weeks for in vivo experiments. The Z-average diameter and polydispersity index (PDI) of the liposomes were measured by dynamic light scattering (DLS) using a NanoZS Zetasizer (Malvern Ltd., Malvern, UK). For this, 10 pL of liposomes were diluted in 990 pL HEPES buffer pH 7.2. The ^-potential was measured by laser Doppler electrophoresis (Malvern Ltd.) using a universal dip cell. To determine the concentration of loaded hPG, hPG-Dex, OVA323, or OVA323-Dex RP-UPLC was used. For this, 20 pL of liposome suspension was dissolved in 180 pL of methanol, and the sample was vortexed. Sample injections were 7.5 pL in volume and the column used was a 1 .7 pm BEH C18 column (2.1 x 50 mm, Waters ACQUITY UPLC, Waters, MA, USA). Column and sample temperatures were 40 °C and 20 °C, respectively. The mobile phases were Milli-Q water with 0.1 % TFA (solvent A) and acetonitrile with 0.1 % TFA (solvent B). For separation, the mobile phases were applied in a linear gradient from 5% to 95% solvent B over 10 min at a flow rate of 0.25 mL/min. Peptide content was detected by absorbance at 280 nm, and Dex was detected at 240 nm³⁹ using an ACQUITY UPLC TUV detector (Waters ACQUITY UPLC, Waters, MA, USA). Peptide concentrations were calculated based on the respective calibration curves of antigen-Dex complexes dissolved in Milli-Q water.

Mice

For bone marrow isolation, 8-week-old WT mice on Balb/cAnNCrl background (male and female) were purchased from Charles River laboratories. Tyh1 .1 ⁺ hPG-TCR transgenic mice were bred inhouse at the central animal laboratory of Utrecht University, the Netherlands. For proteoglycan- induced arthritis (PGIA) studies, 16-week-old female Balb/cAnNCrl mice were purchased from Charles River laboratories. Mice were randomized into experimental groups based on weight or arthritis score using RandoMice. Humane end-points were adhered to, and the physical discomfort of arthritic animals was relieved by providing easy-to-reach water and food, and additional soft bedding materials. Animals were kept under standard conditions of the animal facility and all experiments were approved by the relevant Animal Experiment Committee.

Murine bone marrow-derived dendritic cell (BMDC) isolation, dendritic cell culture, and stimulation Bone marrow isolated from femurs and tibias of Balb/cAnNCrl WT mice were homogenized and seeded in 6-well plates at a cell density of 450,000 cells/mL in 2 mL IMDM (Gibco, ThermoFisher Scientific) supplemented with 10% FCS (Fetal Calf Serum; Bodinco, Alkmaar, The Netherlands), 100 units/mL of penicillin (Gibco, ThermoFisher Scientific, Landsmeer, The Netherlands), 100 ug/mL of streptomycin (Gibco, ThermoFisher Scientific, Landsmeer, The Netherlands) and 0.5 pM p-mercaptoethanol (Gibco, ThermoFisher Scientific, Landsmeer, The Netherlands). Cells were cultured at 37°C and 5% CO2 in the presence of 20 ng/mL of granulocyte-macrophage colonystimulating factor (GM-CSF, in-house produced) for 6 days. On day 2, IMDM and 20 ng/mL GM- CSF were added to the wells. Extra GM-CSF (20 ng/mL) was supplemented on day 5. On day 6, cells were harvested by scraping and counted. For flow cytometry of DCs, and further co-culture with T cells, cells were transferred to a 96-well F-bottom plate at 50,000 cells/well. The cells were left to adhere for 2 hours. Cells were matured in the presence of 10 ng/mL lipopolysaccharide (LPS, 0111 :B4; Sigma-Aldrich) and treated with free or encapsulated hPG, or free or encapsulated hPG- Dex (200 pL/well). In all cases, the concentration of the peptide was 1 ug/mL. For Dex -containing groups, the concentration was 0.18 ug/mL Dex. After 16 h, DCs were harvested for phenotypic characterization by flow cytometry. For qPCR and ELISA, cells were plated out at 600,000 cells/well in an F-bottom 48-well plate. The cells were left to adhere for 2 hours. Cells were stimulated with the same conditions and concentrations as for flow cytometry, in a total volume of 600 pL/well.

Human monocyte isolation, monocyte-derived dendritic cell culture, and stimulation

Peripheral blood mononuclear cells (PBMCs) were obtained from healthy human donors at Sanquin Blood Bank (Amsterdam, Netherlands). PBMCs were isolated by a Ficoll gradient, and subsequently, monocytes were isolated using anti-CD14 microbeads (Miltenyi Biotech) according to the manufacturer’s instructions. Monocytes were seeded in 6-well plates at 2,000,000 cells/mL in 2 mL RPMI (Gibco) supplemented with 5% FCS (Bodinco, Alkmaar, The Netherlands), 100 units/mL of penicillin (Gibco, ThermoFisher Scientific, Landsmeer, The Netherlands), and 100 ug/mL of streptomycin (Gibco, ThermoFisher Scientific, Landsmeer, The Netherlands). To differentiate monocytes towards DCs, 50 ng/mL hGM-CSF (Miltenyi Biotech) and 50 ng/mL hlL-4 (Miltenyi Biotech) were added. On day 3 of culture, fresh medium, and cytokines were added. On day 6, cells were harvested by scraping, counted, and transferred to a 96-well F-bottom plate at 50,000 cells/well. The cells were left to adhere for 2 hours. Cells were matured in the presence of 100 ng/mL lipopolysaccharide (LPS, O1 11 :B4; Sigma-Aldrich) and treated with free or encapsulated hPG, or free or encapsulated hPG-Dex (200 pL/well). In all cases, the concentration of the peptide was 1 ug/mL. For Dex-containing groups, the concentration was 0.18 ug/mL Dex. After 16h, DCs were harvested for phenotypic characterization by flow cytometry.

