Stabilized MHC I

The present invention relates to modified MHC I molecules that can be expressed in cell culture and are more stable than natural MHC I molecules. Further, the invention relates to analytical processes using the modified MHC I molecules, e.g. for determination of antibodies in a serum sample directed against these MHC I molecules, or in tetramer staining. The MHC I molecules comprise or consist of a heavy chain (alpha chain) and a P2-microglobulin, wherein the alpha chain may be devoid of a transmembrane domain to give a soluble MHC I molecule, or a transmembrane domain may be linked to the C-terminus of the alpha chain, giving a MHC I molecule that is bound to a cell membrane, e.g. anchored in the membrane of the cell expressing the MHC I molecule. Accordingly, the stabilized MHC I molecules can be soluble or can be membrane-bound to a cell membrane.

The MHC I molecules of the invention have the advantage, e.g. in soluble form in which they have no transmembrane domain, that the alpha chain is stably associated with the b2- microglobulin also in the absence of a cognate antigen that has affinity to the antigen binding site of the alpha chain. The MHC I molecules of the invention can therefore preferably be produced in a state that is free from cognate antigen, e.g. free from antigen bound to the alpha chain. MHC I molecules of the invention which are devoid of cognate antigen bound to the alpha chain can e.g. be used for analytical processes in which the MHC I molecule is contacted with an antigen after synthesis of the MHC I molecule. An analytical process includes contacting an MHC I molecule according to the invention with a sample previously obtained from a patient, and preferably providing the result of the analytical process to the patient or to a medic appointed by the patient.

State of the art

Shields et al., The Journal of Biological Chemistry 28010-28018 (1998) describes that exogenous P2-microglobulin, especially a S55V mutant of P2-microglobulin, stabilizes human cell surface MHC I molecules and facilitates their loading with peptides.

US6 682 741 B1 describes a mutant of the class I major histocompatibility complex protein (MHC I), which comprises a heavy chain (alpha chain) in association with a light chain (b2- microglobulin), wherein compared to wild-type p2-microglobulin (S55) the mutation S55V in the p2-microglobulin confers higher stability to HLA-Al, HLA-A2 and HLA-A3 by a factor of 2 to 3 in the presence of a specific binding peptide and an appropriate HLA-specific antibody.

Amino acid sequences of MHC I molecules, especially of HLA I molecules are known and are available from the European Bioinformatics Institute, e.g. at

https://ebi.ac.uk/ipd/imgt/hla/.

WO2017/201131 Al describes an MHC I molecule that is covalently linked by a disulphide bond to a P2-microglobulin, which p2-microglobulin is fused to a peptide forming an epitope binding to the MHC I.

WO2019/051091 Al describes an MHC I molecule that is covalently linked by a disulphide bond to a p2-microglobulin, wherein one of these is fused to a peptide forming an epitope binding to the MHC I. Object of the invention

It is an object of the invention to provide alternative MHC I molecules which preferably in comparison to natural MHC I molecules have an increased stability, e.g. against dissociation. More preferably, the invention shall provide MHC I molecules which are stable, e.g. against dissociation, in the absence of a peptide or other molecule binding to the MHC I molecule, and which preferably is soluble, e.g. without a transmembrane domain. It is a further object to provide a process for producing MHC I molecules which are stable and free from a peptide or other molecule bound to the MHC I. A further object is to provide an analytical process which uses a stable MHC I molecule, which optionally has no antigen bound to the alpha chain, and which is soluble, e.g. without a transmembrane domain. The MHC I molecules of the invention shall be free from a covalently bound cognate antigenic peptide binding to the alpha chain bound by at least one disulphide bond to the p2-microglobulin.

Description of the invention

Herein, the term major histocompatibility complex I (MHC I) includes the human leukocyte antigen I (HLA I), especially for proteins comprising or consisting of a heavy chain (alpha chain) of MHC I associated with a p2-microglobulin, which are e.g. available at

https://ebi.ac.uk/ipd/imgt/hla/. Generally, the MHC I can be a HLA A, HLA B, or HLA C, or, less preferably, HLA E, HLA F, or HLA G. Preferably, the MHC I molecules have the amino acid sequence of a human MHC I molecule, but including the mutations resulting in the changes of amino acids to Cystein as described in the following.

The invention achieves the object by the features of the claims, especially by providing a MHC I which has at least one covalent bond formed between an Cystein in the alpha chain and an amino acid residue of the p2-microglobulin, and which preferably has at least one pair of Cystein residues, of which one is in the alpha chain and one is in the P2-microglobulin, which Cystein residues form a disulfide bond between the alpha chain and the b2- microglobulin. The bond formed between at least one pair of amino acids, one of which amino acids of the pair is a Cystein in the alpha chain and one of which amino acids of the pair is an amino acid having a group that is reactive with a thiol group, which amino acid is located in the P2-microglobulin. The MHC I molecule of the invention specifically is free from a covalently bound peptide epitope that is a cognate antigen binding into the peptide binding cleft formed by the alpha chain that is bound by at least one disulphide bond to the P2-microglobulin. Accordingly, the MHC I molecule of the invention does not contain a cognate antigenic peptide that is linked to a terminus of the alpha chain or of the p2-microglobulin chain.

Accordingly, the MHC I molecule of the invention comprises or consists of an alpha chain and a p2-microglobulin which are linked by at least one disulfide bond formed between them, without a cognate antigenic peptide covalently bound to the MHC I molecule, and preferably without a transmembrane domain. As the MHC I molecule of the invention is devoid of a covalently bound antigenic peptide that could be bound by the MHC I molecule, e.g. the binding cleft of its alpha chain, only a free peptide, e.g. a separately solute peptide, to which the MHC I molecule has affinity, can be associated with the MHC I molecule, e.g. with the binding cleft of the alpha chain.

Generally herein, a cognate antigenic peptide is also referred to as antigen peptide or cognate antigen, and these terms define a peptide that has at least some affinity to the binding cleft of the MHC I molecule, especially to the binding cleft of the alpha chain of the MHC I molecule, when the alpha chain is associated with a p2-microglobulin Preferably, the cognate antigenic peptide is defined as a peptide that has a dissociation constant (K _D ) of 1 mM or lower. Determination of the dissociation constant preferably is in PBS buffer (7 mM

NaH2P04, 150 mM NaCl, pH 7.4) at 20°C and/or at 37°C.

The MHC I molecules of the invention due to the disulfide bond between the alpha chain and the p2-microglobulin can have increased stability and still have a native MHC I conformation as determined by binding of W6/32 antibody, e.g. when lacking the transmembrane domain, in the absence or presence of an antigenic peptide arranged in the binding cleft of the alpha chain. This increased stability was e.g. found in comparison to wild-type MHC I lacking the transmembrane domain in SDS-PAGE under non-reducing conditions.

Further, certain of the MHC I molecules of the invention have the advantage of being stable, e.g. stable in solution when devoid of a transmembrane domain, when expressed in animal cells, especially in cultivated human cells, and allow for loading of the MHC I binding site with a soluble cognate antigen, e.g. replacing a peptide bound to the MHC I by a cognate antigen by adding the cognate antigen to the solution containing the MHC I molecule.