Enrichment of CD4⁺ T cells from murine spleens and co-culture with BMDCs

Spleens were isolated from Thy1.1⁺ hPG-TCR mice. A single-cell suspension of splenocytes was obtained by mashing spleens through a 70 pM filter (Falcon, Corning, New York, USA). Erythrocytes were lysed with Ammonium-Chloride-Potassium (ACK) lysis buffer (0.15 M NH4CI, 1 mM KHCO3, 0.1 mM Na2EDTA; pH 7.3). CD4⁺ T cells were negatively selected by magnetic separation using Dynabeads™ (sheep anti-rat IgG, ThermoFisher) and anti-CD8 (YTS169), anti-CD11 b (M1/70), anti-MHCll (M5/114) and anti-B220 (RA3-6B2, all in-house produced). The enriched CD4⁺ T cells were labeled with carboxyfluorescein succinimidyl ester (CFSE, 0.5 nM) according to the manufacturer’s protocol (ThermoFisher). BMDCs were plated out into a 96-well F-bottom plate (50,000 cells/well) and stimulated as described above. After 16h stimulation, cells were washed 4 times with 200 pL PBS/well to remove any free stimuli. To this, 100,000 cells/well of CFSE-labelled CD4⁺ T cells suspended in 200 pL RPMI (Gibco), supplemented with 5% FCS (Bodinco, Alkmaar, The Netherlands), 100 units/mL of penicillin (Gibco, ThermoFisher Scientific, Landsmeer, The Netherlands), and 100 ug/mL of streptomycin (Gibco, ThermoFisher Scientific, Landsmeer, The Netherlands) were added and incubated for 3 days. Subsequently, CD4⁺ T cells were harvested for phenotypic characterization by flow cytometry. Inflammatory adoptive transfer in vivo

CD4⁺ T cells were purified from the spleens and lymph nodes of Thy1 .1 ⁺ hPG-TCR transgenic mice as described above. At tO, WT Balb/cAnNCrl mice received an intramuscular injection of 50 uL PBS containing 10Oug hPG protein to induce a strong inflammatory response against hPG. After 2 hours mice received 500,000 CD4⁺ T cells intravenous via the tail vein. After 16 hours, mice were immunized intravenously with 200 uL PBS, 1 nmol free hPG, 1 nmol hPG liposomes, or 1 nmol hPG-Dex liposomes. 3 days after immunization, mice were sacrificed, and spleens were removed and processed as described above.

Preventative arthritis study in vivo

To induce arthritis in mice, female Balb/c mice were injected on days 0 and 21 intraperitoneally with a mixture of 2 mg dimethyldiotadecylammonium bromide (DDA) and 250 pg human proteoglycan. For the treatment of mice with hPG-Dex tolDCs, BMDCs were cultured in 6-well plates as described above. On day 6, 40 ug/mL hPG-Dex and 10 ng/mL LPS were added to the cells. DCs were harvested after 16 hours. The viability, purity, and phenotype of the DCs were confirmed using flow cytometry before injection in mice. Mice were treated on day 17 via an intravenous injection in the tail vein with 200 pL PBS, 200 pL 1 x 10⁶ hPG-Dex tolDCs (equivalent to 20 nmol of hPG-Dex) in PBS, or 200 pL hPG-Dex liposomes (2 nmol hPG-Dex) in PBS. Arthritis scores were determined 3 times per week starting from day 21 until day 55 in a blinded fashion by two researchers independently using a visual scoring system based on swelling and redness of paws. At the end of the experiment, mice were sacrificed by cervical dislocation.

Curative arthritis study in vivo

To induce arthritis in mice, female Balb/c mice were injected twice intraperitoneally with a mixture of 2 mg DDA and 250 pg human proteoglycan as described above. Arthritis scores were determined 3 times per week as described above. Mice were enrolled in the experiment (day -1) when they had a score of >2 for 2 consecutive scoring moments. Mice were treated on days 0 and 7 via intravenous injection in the tail vein with 200 pL PBS, 200 pL hPG-Dex liposomes (2 nmol hPG-Dex) in PBS, or 200 pL OVA-Dex liposomes (2 nmol OVA323-Dex) in PBS. Mice were scored during a period of 25 days after enrollment. At the end of the experiment, mice were sacrificed by cervical dislocation. Spleens were collected for flow cytometry, paws were collected for qPCR, and blood was collected in 0.8 mL z-serum separation tubes (Greiner Bio-One, Kremsmunster, Austria). Serum was separated from cells by centrifuging the blood samples at 10,000 x g for 5 minutes at 4 °C, collected into separate tubes, and stored at -20°C.

ELISA of stimulated BMDCs

BMDCs were stimulated as described above and the supernatant was harvested and either used directly for ELISA or stored at -80°C for future analysis. IL-10 (U-CyTech, Utrecht, the Netherlands) and IL-12p70 (9A5 and C17.8, BD Biosciences) was measured in the supernatants by ELISA according to the manufacturer’s instructions. Briefly, F-bottom Costar assay 96-well plates (Corning, Kennebunk, ME, USA) were coated with capture antibody at 4°C overnight. Plates were washed thoroughly with 0.01 % Tween-20 in PBS and blocked with 1 % BSA in PBS for 30 min at RT. Subsequently, plates were washed, and (diluted) samples and standard curves were incubated for 2h at RT. Then, plates were washed and the biotinylated detection antibody and streptavidin-HRP (BD Biosciences) were incubated for 1 hour at RT. Finally, plates were washed, and the samples were reacted with TMB substrate solution (BioLegend). The reaction was stopped with 2N H2SO4 solution, and the plates were measured using an iMark™ Microplate Absorbance Reader (Bio-Rad). Cytokine concentrations were calculated based on the respective calibration curves prepared with purified cytokines. qPCR of stimulated BMDCs