During the preparation of the invention, it was found that some combinations of an alpha chain and a p2-microglobulin that contain Cysteines for forming disulphide bridges between these, devoid of cognate antigenic peptide fused to one of these, cannot be expressed in cultivated cells. For example, an MHC I consisting of an alpha chain of HLA I having a Cy stein in amino acid position 119 in combination with a p2-microglobulin containing a Cystein in amino acid position 1 was found to be not expressed in functional conformation, as determined by binding of W6/32 antibody. As additional examples, an MHC I consisting of an alpha chain of HLA I having a Cystein in amino acid position 120 in combination with a P2-microglobulin containing a Cystein in amino acid position 3, an MHC I consisting of an alpha chain of HLA I having a Cystein in amino acid position 96 in combination with a b2- microglobulin containing a Cystein in amino acid position 31, an MHC I consisting of an alpha chain of HLA I having a Cystein in amino acid position 96 in combination with a b2- microglobulin containing a Cystein in amino acid position 60, and an MHC I consisting of an alpha chain of HLA I having a Cystein in amino acid position 234 in combination with a b2- microglobulin containing a Cystein in amino acid position 8 were found to be not expressed in functional conformation, as determined by binding of W6/32 antibody. Determination of the binding of W6/32 antibody preferably is in PBS buffer (7 mM NaH2P04, 150 mM NaCl, pH 7.4) at 20°C and/or at 37°C.

MHC I, especially human HLA I and murine MHC I are known, e.g. from

https://www.ebi.ac.uk/ipd/imgt/hla/download.htmn. release 0.3.35, which can be downloaded from ftp://ftp.ebi.ac.uk/pub/databases/ipd/imgt/hla/ or from

https :// github . com/ANHIG/IMGTHL A/tree/Latest/ ali gnments. showing the homology of MHC I, especially of its alpha chains. The alpha chain of MHC I, especially of HLA I, includes amino acid sequences having a sequence identity of at least 90 %, at least 91 %, at least 92 %, at least 93 %, at least 94%, preferably of at least 95 %, e.g. at least 96 % or at least 97 % or at least 98 % or at least 99 % with one or more of SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 9, or SEQ ID NO: 10 to SEQ ID NO: 78. SEQ ID NO: 10 to SEQ ID NO: 78 are the amino acid sequences of A*01 :01, A*02:01, A*03:01, A* 11 :01, A*23:01, A*24:02, A*25:01, A*26:01, A*29:01, A*30:01, A*31 :01, A*32:01, A*33:01, A*34:01, A*36:01, A*43:01, A*66:01, A*68:01, A*69:01, A*74:01, A*80:01, B*07:02, B*08:01, B*13:01, B*14:01, B*15:01, B*18:01, B*27:01, B*35:01, B*37:01, B*38:01, B*39:01, B*40:01, B*41 :01, B*42:01, B*44:02, B*45:01, B*46:01, B*47:01, B*48:01, B*49:01, B*50:01, B*51 :01, B*52:01, B*53:01, B*54:01, B*55:01, B*56:01, B*57:01, B*58:01, B*59:01, B*67:01, B*78:01, B*81 :01, B*82:01, C*01 :02, C*02:02, C*03:02, C*04:01, C*05:01, C*06:02, C*07:01, C*08:01, C*12:02, C*14:02, C*15:02, C*16:01, C*17:01, C*18:01, each without the N-terminal signal peptide. Generally herein, amino acid sequences are related to from their N-terminus to their C-terminus. Preferably, the alpha chain of MHC I may at least comprise or consist of amino acids No. 3 to 272 of the alpha chain, e.g. of SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 9 or one of SEQ ID NO: 10 to SEQ ID NO: 78, which section of amino acids No. 3 to 272 can be determined to exclude its signal peptide, its transmembrane domain and/or its intracellular domain.

Optionally, the alpha chain may additionally contain one or more mutations contained in the group comprising or consisting of 74L, 223 A, 224E, 225D, 226A, 227K, 229A, 245V

(weakening or abolishing interaction with CD8), 227K and/or 228A, 139C and/or one of 84C and 85C (forming a disulfide bond within the alpha chain), and/or 115E.

The P2-microglobulin, especially the human or murine P2-microglobulin, includes amino acid sequences having a sequence identity of at least 90 %, at least 91 %, at least 92 %, at least 93 %, at least 94%, preferably of at least 95 %, e.g. at least 96 % or at least 97 % or at least 98 % with SEQ ID NO: 4.

The P2-microglobulin, especially the human or murine P2-microglobulin, may at least comprise or consist of amino acids No. 3 to 97 of the P2-microglobuin, determined excluding its signal peptide.

Optionally, the p2-microglobulin may additionally contain one or more mutations contained in the group comprising or consisting of K58E and/or S55V, with its amino acid sequence determined excluding its signal peptide.

For the p2-microglobulin, the Cy stein residue that forms a covalent bond to a residue of the alpha chain can be replaced by another amino acid, which may be a natural or a synthetic amino acid, which carries a group that is reactive with the thiol group of the Cystein of the alpha chain. The Cystein of the alpha chain and the amino acid of the p2-microglobulin which form a bond between the alpha chain and the p2-microglobulin, are defined as pairs of amino acids in positions of the alpha chain and of the P2-microglobulin. In the p2-microglobulin the amino acid that carries a group which is reactive with a thiol group can e.g. be selected from a thiol group, a maleimide group, a iodoacetamide group, and an alkene group. The amino acid that carries a group which is reactive with a thiol group can e.g. be introduced into the b2- microglobulin by derivatisation of a Cystein or by chemical, e.g. non-ribosomal, synthesis of the p2-microglobulin using a natural or a synthetic amino acid in the respective position of the at least one pair of amino acids. Chemical synthesis of peptides is generally known, e.g. by solution using protective group chemistry, or by solid-phase synthesis, e.g. according to Merrifield and further developments of this method. In embodiments in which the b2- microglobulin is produced by derivatisation of a Cystein or by chemical, e.g. non-ribosomal synthesis, the b2-hήϋp¾1oI>u1ΐh can be associated with the alpha chain to produce an MHC I molecule in which at least one covalent bond formed between the pairs of amino acids connects the alpha chain and the b2-hήϋp¾1oI>u1ΐh.

Herein, the amino acid in the position of the b2-hήϋp¾1oI>u1ΐh which is part of the at least one pair of amino acids, is generally represented by Cystein. In an embodiment, in the b2- microglobulin, the representative Cystein of the pair of amino acids is therefore Cystein or another natural or artificial, i.e. non-natural, amino acid having a group that is reactive with the thiol group of the Cystein of the alpha chain. The amino acids forming a covalent bond between the alpha chain and the b2-hήϋp¾1oI>u1ΐh herein are identified as pairs of amino acids, and as pairs of amino acid positions.

The amino acid (AA) positions of the alpha chain have been found in HLA A, as exemplified by SEQ ID NO: 1 (A*01 :01 :01 :01), in HLA B, as exemplified by SEQ ID NO: 2

(B*07:02:01 :01), as well as in HLA C, as exemplified by SEQ ID NO: 3 (C*01 :02:01 :01). SEQ ID NO: 1, NO: 2, NO: 3, and SEQ ID NO: 10 to SEQ ID NO: 78 are given without the N-terminal signal peptide. Fig. 6 shows a sequence alignment of SEQ ID NO: 1, NO: 2, and NO: 3, indicating some of the amino acid positions that can be mutated to Cystein for forming a disulfide bond with a Cystein of the b2-hήϋp¾1oI>u1ΐh. Preferably, with reference to SEQ ID NO: 1 for the alpha chain and with reference to SEQ ID NO: 4 as an example for the b2- microglobulin (given without the N-terminal signal peptide), at least one of the following pairs of Cystein residues as given in the table below is contained in the MHC I. The following table gives the positions of amino acids forming covalent bonds between the alpha chain and the p2-microglobulin, especially Cysteins forming pairs that each form a covalent bond, especially a disulfide bond between alpha chain and p2-microglobulin of MHC I, wherein the numbering refers to the AA position of the respective peptide chain without signal peptide. A preferred alpha chain has SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, and more preferably has SEQ ID NO: 9, or one of SEQ ID NO: 10 to SEQ ID NO: 78, which are given without a signal peptide.

The additional amino acid, which preferably is Cystein, which is arranged N-terminally of AA position No. 1 in the p2-microglobulin may be linked directly to the N-terminus of AA No. 1 of the p2-microglobulin, or may be linked with an intermediate spacer of 1 to 4 amino acids to the N-terminus of AA No. 1 of the p2-microglobulin According to the invention, no antigenic peptide is covalently bound to the MHC I molecule, e.g. to its alpha chain or its b2- microglobulin.

The following alleles of HLA I contain a Cystein in position 120 (counted without the signal peptide) of the alpha chain: A*03:227, A*36:05, B*08: 127, B*13:95, C*04:78, C*05:41 and C*07: 144.

For example, in the alpha chain

the Cystein in position No. 120 can be caused by a mutation resulting in G120C,

the Cystein in position No. 35 can be caused by a mutation resulting in R35C or Q35C, the Cystein in position No. 119 can be caused by a mutation resulting in D119C

the Cystein in position No. 236 can be caused by a mutation resulting in A236C,

the Cystein in position No. 23 can be caused by a mutation resulting in 123 C,

the Cystein in position No. 10 can be caused by a mutation resulting in T10C,

the Cystein in position No. 192 can be caused by a mutation resulting in H192C,

the Cystein in position No. 237 can be caused by a mutation resulting in G237C,

the Cystein in position No. 231 can be caused by a mutation resulting in V231C,

the Cystein in position No. 244 can be caused by a mutation resulting in W244C.

For example, in the P2-microglobulin

the Cystein in position No. 31 can be caused by a mutation resulting in H31C,

the Cystein in position No. 1 can be caused by a mutation resulting in 11C,

the Cystein in position No. 24 can be caused by a mutation resulting in N24C,

the Cystein in position No. 12 can be caused by a mutation resulting in R12C,

the Cystein in position No. 53 can be caused by a mutation resulting in D53C

the Cystein in position No. 54 can be caused by a mutation resulting in L54C, the Cystein in position No. 55 can be caused by a mutation resulting in S55C, the Cystein in position No. 98 can be caused by a mutation resulting in D98C.

In an embodiment, the amino acid in position No. 31, in position No. 1, in position No. 24, in position No. 12, in position No. 53, in position No. 54, in position No. 55, and/or in position No. 98 can be an amino acid other than Cystein, which amino acid has a group that is reactive with a thiol group, forming a bond by at least one pair of amino acids.

Optionally, an amino acid having a group that is reactive with the thiol group can be present as an additional amino acid at the C-terminus of p2-microglobulin Such an additional C- terminal amino acid that is reactive with a thiol group can form a bond with the Cystein of the alpha chain, especially with a Cystein at amino acid position No. 231 or 244 of the alpha chain.

Herein, especially in the following, sequences SEQ ID NO: 1, NO: 2, NO: 3, NO: 9 and NO: 10 to SEQ ID NO: 78, are simply designated and related to as the alpha chain.

Preferably, the MHC I molecule contains a Cystein in position No. 120 of the alpha chain and a Cystein in position No. 31 of SEQ ID NO: 4, resulting in pair No. 1,

and/or the MHC I molecule contains a Cystein in position No. 119 of the alpha chain and a Cystein in position No. 31 of SEQ ID NO: 4, resulting in pair No. 4,

and/or, less preferred, the MHC I molecule contains a Cystein in position No. 120 of the alpha chain and a Cystein in position No. 1 of SEQ ID NO: 4, resulting in pair No. 2, or also less preferred the MHC I molecule contains a Cystein in position No. 119 of the alpha chain and a Cystein in position No. 1 of SEQ ID NO: 4, resulting in pair No. 3,

and/or the MHC I molecule contains a Cystein in position No. 236 of the alpha chain and a Cystein in position No. 24 of SEQ ID NO: 4, resulting in pair No. 5 or the MHC I molecule contains a Cystein in position No. 236 of the alpha chain and a Cystein in position No. 12 of SEQ ID NO: 4, resulting in pair No. 6,

and/or the MHC I molecule contains a Cystein in position No. 237 of the alpha chain and a Cystein in position No. 24 of SEQ ID NO: 4, resulting in pair No. 7, or the MHC I molecule contains a Cystein in position No. 237 of the alpha chain and a Cystein in position No. 12 of SEQ ID NO: 4, resulting in pair No. 8, and/or the MHC I molecule contains a Cystein in position No. 23 of the alpha chain and a

Cy stein in position No. 54 of SEQ ID NO: 4, resulting in pair No. 9,

and/or the MHC I molecule contains a Cystein in position No. 10 of the alpha chain and a

Cystein in position No. 54 of SEQ ID NO: 4, resulting in pair No. 10,

or the MHC I molecule contains a Cystein in position No. 120 of the alpha chain and an additional Cystein arranged N-terminally of amino acid (AA) No. 1, which additional N- terminal Cystein could also be referred to as position No. 0 of SEQ ID NO: 4, resulting in pair

No. 11,

and/or the MHC I molecule contains a Cystein in position No. 192 of the alpha chain and a

Cystein in position No. 98 of SEQ ID NO: 4, resulting in pair No. 12,

and/or the MHC I molecule contains a Cystein in position No. 23 of the alpha chain and a

Cystein in position No. 55 of SEQ ID NO: 4, resulting in pair No. 13,

and/or the MHC I molecule contains a Cystein in position No. 10 of the alpha chain and a

Cystein in position No. 55 of SEQ ID NO: 4, resulting in pair No. 14,

and/or the MHC I molecule contains a Cystein in position No. 231 of the alpha chain and a

Cystein directly C-terminally to position No. 99 of SEQ ID NO: 4, resulting in pair No. 15, and/or the MHC I molecule contains a Cystein in position No. 244 of the alpha chain and a

Cystein directly C-terminally to position No. 99 of SEQ ID NO: 4, resulting in pair No. 16.

The MHC I molecule of the invention can contain exactly one of these pairs of Cysteins forming a disulfide bond between the alpha chain and the p2-microglobulin, or more of these pairs, e.g. exactly two or more of these pairs of Cysteins forming a disulfide bond between the alpha chain and the p2-microglobulin.

Two pairs of Cysteins forming a disulfide bond between the alpha chain and the b2- microglobulin can e.g. be selected from combinations of one pair of the group consisting of pairs 1, 2, 3, 4, 9, 10, 11, 12, 13, 14, 15, and 16, and one pair of the group consisting of pairs 5, 6, 7, 8; or combinations of one pair of the group consisting of pairs 1, 2, 4, 5, 6, 7, 8, 11,

12, 15, and 16, and one pair of the group consisting of pairs 3, 9, 10, 13, 14; or combinations of one pair of the group consisting of pairs 3, 5, 6, 7, 8, 9, 10, 12, 13, 14, 15, and 16, and one pair of the group consisting of pairs 1, 2, 4, and 11.

Examples of combinations of two of these pairs are e.g. pair No. 1 and pair No. 5, or pair No.

1 and pair No. 6, or pair No. 1 and pair No. 7, or pair No. 1 and pair No. 8, or pair No. 2 and pair No. 5, or pair No. 2 and pair No. 6, or pair No. 2 and pair No. 7, or pair No. 2 and pair No. 8, or pair No. 3 and pair No. 5, or pair No. 3 and pair No. 6, or pair No. 3 and pair No. 7, or pair No. 3 and pair No. 8, or pair No. 4 and pair No. 5, or pair No. 4 and pair No. 6, or pair No. 4 and pair No. 7, or pair No. 4 and pair No. 8, or pair No. 11 and pair No. 5, or pair No.