BMDCs were stimulated as described above, and 350 pL RLT buffer (Qiagen Benelux B.V., Venlo, the Netherlands) was added to the cells. The lysate was either used directly for mRNA extraction or stored at -80°C for future analysis. Total mRNA was extracted from stimulated BMDCs using the RNeasy kit (Qiagen) according to the manufacturer’s instructions. DNase treatment was performed on-column (Qiagen). The yield of mRNA extraction was measured using a Nanodrop (ThermoFisher). Transcription into cDNA was performed using the iScript™ cDNA Synthesis Kit (Bio-Rad Laboratories B.V., Veenendaal, The Netherlands). PCR and Real-Time detection were performed using a Bio-Rad MyiQ iCycler (Bio-Rad). Amplification was performed using IQ™ SYBR Green® Supermix (Bio-Rad) with 0.25 pM final concentrations of primers specific for IL1B (5 -TCC ATC TTC TTC TTT GGG TAT TG-3’ (SEQ ID NO: 44) and 5’-TTC CCG TGG ACC TTC CAG-3’ SEQ ID NO: 45) and Indoleamine 2,3-dioxygenase 1 (IDO) (5’-GCA GAC TGT GTC CTG GCA AAC T-3’ (SEQ ID NO: 46) and 5’-AGA GAC GAG GAA GAA GCC CTT G-3’ (SEQ ID NO: 47)), and hypoxanthine-guanine phosphoribosyl transferase (HPRT) (5 -CTG GTG AAA AGG ACC TCT COS’ (SEQ ID NO: 48) and 5'-TGA AGT ACT CAT TAT AGT CAA GGG CA-3' (SEQ ID NO: 49)). The following PCR program was used: pre-soaking at 95 °C for 3 min, [denaturation at 95 °C for 20 sec, annealing at 59°C for 30 sec] repeated 40 times. Melting curves and primer efficiencies were measured for each sample. For each sample mRNA expression was normalized to the detected Ct value of HPRT and expressed relative to the average of the DCs incubated with hPG + LPS. qPCR of paws

Paws were harvested and pooled per mouse in 6-well plates containing ice-cold sterile PBS. The skin was removed using scissors and tweezers, and the paws were agitated to release synovial fluid. The resulting suspension was passed through a 70 pM filter (Falcon, Corning, New York, USA) and cells were pelleted by centrifugation. After removal of supernatant, cells were lysed using 350 pL RLT buffer (Qiagen Benelux B.V., Venlo, the Netherlands). Total mRNA was immediately extracted using the RNeasy kit (Qiagen) according to the manufacturer’s instructions. Transcription into cDNA was performed using the iScript™ cDNA Synthesis Kit (Bio-Rad Laboratories B.V., Veenendaal, The Netherlands). PCR and Real-Time detection were performed using a Bio-Rad MyiQ iCycler (Bio-Rad). Amplification was performed using IQ™ SYBR Green® Supermix (Bio- Rad) with 0.25 pM final concentrations of primers specific for MPO (5 -GCT ACC CGC TTC TCC TTC TT-3’ (SEQ ID NO: 50) and 5’-GGT TCT TGA TTC GAG GGT CA-3’ (SEQ ID NO: 51)), IL1B (SEQ ID NOs 44 and 45), IL10 (5’-GGT TGC CAA GCC TTA TCG GA-3’ (SEQ ID NO: 52) and 5’- ACC TGC TCC ACT GCC TTG CT-3’ (SEQ ID NO: 53)), and hypoxanthine-guanine phosphoribosyl transferase (HPRT) (SEQ ID NOs 48 and 49'). The following PCR program was used: pre-soaking at 95 °C for 3 min, [denaturation at 95 °C for 20 sec, annealing at 59°C for 30 sec] repeated 40 times. Melting curves and primer efficiencies were measured for each sample. The Pfaffl method was used to calculate the gene expression ratio of each gene of interest vs. HPRT, using the PBS group as control.

ELISA of serum anti-hPG lgG1 and lgG2a

ELISA 96-well plates (Corning) were coated overnight with hPG (5pg/mL per well) in 0.1 M Carbonate buffer (pH = 9.5). Subsequently, wells were blocked with a blocking buffer consisting of 1 .5% milk powder (Campina, Zaltbommel, The Netherlands) dissolved in 1X PBS for 2 hours at RT. Mouse serum was added to the wells at different dilutions (IgG 1 : 1 :12500, 1 :25000, 1 :50000; lgG2a: 1 :500, 1 :2500, 1 :12500). On each plate, a standard curve composed of serum of a mouse that reached the humane endpoint for arthrtitis development (PGIA induction, no treatment) at dilutions 0, 1 :6250, 1 :12500, 1 :25000, 1 :50000, 1 :100000 for lgG1 and 0, 1 :250, 1 :500, 1 :1000, 1 :2000, 1 :4000 for lgG2a was Included. After two hours, lgG1 -HRP (X56; BD Biosciences) and lgG2a- HRP(19-15; BD Biosciences) antibodies were added to the wells in the blocking buffer at a 1 :1000 dilution. After 1 hour of incubation at RT, wells were washed, and TMB (Thermo Fisher Scientific) was added. The reaction was stopped using 2M H2SO4. ELISA data was read on the using an iMark™ Microplate Absorbance Reader (Bio-Rad) at 450 nm. The background signal (550 nm) was subtracted and serum levels of anti-hPG IgG 1 and lgG2a were calculated using the standard curve.

Flow cytometry

BMDCs or moDCs were stimulated as described above and harvested, washed 3 times with 200 pL of 4 mM EDTA and once with 200 pL PBS to remove any free antigen or liposomes, and transferred to a V-bottom 96-well plate. Co-cultured CFSE-labelled CD4⁺ T cells were harvested and transferred to a V-bottom 96-well plate. For splenocytes from in vivo experiments, 2 x 10⁶ splenocytes were plated out in 96-well U-bottom plates.