11 and pair No. 6, or pair No. 11 and pair No. 7, or pair No. 11 and pair No. 8, or pair No. 12 and pair No. 5, or pair No. 12 and pair No. 6, or pair No. 12 and pair No. 7, or pair No. 12 and pair No. 8, or pair No. 9 and pair No. 5, or pair No. 9 and pair No. 6, or pair No. 9 and pair No. 7, or pair No. 9 and pair No. 8, or pair No. 10 and pair No. 5, or pair No. 10 and pair No. 6, or pair No. 10 and pair No. 7, or pair No. 10 and pair No. 8, or pair No. 13 and pair No. 5, or pair No. 13 and pair No. 6, or pair No. 13 and pair No. 7, or pair No. 13 and pair No. 8, or pair No. 14 and pair No. 5, or pair No. 14 and pair No. 6, or pair No. 14 and pair No. 7, or pair No. 14 and pair No. 8, or pair No. 15 and pair No. 5, or pair No. 15 and pair No. 6, or pair No. 15 and pair No. 7, or pair No. 15 and pair No. 8, or pair No. 16 and pair No. 5, or pair No. 16 and pair No. 6, or pair No. 16 and pair No. 7, or pair No. 16 and pair No. 8, or pair No. 1 and pair No. 12, or pair No. 2 and pair No. 12, or pair No. 4 and pair No. 12, or pair No. 5 and pair No. 12, or pair No. 6 and pair No. 12, or pair No. 7 and pair No. 12, or pair No. 8 and pair No. 12, or pair No. 9 and pair No. 12, or pair No. 10 and pair No. 12, or pair No. 11 and pair No. 12, or pair No. 13 and pair No. 12, or pair No. 14 and pair No. 12, or pair No. 1 and pair No. 15, or pair No. 2 and pair No. 15, or pair No. 4 and pair No. 15, or pair No. 5 and pair No. 15, or pair No. 6 and pair No. 15, or pair No. 7 and pair No. 15, or pair No. 8 and pair No. 15, or pair No. 9 and pair No. 15, or pair No. 10 and pair No. 15, or pair No. 11 and pair No. 15, or pair No. 13 and pair No. 15, or pair No. 14 and pair No. 15, or pair No. 1 and pair No. 16, or pair No. 2 and pair No. 16, or pair No. 4 and pair No. 16, or pair No. 5 and pair No. 16, or pair No. 6 and pair No. 16, or pair No. 7 and pair No. 16, or pair No. 8 and pair No. 16, or pair No. 9 and pair No. 16, or pair No. 10 and pair No. 16, or pair No. 11 and pair No. 16, or pair No. 13 and pair No. 16, or pair No. 14 and pair No. 16, or pair No. 1 and pair No. 9, or pair No. 2 and pair No. 9, or pair No. 4 and pair No. 9, or pair No. 11 and pair No. 9, or pair No. 1 and pair No. 10, or pair No. 2 and pair No. 10, or pair No. 4 and pair No. 10, or pair No. 11 and pair No. 10, or pair No. 1 and pair No. 13, or pair No. 2 and pair No. 13, or pair No. 4 and pair No. 13, or pair No. 11 and pair No. 13, or pair No. 1 and pair No. 14, or pair No. 2 and pair No. 14, or pair No. 4 and pair No. 14, or pair No. 11 and pair No. 14. Optionally, the MHC I molecule of the invention can contain three of these pairs of Cysteins forming a disulfide bond between the alpha chain and the p2-microglobulin, preferably one pair of the group consisting of pairs 1, 2, 4, and 11, and/or one pair of the group consisting of pairs 5, 6, 7, 8, and/or one pair of the group consisting of pairs 3, 9, 10, 13, 14, and/or one pair of the group consisting of pairs 12, 15, and 16. These three pairs can e.g. be