Cell suspensions were blocked for 15 min with 10 ug/mL Fc Block (2.4G2, in-house produced). BMDCs were stained with a monoclonal antibody mix of CD11 c-APC (N418, eBioscience, Thermo Fisher Scientific), TLR2-FITC (6C2, eBioscience, Thermo Fisher Scientific), CD86-FITC (GL1 , BD Biosciences), CD40-PE (3/23, BD Biosciences), LAP-PE (TW7-16B4, eBioscience, Thermo Fisher Scientific), MerTK-APC (2B10C42, BioLegend), and ViaKrome808 (Beckman Coulter, Indianapolis, IN, USA) in FACS Buffer (1X PBS supplemented with 2% FCS, 0.01 % sodium azide, and 2 mM EDTA). moDCs were stained with a monoclonal antibody mix of CD11 c-PE (MJ4-27G12, Miltenyi), CD40 PE-Cy7 (5C3, eBioscience), CD86-BB515 (FUN-1 , BD Biosciences), and ViaKrome808 (Beckman Coulter, Indianapolis, IN, USA) in FACS Buffer. CD4⁺ T cells were stained with a monoclonal antibody mix of CD4-BV785 (RM4-5, BioLegend, USA), LAG- 3-PE (eBioC9B7W, eBioscience, Thermo Fisher Scientific, USA), CD49b-APC-Cy7 (DX5, BioLegend, USA), and CD25-PerCP-Cy5.5 (PC61.5, eBioscience, Thermo Fisher Scientific, USA), and ViaKrome808 (Beckman Coulter, Indianapolis, IN, USA) in FACS Buffer. After 30 min incubation at 4°C in the dark, cells were washed with PBS, and fixed and permeabilized using the FoxP3 transcription factor staining set (eBioscience, San Diego, CA, USA). Subsequently, cells were stained intracellularly according to the manufacturer’s instructions with FoxP3-eFluor450 (FJK-16s, eBioscience, Thermo Fisher Scientific) and T-Bet-APC (4B10, eBioscience, Thermo Fisher Scientific). Splenocytes from the adoptive transfer experiment were stained with CD4-BV510 (RM4-5, BioLegend, USA), Thy1 ,1-PerCP-Cy5.5 (HIS51 , eBioscience), Thy1.1 -FITC (HIS51 , eBioscience), LAG-3-APC (C9B7W, eBioscience), CD49b-APC-Cy7 (DX5, Biolegend), PD-L1- BV650 (10F.9G2, Biolegend), CD11 c-FITC (N418, eBioscience), CD86-PE-Cy5 (GL1 , eBioscience), and ViaKrome808 (Beckman Coulter, Indianapolis, IN, USA) in FACS Buffer. After 30 min incubation at 4°C in the dark, cells were washed with PBS, and fixed and permeabilized using the FoxP3 transcription factor staining set (eBioscience, San Diego, CA, USA). Subsequently, cells were stained intracellularly according to the manufacturer’s instructions with FoxP3-eFluor450 (FJK-16s, eBioscience, Thermo Fisher Scientific) and T-bet-APC (4B10, eBioscience). Splenocytes from the curative arthritis study experiment were stained with a monoclonal antibody mix of CD4- BV785 (RM4-5, BioLegend, USA) and CD25-PerCPCy5.5 (PC61.5, eBioscience, Thermo Fisher Scientific, USA), and ViaKrome808 (Beckman Coulter, Indianapolis, IN, USA) in FACS Buffer. After 30 min incubation at 4°C in the dark, cells were washed with PBS, and fixed and permeabilized using the FoxP3 transcription factor staining set (eBioscience, San Diego, CA, USA). Subsequently, cells were stained intracellularly according to the manufacturer’s instructions with FoxP-eFluor450 (FJK-16s, eBioscience, Thermo Fisher Scientific), RORyT-PE (AFKJS-9, eBioscience, Thermo Fisher Scientific), GATA-3-PE-Cy7 (TWAJ, eBioscience, Thermo Fisher Scientific) and T-Bet-APC (4B10, eBioscience, Thermo Fisher Scientific). After 30 min incubation at 4 °C in the dark, cells were washed and resuspended in 100 pL PBS for measurement. To ensure correct analysis, relevant single-stain, and fluorescence minus one (FMO) controls were used. Samples were measured on a Beckman Coulter Cytoflex LX at the Flow Cytometry and Cell Sorting Facility located at the Faculty of Veterinary Medicine at Utrecht University. The total measured volume was 85 pL per sample, at a measurement speed of 60 pL/min. Acquired data were analyzed using FlowJo Software v.10.7 (FlowJo LLC, Ashland, OR, USA).

Mice for adoptive transfer assay using liposome-encapsulated OVAjzs-Dex

8-week old female C57BL/6-Ly5.1 and C57BL/6-Tg(TcraTcrb)425Cbn/Crl (OTII) mice used for adoptive transfer experiments were purchased from Charles River laboratories.

Adoptive transfer in vivo using liposome-encapsulated OVAszs-Dex

CD4⁺ T cells were purified from OT-II transgenic mice using a CD4⁺ T cell enrichment kit according to the manufacturer’s instructions (Miltenyi, Netherlands). On day -1 , all CD45.1 + Ly5.1 mice received 500,000 CD4⁺ T cells intravenously via the tail vein. On day 0, mice were immunized subcutaneously by injection into the left and right flanks (50 pL each side) of 85 nmol OVA323, 85 nmol OVA323-Dex, or 1 nmol OVA323-Dex encapsulated in liposomes. Seven days after immunization, mice were sacrificed and spleens were removed and processed as described above. For FACS analysis, 2 x 10⁶ splenocytes were plated out in 96-well U-bottom plates. Cells were blocked for 15 min with Fc Block (2.4G2, in-house produced). CD4⁺ T cells were stained with a monoclonal antibody mix of CD4-BV785 (RM4-5, BioLegend, USA), CD45.2-PerCP-Cy5.5 (104, eBioscience), and CD25-BV650 (PC61 , Biolegend), and ViaKrome808 (Beckman Coulter, Indianapolis, IN, USA) in FACS Buffer (1X PBS supplemented with 2% FCS and 2 mM EDTA). After 30 min incubation at 4 °C in the dark, cells were washed with PBS, and fixed and permeabilized using the FoxP3 transcription factor staining set (eBioscience, San Diego, CA, USA). Subsequently, cells were stained intracellularly according to the manufacturer’s instructions with FoxP-eFluor450 (FJK-16s, eBioscience, Thermo Fisher Scientific). Finally, cells were washed and resuspended in 100 pL PBS for measurement. To ensure correct analysis, relevant single-stain, and fluorescence minus one (FMO) controls were used. Samples were measured on a Beckman Coulter Cytoflex LX at the Flow Cytometry and Cell Sorting Facility at the Faculty of Veterinary Medicine at Utrecht University. Acquired data were analyzed using FlowJo Software v.10.7 (FlowJo LLC, Ashland, OR, USA).

Dose determination of encapsulated or unencapsulated conjugates

CD45.1⁺ BI6 mice were injected intravenously with 500,000 OT-II CD45.2⁺CD4⁺ T cells. 1 day later, mice were injected subcutaneously with OVA323, OVA323E4-Dex, OVA323K4-Dex, or OVA323K4-Dex encapsulated in liposomes. 7 days later, mice were sacrificed and spleens were isolated for flow cytometry. Cells were stained for Viakrome, CD45.2, CD4, CD25, and Foxp3 and measured with flow cytometry.

Statistical analysis

Statistical analysis was performed in GraphPad Prism v.9.3.1 . Details of the analyses are indicated in the figure legends.

Example 2. Conjugates induce tolerogenic DCs, which induce antigen-specific Tregs Dexamethasone, free or encapsulated in liposomes, induces a tolerogenic phenotype in DCs in vitro.

The arthritis-relevant MHC-II autoantigen hPG and the ovalbumin-derived MHC-ll-restricted OVA323-339 antigen were extended with a linker to couple to Dex (forming hPG-Dex and OVA323- Dex, respectively). The antigens were encapsulated into anionic DSPG liposomes. The liposomes were below 200 nm in size and had a negative charge. The LE of the antigen-Dex complexes were between 46.5 and 49.6% (see Table 1). Conjugates without a linker had a loading efficiency of around 10%. Table 1 : Physicochemical characterization of Dex-loaded liposomes

^aZ-average diameter (Zave), mean ± SD, n = 3. ^b %LE (loading efficiency) was calculated as the total amount of peptide before extrusion/total amount of peptide after purification * 100%.

It was found that liposomes were particularly stable (see Fig. 12) and did not leak. Ultracentrifugation of the liposomes tested in Fig. 12, and measurement of supernatants by UPLC, showed no leakage of cargo.

To evaluate the tolerance induction by free or encapsulated Dex, immature BMDCs were stimulated overnight with LPS and either free or encapsulated hPG and hPG-Dex. The free hPG control was coupled to the same linker that was used to couple hPG to Dex. Gene expression of IL1B and secretion of IL-12p70 were greatly reduced when BMDCs were incubated with hPG-Dex or hPG- Dex encapsulated in liposomes compared to the hPG control. Encapsulating the antigen in liposomes without Dex had the same effect (Fig. 1A and D). hPG-Dex, free and encapsulated, increased the expression of Toll-like receptor 2 (TLR2) (Fig. 1 E), and latency-associated protein (LAP, the membrane-bound form of TGF-p) (Fig. 1 F). This coincided with a reduction in the expression of costimulatory molecules CD86 and CD40 in BMDCs and moDCs (Fig. 1 G, H, I, J). Surprisingly, encapsulation in liposomes enhanced the tolerogenic capacity of hPG-Dex, as evidenced by the increased gene expression of IDO (Fig. 1 B), the release of IL-10 (Fig. 1 C). We also observed that hPG-Dex liposomes had more efficient uptake by BMDCs as compared to Dex- free hPG liposomes (Fig. 2A), possibly due to an increase in the phagocytic receptor MerTK (Fig. 2B).

Dex-linked hPG induces antigen-specific Tregs in vitro and in vivo.

As shown above, conjugates (here: hPG-Dex liposomes) can induce phenotypical tolDCs with a cytokine secretion and gene expression profile that can promote tolerance in T cells. Next, the effect of these tolDCs on antigen-specific T cells was assessed in vitro and in vivo. Both free and encapsulated hPG-Dex-pulsed DCs increased antigen-specific CD25⁺Foxp3⁺ Tregs (Fig. 3A). Only encapsulated hPG-Dex increased CD49b⁺LAG-3⁺ Tr1 cells (Fig. 3B). Both free and encapsulated hPG-Dex decreased antigen-specific T-bet⁺ Th1 cells, although free hPG-Dex was more potent than encapsulated hPG-Dex. In a naive adoptive transfer model, OVA323-Dex liposomes were able to expand antigen-specific CD25⁺Foxp3⁺ Tregs (Fig. 4). In an inflammatory in vivo adoptive transfer model, intravenous injection of hPG liposomes induced antigen-specific CD25⁺Foxp3⁺ Tregs, as did hPG-Dex liposomes, albeit to a lesser extent (Fig. 3D). The hPG-Dex liposomes did greatly enhance antigen-specific CD49b⁺LAG-3⁺ Tr1 cells (Fig. 3E) compared to free hPG antigen and Dex-free hPG liposomes. This shows that hPG-Dex liposomes can induce strong Tr1 responses, even in an inflammatory environment, which is hypothesized to be necessary for suppressing the responses in an arthritis model.

Arthritis development in mice is inhibited by hPG-Dex liposomes

Based on the results, we hypothesized that the hPG-Dex liposomes would give the best protection in a model of arthritis. To assess the pre-clinical efficacy of the hPG-Dex liposomes we employed the PGIA mouse model. First, we performed a preventative study, whereby mice were injected intravenously with PBS, 1 x 10⁶ hPG-Dex tolDCs (equivalent to 20 nmol of hPG-Dex), or hPG-Dex liposomes (2 nmol) before they had developed arthritis (Fig. 5A). Mice that were treated with hPG- Dex liposomes developed significantly less arthritis compared to the PBS and even the hPG-Dex- pulsed tolDC-treated mice (Fig. 5A). Furthermore, while 100% of mice in the PBS group had developed arthritis (score of 2 or higher), 75% and 27% of mice in the hPG-Dex-pulsed tolDC and hPG-Dex liposomes groups, respectively, developed arthritis (Table E1).

Table E1 : Additional data PGIA mouse studies

All data shown as mean or mean ± SD. ^a at the end of the study.

Next, we assessed whether the hPG-Dex liposomes could halt the progression of arthritis in mice that had ongoing inflammation (Fig. 5B). In this model, we observed that they could indeed stabilize arthritis in mice, compared to PBS and compared to OVA323-Dex liposomes (Fig. 5B). 100% of mice in the PBS and OVA323-Dex liposomes groups had increased arthritis scores compared to the day of the first injection, while 33% of mice treated with hPG-Dex liposomes had a lower score at the end of the experiment compared to the start (Table E1). Anti-hPG IgG 1 and lgG2a were measured in the serum of mice after sacrifice. Mice that received hPG-Dex liposomes overall had lower amounts of anti-hPG lgG1 compared to the other groups, but not lgG2a (Fig. 5C and D).