pair No. 1 and pair No. 5 and pair No. 3, or

pair No. 1 and pair No. 6 and pair No. 3, or

pair No. 1 and pair No. 7 and pair No. 3, or

pair No. 1 and pair No. 8 and pair No. 3, or

pair No. 2 and pair No. 5 and pair No. 3, or

pair No. 2 and pair No. 6 and pair No. 3, or

pair No. 2 and pair No. 7 and pair No. 3, or

pair No. 2 and pair No. 8 and pair No. 3, or

pair No. 4 and pair No. 5 and pair No. 3, or

pair No. 4 and pair No. 6 and pair No. 3, or

pair No. 4 and pair No. 7 and pair No. 3, or

pair No. 4 and pair No. 8 and pair No. 3, or

pair No. 11 and pair No. 5 and pair No. 3, or

pair No. 11 and pair No. 6 and pair No. 3, or

pair No. 11 and pair No. 7 and pair No. 3, or

pair No. 11 and pair No. 8 and pair No. 3, or

pair No. 1 and pair No. 5 and pair No. 9, or

pair No. 1 and pair No. 6 and pair No. 9, or

pair No. 1 and pair No. 7 and pair No. 9, or

pair No. 1 and pair No. 8 and pair No. 9, or

pair No. 2 and pair No. 5 and pair No. 9, or

pair No. 2 and pair No. 6 and pair No. 9, or

pair No. 2 and pair No. 7 and pair No. 9, or

pair No. 2 and pair No. 8 and pair No. 9, or

pair No. 4 and pair No. 5 and pair No. 9, or

pair No. 4 and pair No. 6 and pair No. 9, or

pair No. 4 and pair No. 7 and pair No. 9, or

pair No. 4 and pair No. 8 and pair No. 9, or

pair No. 11 and pair No. 5 and pair No. 9, or pair No. 11 and pair No. 6 and pair No. 9, or pair No. 11 and pair No. 7 and pair No. 9, or pair No. 11 and pair No. 8 and pair No. 9, or pair No. 12 and pair No. 5 and pair No. 9, or pair No. 12 and pair No. 6 and pair No. 9, or pair No. 12 and pair No. 7 and pair No. 9, or pair No. 12 and pair No. 8 and pair No. 9, or pair No. 1 and pair No. 5 and pair No. 10, or pair No. 1 and pair No. 6 and pair No. 10, or pair No. 1 and pair No. 7 and pair No. 10, or pair No. 1 and pair No. 8 and pair No. 10, or pair No. 2 and pair No. 5 and pair No. 10, or pair No. 2 and pair No. 6 and pair No. 10, or pair No. 2 and pair No. 7 and pair No. 10, or pair No. 2 and pair No. 8 and pair No. 10, or pair No. 4 and pair No. 5 and pair No. 10, or pair No. 4 and pair No. 6 and pair No. 10, or pair No. 4 and pair No. 7 and pair No. 10, or pair No. 4 and pair No. 8 and pair No. 10, or pair No. 11 and pair No. 5 and pair No. 10, or pair No. 11 and pair No. 6 and pair No. 10, or pair No. 11 and pair No. 7 and pair No. 10, or pair No. 11 and pair No. 8 and pair No. 10, or pair No. 12 and pair No. 5 and pair No. 10, or pair No. 12 and pair No. 6 and pair No. 10, or pair No. 12 and pair No. 7 and pair No. 10, or pair No. 12 and pair No. 8 and pair No. 10, or pair No. 1 and pair No. 5 and pair No. 13, or pair No. 1 and pair No. 6 and pair No. 13, or pair No. 1 and pair No. 7 and pair No. 13, or pair No. 1 and pair No. 8 and pair No. 13, or pair No. 2 and pair No. 5 and pair No. 13, or pair No. 2 and pair No. 6 and pair No. 13, or pair No. 2 and pair No. 7 and pair No. 13, or pair No. 2 and pair No. 8 and pair No. 13, or pair No. 4 and pair No. 5 and pair No. 13, or pair No. 4 and pair No. 6 and pair No. 13, or pair No. 4 and pair No. 7 and pair No. 13, or pair No. 4 and pair No. 8 and pair No. 13, or pair No. 11 and pair No. 5 and pair No. 13, or pair No. 11 and pair No. 6 and pair No. 13, or pair No. 11 and pair No. 7 and pair No. 13, or pair No. 11 and pair No. 8 and pair No. 13, or pair No. 1 and pair No. 5 and pair No. 14, or pair No. 1 and pair No. 6 and pair No. 14, or pair No. 1 and pair No. 7 and pair No. 14, or pair No. 1 and pair No. 8 and pair No. 14, or pair No. 2 and pair No. 5 and pair No. 14, or pair No. 2 and pair No. 6 and pair No. 14, or pair No. 2 and pair No. 7 and pair No. 14, or pair No. 2 and pair No. 8 and pair No. 14, or pair No. 4 and pair No. 5 and pair No. 14, or pair No. 4 and pair No. 6 and pair No. 14, or pair No. 4 and pair No. 7 and pair No. 14, or pair No. 4 and pair No. 8 and pair No. 14, or pair No. 11 and pair No. 5 and pair No. 14, or pair No. 11 and pair No. 6 and pair No. 14, or pair No. 11 and pair No. 7 and pair No. 14, or pair No. 11 and pair No. 8 and pair No. 14, or pair No. 1 and pair No. 5 and pair No. 12, or pair No. 1 and pair No. 6 and pair No. 12, or pair No. 1 and pair No. 7 and pair No. 12, or pair No. 1 and pair No. 8 and pair No. 12, or pair No. 2 and pair No. 5 and pair No. 12, or pair No. 2 and pair No. 6 and pair No. 12, or pair No. 2 and pair No. 7 and pair No. 12, or pair No. 2 and pair No. 8 and pair No. 12, or pair No. 4 and pair No. 5 and pair No. 12, or pair No. 4 and pair No. 6 and pair No. 12, or pair No. 4 and pair No. 7 and pair No. 12, or pair No. 4 and pair No. 8 and pair No. 12, or pair No. 11 and pair No. 5 and pair No. 12, or pair No. 11 and pair No. 6 and pair No. 12, or pair No. 11 and pair No. 7 and pair No. 12, or pair No. 11 and pair No. 8 and pair No. 12, or pair No. 1 and pair No. 5 and pair No. 15, or pair No. 1 and pair No. 6 and pair No. 15, or pair No. 1 and pair No. 7 and pair No. 15, or pair No. 1 and pair No. 8 and pair No. 15, or pair No. 2 and pair No. 5 and pair No. 15, or pair No. 2 and pair No. 6 and pair No. 15, or pair No. 2 and pair No. 7 and pair No. 15, or pair No. 2 and pair No. 8 and pair No. 15, or pair No. 4 and pair No. 5 and pair No. 15, or pair No. 4 and pair No. 6 and pair No. 15, or pair No. 4 and pair No. 7 and pair No. 15, or pair No. 4 and pair No. 8 and pair No. 15, or pair No. 11 and pair No. 5 and pair No. 15, or pair No. 11 and pair No. 6 and pair No. 15, or pair No. 11 and pair No. 7 and pair No. 15, or pair No. 11 and pair No. 8 and pair No. 15, or pair No. 1 and pair No. 5 and pair No. 16, or pair No. 1 and pair No. 6 and pair No. 16, or pair No. 1 and pair No. 7 and pair No. 16, or pair No. 1 and pair No. 8 and pair No. 16, or pair No. 2 and pair No. 5 and pair No. 16, or pair No. 2 and pair No. 6 and pair No. 16, or pair No. 2 and pair No. 7 and pair No. 16, or pair No. 2 and pair No. 8 and pair No. 16, or pair No. 4 and pair No. 5 and pair No. 16, or pair No. 4 and pair No. 6 and pair No. 16, or pair No. 4 and pair No. 7 and pair No. 16, or pair No. 4 and pair No. 8 and pair No. 16, or pair No. 11 and pair No. 5 and pair No. 16, or pair No. 11 and pair No. 6 and pair No. 16, or pair No. 11 and pair No. 7 and pair No. 16, or pair No. 11 and pair No. 8 and pair No. 16, or pair No. 1 and pair No. 12 and pair No. 3, or pair No. 1 and pair No. 15 and pair No. 3, or pair No. 1 and pair No. 16 and pair No. 3, or pair No. 2 and pair No. 12 and pair No. 3, or pair No. 2 and pair No. 15 and pair No. 3, or pair No. 2 and pair No. 16 and pair No. 3, or pair No. 4 and pair No. 12 and pair No. 3, or pair No. 4 and pair No. 15 and pair No. 3, or pair No. 4 and pair No. 16 and pair No. 3, or pair No. 11 and pair No. 12 and pair No. 3, or pair No. 11 and pair No. 15 and pair No. 3, or pair No. 11 and pair No. 16 and pair No. 3, or pair No. 1 and pair No. 12 and pair No. 9, or pair No. 1 and pair No. 15 and pair No. 9, or pair No. 1 and pair No. 16 and pair No. 9, or pair No. 2 and pair No. 12 and pair No. 9, or pair No. 2 and pair No. 15 and pair No. 9, or pair No. 2 and pair No. 16 and pair No. 9, or pair No. 4 and pair No. 12 and pair No. 9, or pair No. 4 and pair No. 15 and pair No. 9, or pair No. 4 and pair No. 16 and pair No. 9, or pair No. 11 and pair No. 12 and pair No. 9, or pair No. 11 and pair No. 15 and pair No. 9, or pair No. 11 and pair No. 16 and pair No. 9, or pair No. 1 and pair No. 12 and pair No. 10, or pair No. 1 and pair No. 15 and pair No. 10, or pair No. 1 and pair No. 16 and pair No. 10, or pair No. 2 and pair No. 12 and pair No. 10, or pair No. 2 and pair No. 15 and pair No. 10, or pair No. 2 and pair No. 16 and pair No. 10, or pair No. 4 and pair No. 12 and pair No. 10, or pair No. 4 and pair No. 15 and pair No. 10, or pair No. 4 and pair No. 16 and pair No. 10, or pair No. 11 and pair No. 12 and pair No. 10, or pair No. 11 and pair No. 15 and pair No. 10, or pair No. 11 and pair No. 16 and pair No. 10, or pair No. 1 and pair No. 12 and pair No. 13, or pair No. 1 and pair No. 15 and pair No. 13, or pair No. 1 and pair No. 16 and pair No. 13, or pair No. 2 and pair No. 12 and pair No. 13, or pair No. 2 and pair No. 15 and pair No. 13, or pair No. 2 and pair No. 16 and pair No. 13, or pair No. 4 and pair No. 12 and pair No. 13, or pair No. 4 and pair No. 15 and pair No. 13, or pair No. 4 and pair No. 16 and pair No. 13, or pair No. 11 and pair No. 12 and pair No. 13, or pair No. 11 and pair No. 15 and pair No. 13, or pair No. 11 and pair No. 16 and pair No. 13, or pair No. 1 and pair No. 12 and pair No. 14, or pair No. 1 and pair No. 15 and pair No. 14, or pair No. 1 and pair No. 16 and pair No. 14, or pair No. 2 and pair No. 12 and pair No. 14, or pair No. 2 and pair No. 15 and pair No. 14, or pair No. 2 and pair No. 16 and pair No. 14, or pair No. 4 and pair No. 12 and pair No. 14, or pair No. 4 and pair No. 15 and pair No. 14, or pair No. 4 and pair No. 16 and pair No. 14, or pair No. 11 and pair No. 12 and pair No. 14, or pair No. 11 and pair No. 15 and pair No. 14, or pair No. 11 and pair No. 16 and pair No. 14, or pair No. 5 and pair No. 12 and pair No. 3, or pair No. 5 and pair No. 15 and pair No. 3, or pair No. 5 and pair No. 16 and pair No. 3, or pair No. 5 and pair No. 12 and pair No. 9, or pair No. 5 and pair No. 15 and pair No. 9, or pair No. 5 and pair No. 16 and pair No. 9, or pair No. 5 and pair No. 12 and pair No. 10, or pair No. 5 and pair No. 15 and pair No. 10, or pair No. 5 and pair No. 16 and pair No. 10, or pair No. 5 and pair No. 12 and pair No. 13, or pair No. 5 and pair No. 15 and pair No. 13, or pair No. 5 and pair No. 16 and pair No. 13, or pair No. 5 and pair No. 12 and pair No. 14, or pair No. 5 and pair No. 15 and pair No. 14, or pair No. 5 and pair No. 16 and pair No. 14, or pair No. 6 and pair No. 12 and pair No. 3, or pair No. 6 and pair No. 15 and pair No. 3, or pair No. 6 and pair No. 16 and pair No. 3, or pair No. 6 and pair No. 12 and pair No. 9, or pair No. 6 and pair No. 15 and pair No. 9, or pair No. 6 and pair No. 16 and pair No. 9, or pair No. 6 and pair No. 12 and pair No. 10, or pair No. 6 and pair No. 15 and pair No. 10, or pair No. 6 and pair No. 16 and pair No. 10, or pair No. 6 and pair No. 12 and pair No. 13, or pair No. 6 and pair No. 15 and pair No. 13, or pair No. 6 and pair No. 16 and pair No. 13, or pair No. 6 and pair No. 12 and pair No. 14, or pair No. 6 and pair No. 15 and pair No. 14, or pair No. 6 and pair No. 16 and pair No. 14, or pair No. 7 and pair No. 12 and pair No. 3, or pair No. 7 and pair No. 15 and pair No. 3, or pair No. 7 and pair No. 16 and pair No. 3, or pair No. 7 and pair No. 12 and pair No. 9, or pair No. 7 and pair No. 15 and pair No. 9, or pair No. 7 and pair No. 16 and pair No. 9, or pair No. 7 and pair No. 12 and pair No. 10, or pair No. 7 and pair No. 15 and pair No. 10, or