Further analysis of the spleens of mice sacrificed on day 25 after the first injection of liposomes or control revealed that, in the mice that received the hPG-Dex liposomes, the %CD11 c⁺CD86⁺ DCs were reduced compared to the other groups (Fig. 6A). This coincided with an increase in CD25⁺Foxp3⁺ and PD-1⁺ regulatory T cells (Fig. 6C and D). PD-L1 , RORyT⁺, and Tbet⁺ populations were mildly affected (Fig. 6B, E, and F). Interestingly, in the paws of mice, the expression of IL10 was enhanced in mice that received hPG-Dex liposomes (Fig. 6I), along with a non-significant reduction in MPO and IL1B (Fig. 6G and H).

Multiple sclerosis development in mice is inhibited by MOG-Dex liposomes

Encapsulation of MOG-Dex conjugates in liposomes prevented the development of EAE (autoimmune encephalomyelitis) in mice. Eleven-week-old mice were injected intravenously with dexamethasone-conjugated MOG (SEQ ID NO: 35) in DSPC:DSPG:CHOL liposomes. In parallel, further mice were injected with dexamethasone-conjugated OVA323 in DSPC:DSPG:CHOL liposomes, or with an equivalent amount of DSPC:DSPG:CHOL liposomes, or with saline. Four days later, EAE was induced in in mice by subcutaneous injection at the base of the tail with 200 pg of recombinant human myelin oligodendrocyte glycoprotein MOG35-55 emulsified in 100 pl complete Freund’s adjuvant supplemented with 4 mg/ml of Mycobacterium tuberculosis (H37RA strain) according to manufacturer’s guidelines (Hooke Laboratories, Lawrence, USA). Within 2 h and after 24 h, mice were intraperitoneally injected with 0.1 ml pertussis toxin. 2 days after EAE induction, mice were injected again with liposomes or controls. Mice were weighed and scored daily by following a five-point standardized rating of clinical symptoms: 0, no signs; 1 , loss of tail tonus; 2, flaccid tail; 3, hind limb paresis; 4, hind limb paralysis; 5, death.

Discussion

Restoring immune tolerance is relevant for the treatment of autoimmune and chronic inflammatory diseases. Current clinical trials are making use of DCs pulsed with immunomodulators and diseaserelevant antigens to achieve this (Bell et al. 2015). Despite encouraging results, the production of these tolDCs requires specialized research centers and is costly, making them less accessible to the large groups of patients that need treatment. Accordingly, there’s a need for a strategy that overcomes these limitations. Nanoparticles, including liposomes, are promising drug-delivery vehicles that can surpass the need for ex vivo culturing of tolDCs (Benne et al. 2022). Here, we prepared liposomes for the delivery of conjugates to induce antigen-specific immune tolerance in vitro, in vivo, and in a pre-clinical model for RA. We coupled Dex to our antigens of interest, intending to prevent the uptake of Dex by APCs in a non-antigen-specific context, which prevents non-antigen-specific effects. In addition, we hypothesized that the liposomes facilitated more efficient uptake of antigen-Dex conjugates by APCs than free antigen-Dex conjugates, allowing us to greatly reduce the required dose, further minimizing side effects.

BMDCs and moDCs exhibit a tolerogenic phenotype upon stimulation with free hPG-Dex (Fig. 1). This indicates that the linking of Dex to an antigen, such as hPG, surprisingly does not hinder the functionality of Dex. Encapsulation of hPG-Dex into liposomes similarly was found to surprisingly not impede the immunomodulatory effects of Dex (Fig. 1 and Fig. 2). On the contrary, stimulation of DCs with hPG-Dex liposomes increased gene expression of IDO (Fig. 1 B), the release of IL-10 (Fig. 1 C), and protein expression of LAP (Fig. 1 F) compared to free hPG-Dex. Interestingly, some tolerogenic properties of DSPG-liposomes were enhanced by the addition of Dex (Fig. 1). Furthermore, Dex stimulates uptake of liposomes, possibly due to an increase in the phagocytic receptor MerTK (Fig. 2). This increased uptake of the inherently tolerogenic liposomes, together with the tolerogenic properties of Dex, could explain why the Dex -containing liposomes show a more favorable tolDC phenotype. Encapsulation was further shown to be more efficient for conjugates than for the individual components (Fig. 7), and aspects of conjugate design such as a choice of linker was shown to not influence this encapsulation (Fig. 8).

Our studies using TCR-specific transgenic CD4⁺ T cells indicate that the coupling of Dex to an antigen via a linker does not prevent MHC presentation by DCs and subsequent TCR recognition by antigen-specific T cells as we observe potent T cell responses. The most striking result is that hPG-Dex liposome-treated BMDCs increase CD49b⁺LAG-3⁺ Tr1 numbers in in vitro co-culture experiments (Fig. 3B), while free hPG-Dex and hPG liposomes without Dex do not induce Tr1 cells. In an in vivo inflammatory adoptive transfer model, it was confirmed that hPG-Dex liposomes induce antigen-specific CD49b⁺LAG-3⁺ Tr1 cells (Fig. 3E).

Since the hPG-Dex liposomes induced potent Tr1 responses, and these responses are important for protection to arthritis (Volz et al. 2013), we tested these liposomes in a murine PGIA disease model for arthritis. hPG-Dex liposomes significantly reduced arthritis development compared to mice that received hPG-Dex tolDCs and PBS (Fig. 5A). It should be stated that antigen-loaded dexamethasone-induced tolDCs are a known therapy against rheumatoid arthritis (Jansen et al. 2019), yet the present invention yields better results (Fig. 5A). Importantly, hPG-Dex liposomes could prevent the further progression of arthritis in mice that had established diseases (Fig. 5B). Furthermore, the lack of arthritis development in mice that received hPG-Dex liposomes is more pronounced than in previous research using arthritis-related agent-pulsed tolDC therapy in a mouse model (Jansen et al. 2019; Hilkens et al. 2013).