pair No. 7 and pair No. 16 and pair No. 10, or

pair No. 7 and pair No. 12 and pair No. 13, or

pair No. 7 and pair No. 15 and pair No. 13, or

pair No. 7 and pair No. 16 and pair No. 13, or

pair No. 7 and pair No. 12 and pair No. 14, or

pair No. 7 and pair No. 15 and pair No. 14, or

pair No. 7 and pair No. 16 and pair No. 14, or

pair No. 8 and pair No. 12 and pair No. 3, or

pair No. 8 and pair No. 15 and pair No. 3, or

pair No. 8 and pair No. 16 and pair No. 3, or

pair No. 8 and pair No. 12 and pair No. 9, or

pair No. 8 and pair No. 15 and pair No. 9, or

pair No. 8 and pair No. 16 and pair No. 9, or

pair No. 8 and pair No. 12 and pair No. 10, or

pair No. 8 and pair No. 15 and pair No. 10, or

pair No. 8 and pair No. 16 and pair No. 10, or

pair No. 8 and pair No. 12 and pair No. 13, or

pair No. 8 and pair No. 15 and pair No. 13, or

pair No. 8 and pair No. 16 and pair No. 13, or

pair No. 8 and pair No. 12 and pair No. 14, or

pair No. 8 and pair No. 15 and pair No. 14, or

pair No. 8 and pair No. 16 and pair No. 14.

Generally, as an option, the p2-microglobulin can comprise a mutation resulting in an S55V amino acid exchange as an additional mutation.

Preferably, the alpha chain contains a signal peptide sequence directly adjacent to the N- terminus the alpha chain, e.g. at the N-terminus of amino acid No. 1 of SEQ ID NO: 1, SEQ ID NO: 2, or SEQ ID NO: 3, or one of SEQ ID NO: 10 to SEQ ID NO: 78.

Preferably, the p2-microglobulin contains a signal peptide sequence directly adjacent to its N- terminus, or in case of an additional Cystein arranged at the N-terminus of the b2- microglobulin (e.g. pair No. 11 or No. 12) a signal peptide sequence at the N-terminus of the Cystein bound to the N-terminus of the P2-microglobulin.

It was found that the MHC I molecules of the invention can be expressed in eurkaryotic cells, preferably in mammalian cells or insect cells or yeast or fungal cells, with the disulfide bonds formed between the Cystein residues formed by the pairs of Cystein residues.

It is an advantage of MHC I molecules according to the invention that they are stable to dissociation also in the absence of a cognate antigen binding to the alpha chain of the MHC I, wherein the MHC I can be composed of its alpha chain and p2-microglobulin. Therefore, the invention preferably allows the expression of the MHC I molecules in the absence of peptide or other molecule binding to the MHC I for production of the MHC I molecule without a peptide or other molecule bound to the binding groove of the alpha chain, alternatively, the MHC I molecules may be expressed with a peptide binding to its binding groove, followed by removing the bound peptide, e.g. by heating for dissociating the peptide from the binding groove of the MHC I, optionally followed by separating the peptide, e.g. by chromatography. Herein, a peptide or other molecule binding to the MHC I, especially to the alpha chain of MHC I, is also termed cognate antigen.

Generally preferred, MHC I molecules of the invention can be produced by expression from nucleic acid sequences encoding the MHC I molecule in mammalian cell culture, in insect cell culture, or in another eukaryotic cell culture, e.g. yeast or in a fungus. Optionally, the cell expressing the MHC I molecule and the culture, e.g. cultivating medium, does not contain a cognate antigen binding to the MHC I molecule, e.g. when expressing the MHC I molecule in yeast. Alternatively, the MHC I molecules can be expressed in a prokaryote, i.e. in a bacterium, both the alpha chain and the p2-microglobulin together in one bacterial cell or in separate bacterial cells, e.g. in separate bacterial cultures, followed by isolation of the alpha chain and the p2-microglobulin of the MHC I molecules, e.g. as inclusion bodies, respectively of the separate alpha chain and p2-microglobulin, with subsequent re-folding of the alpha chain and the P2-microglobulin combined to form MHC I molecules having the disulfide bond or binds, and separation of these molecules from other protein associates, e.g. from denatured single alpha chain or denatured P2-microglobulin or from multimers. For expression of MHC I molecules in bacterial cells, it is preferred that the alpha chain is devoid of a transmembrane domain. The MHC I molecules of the invention preferably have a conformation that is considered close to or essentially identical to the natural conformation of natural MHC I consisting of alpha chain, p2-microglobulin in association with an antigen bound to the alpha chain. This natural conformation of the MHC I molecules of the invention is e.g. shown by the binding of W6/32 antibody to these MHC I molecules.