Mechanistically, we show that mice treated with hPG-Dex liposomes had a significantly lower proportion of %CD11 c⁺CD86⁺ DCs compared to the PBS group (Fig. 6A). At the same time, there was a greater amount of CD4⁺CD25⁺Foxp3⁺ and CD4⁺PD-1⁺ T cells in the spleens of these mice (Fig. 6C and D). This suggests that there were some long-lasting systemic effects of the therapy, which seem to be antigen-specific, since it was not observed in the mice receiving OVA323- Dex liposomes. The CD4⁺CD25⁺Foxp3⁺ and CD4⁺PD-1⁺ T cells possibly contribute to protection against arthritis or are the result of reduced inflammation in mice. Nguyen et al. also observed a decrease in CD86⁺ APCs and an increase in Foxp3⁺CD4⁺ T cells in spleens of mice after nanoparticle treatment, which coincided with protection against experimental autoimmune encephalomyelitis (EAE, see Nguyen et al. 2022). Interestingly, we saw a significant increase in the expression of IL 10 in the paws of mice treated with hPG-Dex liposomes (Fig. 6I), which can explain why these mice had a lower arthritis score than controls.

Liposomes have been used as delivery vehicles to target Dex towards the inflamed joints, thereby reducing arthritis symptoms through the broad inflammation-inhibiting properties of Dex. To test the antigen-specificity of our treatment, and the effect of delivering liposomal Dex in arthritis mice, we treated mice with OVA323-Dex liposomes, OVA323 being a disease-irrelevant MHC-II antigen. We observed no changes in any assays between OVA323-Dex liposomes to PBS (Fig. 5 and 6). This was expected, since to obtain an accumulation of liposomes in inflamed joints through the leaky vasculature that is associated with inflammation, it is necessary to functionalize the liposomes through e.g., PEGylation, peptides (Meka et al. 2019), or other small molecules on the surface of the liposomes. Furthermore, the dose of Dex reported in studies that deliver Dex to the site of inflammation via nanoparticles is 5 to 30 times higher than the dose used in this study (0.1 - 1.2 mg/kg vs 0.02 mg/kg), and often required more than 2 injections. Therefore, we can conclude that the observed effect in our study is not due to the accumulation of Dex liposomes in the joints that affect the inflamed tissue, but rather due to antigen-specific skewing of T cells toward protective responses against arthritis.

Mice treated with MOG-Dex liposomes were able to maintain a healthier weight, and scored particularly low when scored according to a standardized rating of clinical symptoms associated with EAE (Fig. 14). Negative controls using saline, empty liposomes, or liposomes comprising a conjugate unrelated to multiple sclerosis (viz. OVA323-Dex) all showed outcomes comparable to one another, but different from the MOG-Dex treatment. The negative controls all showed a profile of increased five-point scoring, and a profile of weight loss after EAE induction. This corroborates that MOG-Dex liposomes have a beneficial effect related to its peptide, and not to the delivery of Dex.

Our conjugates provide a promising strategy to inhibit for instance arthritis or multiple sclerosis development. The results presented here highlight the therapeutic potential of antigen- Dex-loaded DSPG-containing liposomes in immune therapy against autoimmune diseases.

References

Bell GM, Anderson AE, Diboll J, et al. Autologous tolerogenic dendritic cells for rheumatoid and inflammatory arthritis. Ann Rheum Dis. 2017;76(1):227-234. doi:10.1136/annrheumdis-2015- 208456

Benne N, ter Braake D, Stoppelenburg AJ, Broere F. Nanoparticles for Inducing Antigen- Specific T Cell Tolerance in Autoimmune Diseases. Front Immunol. 2022;13. doi:10.3389/fimmu.2022.864403

Volz T, Skabytska Y, Guenova E, et al. Nonpathogenic bacteria alleviating atopic dermatitis inflammation induce IL-10-producing dendritic cells and regulatory Tr1 cells. Journal of Investigative Dermatology. 2014;134(1):96-104. doi:10.1038/jid.2013.291

Jansen MAA, Spiering R, Ludwig IS, van Eden W, Hilkens CMU, Broere F. Matured tolerogenic dendritic cells effectively inhibit autoantigen specific CD4+ T cells in a murine arthritis model. Front Immunol. 2019;10(AUG). doi:10.3389/fimmu.2019.02068

Hilkens CMU, Isaacs JD. Tolerogenic dendritic cell therapy for rheumatoid arthritis: Where are we now? Clin Exp Immunol. 2013;172(2):148-157. doi :10.1111/cei.12038

Nguyen TL, Choi Y, Im J, et al. Immunosuppressive biomaterial-based therapeutic vaccine to treat multiple sclerosis via re-establishing immune tolerance. Nat Commun. 2022;13(1). doi:10.1038/s41467-022-35263-9

Meka RR, Venkatesha SH, Acharya B, Moudgil KD. Peptide-targeted liposomal delivery of dexamethasone for arthritis therapy. Nanomedicine. 2019;14(11):1455-1469. doi:10.2217/nnm- 2018-0501

Claims

1 . A conjugate comprising:

(i) a glucocorticoid receptor agonist, and

(ii) a peptide that is autoantigenic.

2. The conjugate according to claim 1 , wherein the glucocorticoid receptor agonist is

3. The conjugate according to claim 1 or 2, wherein the peptide is an autoantigen associated with an autoimmune disease, such as rheumatoid arthritis, spondyloarthritis, juvenile idiopathic arthritis, psoriatic arthritis, psoriasis, Sjogren disease, systemic lupus erythematosus, dermatomyositis, systemic sclerosis, idiopathic thrombocytopenic purpura, alopecia, vitiligo, type 1 diabetes, myasthenia gravis, multiple sclerosis, Graves’ disease, or autoimmune hepatitis, preferably associated with rheumatoid arthritis or with multiple sclerosis.

4. The conjugate according to any one of claims 1-3, wherein the peptide is derived from a protein selected from human proteoglycan, insulin, insulin precursor, preproinsulin, proinsulin, melanin, topoisomerase 1 , topoisomerase 2, a glutamate decarboxylase such as glutamate decarboxylase 2 or glutamate decarboxylase 65, collagen such as type II collagen, citrullinated human proteoglycan, alpha-enolase, citrullinated alpha-enolase, cartilage intermediate-layer protein, citrullinated cartilage intermediate-layer protein, fibrinogen, citrullinated fibrinogen, vimentin, citrullinated vimentin, acetylcholine receptor, a myelin protein such as myelin oligodendrocyte glycoprotein, myelin proteolipid protein, or myelin basic protein, thyrotropin receptor, and smooth muscle, preferably from human proteoglycan.

5. The conjugate according to any one of claims 1 -4, wherein the peptide comprises a sequence that has from 6 to 60, preferably 10 to 40, more preferably from 12 to 20 contiguous amino acids from the sequence of a protein as identified in claim 4, wherein the contiguous sequence can have zero, one, two, or three amino acid substitutions.

6. The conjugate according to any one of claims 1 -5, wherein the peptide has a length of 6 to 70 amino acids, preferably of 10 to 40 amino acids, more preferably of 12 to 20 amino acids.

7. The conjugate according to any one of claims 1 -6, wherein the peptide comprises or consists of the sequence of any one of SEQ ID NOs: 3-43, preferably SEQ ID NOs: 3-18, more preferably SEQ ID NOs: 3-11 , even more preferably SEQ ID NO: 3.

8. The conjugate according to any one of claims 1-7, wherein the glucocorticoid receptor agonist is linked to a terminus of the peptide.

9. The conjugate according to claim 8, wherein the glucocorticoid receptor agonist is linked to the N-terminus of the peptide.

10. The conjugate according to any one of claims 1-9, wherein the glucocorticoid receptor agonist and the peptide are linked through a linker, wherein the linker is preferably a peptide linker comprising from 1 to about 12 amino acids, preferably from 2 to about 8 amino acids, more preferably from about 3 to about 6 amino acids, most preferably 4 or 5 amino acids, wherein preferably the amino acids are hydrophilic.

11. The conjugate according to any one of claims 1 -10, wherein the glucocorticoid receptor agonist is dexamethasone, and wherein the peptide is an autoantigen associated with either rheumatoid arthritis or multiple sclerosis.

12. A composition comprising the conjugate according to any one of claims 1 -11 , wherein the conjugate is comprised in a nanoparticle.

13. The composition according to claim 12, wherein the nanoparticle is a micelle, polymeric nanoparticle, polysaccharidic nanoparticle, liposome, lipoplex, unilamellar vesicle, multilamellar vesicle, or a cross-linked or hybrid variant thereof.

14. The composition according to claim 12, wherein the nanoparticle is a liposome.

15. The composition according to any one of claims 12-14, wherein the nanoparticle comprises one or more phospholipids, preferably two phospholipids, wherein the phospholipids are preferably selected from 1 ,2-dilauroyl-sn-glycero-3-phosphate (DLPA), 1 ,2-dilauroyl-sn- glycero-3- phosphoethanolamine (DLPE), 1 ,2-dimyristoyl-sn-glycero-3-phosphate (DMPA), 1 ,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), 1 ,2-dimyristoyl-sn-glycero-3- phosphoethanolamine (DMPE), 1 ,2-dimyristoyl-sn-glycero-3-phosphoglycerol (DMPG), 1 ,2- dimyristoyl-sn-glycero-3-phosphoserine (DMPS), 1 ,2-dipalmitoyl-sn-glycero-3-phosphate (DPPA), 1 ,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), 1 ,2-dipalmitoyl-sn-glycero-3- phosphoethanolamine (DPPE), 1 ,2-dipalmitoyl-sn-glycero-3-phosphoglycerol (DPPG), 1 ,2- dipalmitoyl-sn-glycero-3-phosphoserine (DPPS), 1 ,2-distearoyl-sn-glycero-3-phosphate (DSPA), 1 ,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), 1 ,2-distearoyl-sn-glycero-3- phosphoethanolamine (DSPE), 1 ,2-distearoyl-sn-glycero-3-phosphoglycerol (DSPG), 1 ,2- distearoyl-sn-glycero-3-phosphoserine (DSPS), hydrogenated soy phosphatidylcholine (HSPC), 1 ,2-Dioleoyl-sn-glycero-3-phosphate (DOPA), 1 ,2-Dioleoyl-sn-glycero-3- phosphoethanolamine (DOPE), 1 ,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1 ,2- Dioleoyl-sn-glycero-3-phosphoserine (DOPS), 1 ,2-Dilauroyl-sn-glycero-3-phosphocholine (DLPC), or polymer-conjugated phospholipid such as PEGylated phospholipid or polyglycerol phospholipid.

16. The conjugate according to any one of claims 1 -11 , or the composition according to any one of claims 12-15, for use as a medicament.

17. The conjugate for use or composition for use according to claim 16, wherein the medicament is for the treatment of an immune or autoimmune disease and/or an inflammation, preferably for the treatment of rheumatoid arthritis, systemic lupus erythematosus, thrombocytopenia, Idiopathic thrombocytopenic purpura (ITP), Systemic Sclerosis, Graves’ Disease, Hashimoto’s thyroiditis, vasculitis, inflammatory bowel disease, multiple sclerosis, hemolytic anemia, thyroiditis, stiff person syndrome, pemphigus vulgaris, myasthenia gravis, lupus nephritis, Crohn’s Disease, or Ulcerative Colitis, more preferably for the treatment of rheumatoid arthritis, type 1 diabetes, myasthenia gravis, multiple sclerosis, Graves’ disease, or autoimmune hepatitis.

18. The conjugate for use or composition for use according to claim 16, wherein the medicament is for the treatment of rheumatoid arthritis or multiple sclerosis.

19. The conjugate for use or composition for use according to claim 16, wherein the medicament is for the treatment of rheumatoid arthritis, and wherein the peptide is an autoantigen associated with rheumatoid arthritis.

20. The conjugate for use or composition for use according to claim 16, wherein the medicament is for the treatment of multiple sclerosis, and wherein the peptide is an autoantigen associated with multiple sclerosis.

21. A method for treating an autoimmune disease and/or inflammation, the method comprising the step of administering a conjugate according to any one of claims 1 -11 , or a composition according to any one of claims 12-15, to a subject in need thereof.