MHC I molecules of the invention can be expressed from a first expression cassette encoding the alpha chain and a separate second expression cassette encoding the p2-microglobulin. Optionally, MHC I molecules of the invention are expressed from one expression cassette under the control of one promoter, the expression cassette containing a coding sequence, from 5 '-terminus to 3 '-terminus for an optional first signal peptide - alpha chain - optional tag sequence - ribosomal skip sequence - optional second signal peptide - p2-microglobulin.

The ribosomal skip sequence enables the production of the alpha chain and of the b2- microglobulin as separate peptides in close proximity, the close proximity including proximity of translation of the peptides both in time and localisation. Further, the ribosomal skip sequence preferably results in the stoichiometric expression of the alpha chain and of the P2-microglobulin as two separate proteins.

The MHC I molecules of the invention can be expressed from separate expression cassettes contained in one cell. This indicates that the alpha chain and the p2-microglobulin, when expressed from separate expression cassettes, combine to a complete MHC I molecule, e.g. in the endoplasmatic reticulum.

The optional first signal peptide preferably is a secretion signal peptide sequence, which is preferably directly linked to the N-terminus of the alpha chain, and which may e.g. be selected from natural signal peptides of MHC I molecules. Examples of suitable signal peptide sequences originating from alpha chains of natural HLA molecules are given as SEQ ID NO: 5, SEQ ID NO: 6, and SEQ ID NO: 7.

The optional second signal peptide preferably is a secretion signal peptide sequence, which is preferably directly linked to the N-terminus of the p2-microglobulin A signal peptide of the human p2-microgloblin is given as SEQ ID NO: 8. The second signal peptide may have the same amino acid sequence as the first signal peptide. The optional tag sequence may be selected from the V5 tag, streptavidin binding peptide (SBP) tag, and/or His tag, e.g. Hisio tag.

The ribosomal skip sequence may be the 2 A sequence of porcine teschovirus-1 (P2A), the 2 A sequence of Thoseaasigna virus (T2A), the 2A sequence of equine rhinitis A virus (ERAV), or the 2 A sequence of foot-and-mouth-disease virus (FMDV).

The invention is described in greater detail by way of examples with reference to the figures, which show in

- Fig. 1 a schematic representation of a preferred expression cassette for the MHC I,

- Fig. 2 results of detection of embodiments of MHC I molecules, detected with W6/32 antibody,

- Fig. 3 concentrations of cell culture supernatant of cells expressing embodiments of MHC I molecules and of comparative MHC I molecules,

- Fig. 4 an SDS-PAGE of purified MHC I according to the invention,

- Fig. 5 results of an antibody detection assay, and

- Fig. 6 shows a sequence alignment of exemplary alpha chains of HLA I molecules, indicating AA positions that can be changed to Cystein by a |,

- Fig. 7 a schematic representation of expression cassettes for MHC I alpha chain and P2-microglobulin,

- Fig. 8 concentrations of cell culture supernatant of cells expressing embodiments of MHC I molecules and of comparative MHC I molecules,

- Fig. 9 concentrations of cell culture supernatant of cells expressing embodiments of MHC I molecules and of comparative MHC I molecules,

- Fig. 10 Western Blot of purified cell culture supernatant of cells expressing

embodiments of MHC I molecules and of comparative MHC I molecules

Example 1 : Expression of a stabilized MHC containing one disulfide bond

Fig. 1 shows a preferred arrangement of sequences encoding the MHC I of the invention, which was used in this Example. From N-terminus to C-terminus, the nucleic acid construct contains directly adjacent one another the coding sequence of a first secretion signal peptide (SP) directly linked to the coding sequence of the alpha chain (alpha chain of MHC class I), directly linked to a tag sequence, herein comprising the tags V5 tag (V5), SBP tag (SBP), and His 10 (Hisio), the ribosomal skip sequence P2A, a second secretion signal peptide (SP), and P2-microglobulin (b2-ih). As examples for stabilized MHC I molecules which contain only one disulfide bond connecting the alpha chain and the p2-microglobulin, the coding sequence of the alpha chain of HLA-B*55:01 was modified by primer mutagenesis to obtain the G120C mutant which also contained the natural secretion signal peptide coding sequence, and the coding sequence for p2-microglobulin was modified to the H31C mutant which also contained the natural N- terminal secretion signal peptide coding sequence. This combination of Cystein residues corresponds to the pair No. 1 of Cystein residues. Accordingly, the coding sequence did not encode the transmembrane domain of the alpha chain.

These coding sequences were cloned into an expression vector, containing under the control of a SFFV promoter from 5 '-terminus to 3 '-terminus the coding sequence for the first signal peptide - alpha chain - V5 tag, SBP Tag, HislO tag sequence - the ribosomal skip sequence P2A - second signal peptide - P2-microglobulin.

Similarly, the A236C mutant of the alpha chain of HLA-B*55:01 was expressed in the vector with the R12C mutant of P2-microglobulin, corresponding to pair No. 6 of Cystein residues, and the 123 C mutant of the alpha chain of HLA-B*55:01 was expressed in the vector with the L54C mutant of P2-microglobulin, corresponding to pair No. 9 of Cystein residues, and the A236C mutant of the alpha chain of HLA-B*55:01 was expressed in the vector with the N24C mutant of P2-microglobulin, corresponding to pair No. 5 of Cystein residues.

In addition to these mutations of Cystein residues replacing the original amino acids, the b2- microglobulin contained the S55V mutation as described by Shields et al. (1998).

For expression, HEK293 cells were transfected with a viral particle containing the expression cassette in a viral vector. The cells were cultivated under cell culture conditions in DMEM- F12 medium at 37 °C in a 5% CO2 atmosphere. On day 3, culture supernatant was removed, concentrated by isopropanol precipitation and analysed by polyacrylamide gel electrophoresis (SDS-PAGE) for aliquots under non-reducing conditions without dithiothreitol (DTT) or under reducing conditions with DTT added to the sample loading buffer, with subsequent Western blotting onto a PVDF membrane. For detection, the antibody PR1F11F11 that is specific for the V5 tag on the alpha chain, or the BBM.l antibody that is specific for the b2- microglobulin, each conjugated to horse-radish peroxidase (HRP) for detection, was used. Using detection of the alpha chain in the Western blot, the comparative wild-type MHC I molecule and the mutant MHC I molecules, all without transmembrane domain, expression and secretion of the MHC I molecules was found in soluble form into the culture supernatant. The results indicate that in the mutant MHC I molecules the disulfide bond is formed between the alpha chain and the p2-microglobulin, resulting in a stabilized association of the alpha chain and the P2-microglobulin.

The p2-microglobulin under non-reducing sample preparation conditions could be shown to be associated with the alpha chain, and under reducing sample preparation conditions could be detected at an apparent size corresponding to the separate p2-microglobulin.

Fig. 2 shows the results of an ELISA, normalized to wildtype soluble MHC I, in which culture supernatant of cells expressing MHC I molecules was immobilized on a BBM.1 antibody coated surface and detection was by W6/32 antibody, which is generally regarded as indicating a non-denatured or natural conformation of MHC I molecules. For comparison, supernatant of a HEK293 cell expressing the wild-type alpha chain without transmembrane domain and the S55V variant of p2-microglobulin (WT/S55V) and supernatant of wild-type alpha chain without transmembrane domain and the wild-type of p2-microglobulin (WT/WT) were included. The stabilized MHC I molecules were A236C in the alpha chain and N24C in the p2-microglobulin (pair No. 5, A236C/N24C_S55V), also carrying the S55V mutation, and G120C in the alpha chain and H31C in the P2-microglobulin (pair No. 1,

G120C/H31C S55V), also carrying the S55V mutation in the p2-microglobulin The results show that the MHC I molecules stabilized by a disulfide bridge have a natural conformation, as determined by binding of the W6/32 antibody.

Further, HEK293 cells were transfected with a nucleic acid construct encoding the stabilized mutant of HLA-B* 55:01, carrying the G120C mutation in the alpha chain and the H31C mutation in the p2-microglobulin in the same vector as described above (pair No. 1,

G120C/H31C), or the cells were transfected with a nucleic acid construct encoding the stabilized mutant of HLA-B*55:01, carrying the G120C mutation in the alpha chain and the IIC mutation in the p2-microglobulin in the same vector as described above (pair No. 2, G120C/I1C) as a repetition, cell culture supernatants were shown to contain the stabilized mutant MHC I molecules that have the disulfide bond between the alpha chain and the b2- microglobulin. Using a Western blot of cell culture supernatants with sample preparation under non-reducing conditions, i.e. using SDS containing sample buffer without DTT (-DTT) and detection with HRP-conjugated antibody directed against the p2-microglobulin (BBM.l-HRP), it could be shown that the stabilized MHC I molecules and also the natural B*55:01 molecule used as a comparison, which was expressed in the same cell type from the nucleic acid construct containing the same signal peptides and regulatory elements, contain the p2-microglobulin in association with the alpha chain (band of apparent 58 kDa). In contrast, the wild-type comparative natural B*55:01 molecule shows a more prominent band of P2-microglobulin that is dissociated from the alpha chain. An additional detected band of apparent 14-15 kDa in the lanes of the stabilized MHC I molecules was assumed to be p2-microglobulin that has not formed a disulfide bond with an alpha chain.

Using a Western blot of an SDS-PAGE using reducing sample preparation conditions (+DTT) of the same cell culture supernatants with detection by a HRP-conjugated antibody

(PR1F11F11-HRP) directed against the V5 tag on the alpha chain, it could be shown that the stabilized MHC I molecules under reducing conditions show a lower apparent molecular weight than under non-reducing conditions, which weight corresponds to the size of the alpha chain only, indicating dissociation of the p2-microglobulin chain under reducing conditions.

Analysis of an ELISA as above using immobilized cell culture supernatants and detection by anti-tag antibody (anti-V5His) or detection by W6/32 antibody could show that in this assay the stabilized MHC I molecule containing the disulfide bond formed by pair No. 1,

G120C/H31C (B5501_G120C_b2m_H31C) is strongly recognized by the W6/32 antibody, indicating a conformation corresponding to that of natural MHC I. A signal for the anti-tag antibody was detected at lower intensity, indicating that the tag is present in the stabilized MHC I molecule but is assumed to be less accessible than the binding site for W6/32 antibody.

Fig. 3 shows concentrations of MHC I molecules in stably transduced cell culture supernatant in a 25 cm ² cell culture flask, detected using a sandwich ELISA with the BBM.1 and the W6/32 antibody. This shows that along with the positive control of wildtype MHC I containing both alpha chain and p2-microglobulin in wild-type form (wt/wt), the mutant having the S55V in the p2-microglobulin with a wild-type alpha chain (wt/S55V) and the stabilized MHC I G120C/H31C (G120C/H31C_S55V) including the S55V in the b2- microglobulin, and the stabilized MHC I A236C/N24C (A236C/N24C_S55C), all without transmembrane domain, are expressed in the cell culture from the construct containing the coding sequences under the control of one promoter only with a ribosomal skip sequence arranged between these.

From separate cultures, the cell culture supernatant was collected and buffered into PBS (7 mM NaH2P04, 150 mM NaCl, pH 7.4). HLA-B*55:01 protein was bound to an immobilized metal affinity chromatography (IMAC) column via the histidine tag, the column was washed, and bound protein was eluted with imidazol, to yield (wt/wt), (wt/S55V), and

(G120C/H31C S55V) in separate preparations. The eluates were buffered into PBS, optionally containing a preserving agent, e.g. ProClin 300 (obtainable from Sigma Aldrich), adjusted to 400 to 500 ng/mL, and stored at 4°C.

Fig. 4 shows a Coomassie stained SDS-PAGE of the preparation, under reducing and non reducing conditions. In the lanes of the MHC I (HLA-B*55:01) wild-type (WT) alpha chain associated with the wild-type (WT) P2-microglobulin (B2M) or with the S55V mutant b2- microglobulin, the sample without DTT (-) migrate to apparent lower molecular weights, which is assumed to be due to increased compactness of the alpha chain and p2-microglobulin in their associated state, than the sample with DTT (+). For the MHC I of the alpha chain having the G120C mutation (G120C) and the P2-microglobulin having the H31C and the S55V mutations (H31C/S55V), the lane without DTT (-) shows a prominent band of higher molecular weight than the prominent band of the wild-types, and the lane with DTT (-) shows a lower molecular weight than the one with DTT. Only in the presence of DTT, the MHC I of the alpha chain having the G120C mutation (G120C) and the P2-microglobulin having the H31C and the S55V mutations shows a band of 11 kDa, corresponding to an isolated b2- microglobulin chain, and the lane without DTT shows no band corresponding to the isolated b2-hύϋp¾^u1ΐh chain. The results confirm the disulfide bond between alpha chain and isolated b2-ihΐϋp¾^u1ΐh chain for the exemplary MHC I of the invention.

Example 2: Analysis of antibody from a serum sample

Samples: Positive serum known to contain an anti-HLA B*55:01 antibody; Negative serum without anti-HLA antibodies; Positive control: anti-B2M antibody (BBM. l), HRP-coupled. Protocol:

Wash the required amount of Dynabeads M-280 streptavidin (M280SA; Thermo Fisher, 2.5 mΐ per well of a 96-well-plate) with TSP (50 mM tris(hydroxymethyl)amino methane- hydrochloride, 0.5 mM ethylenediamine tetraacetic acid (EDTA), 100 mM sodium chloride,

0.1% v/v ProClin300, 5% w/v saccharose, adjust to pH 8.0)/10% PBSTB (7 mM NaH2P04, 150 mM NaCl, pH 7.4, 0,05 % v/v Tween 20, 2 % w/v bovine serum albumin), resuspend in starting volume. Distribute 2.5 mΐ M280SA into the required wells of a 96 well plate (Nunc, 96F). Add 0.5 pg of HLA-B*55:01 (G120C/H31C_S55V) to 2.5 mΐ M280SA beads in 100 mΐ TSP/10% PBSTB. Resuspend the beads, incubate overnight at 2 to 8°C, wash 5x with a magnetic washer (HydroFlex microplate washer, Tecan) with PIWP (1.9 mM di-sodium hydrogenphosphate (NaiHPCri), 0.07 mM sodium dihydrogenphosphate (NaHiPCri), 0.05 % w/v Tween 20, 1 mM 2-bromo-2-nitro-l,3-propandiol). Add 100 mΐ serum 1 :8 dilution in BAP (PBS, 1% w/v bovine serum albumin, 0.05% Tween20, 0.1% w/v ProClin300), only BAP, or anti -b2-hiί crogl obul i n antibody at dilution of 1 :2,500 in BAP, respectively. Resuspend the beads and incubate for 45 min at room temperature. Wash 5 times with a magnetic washer with PIWP. Add to all wells 100 mΐ anti human IgG antibody (MCA647P, Bio-Rad, HRP coupled) at dilution of 1 :20,000 in PIKV (50% v/v fetal calf serum, 1.06 %v/v

tris(hydroxymethyl)amino methane, 0.5 % w/v PVP-10 (polyvinyl pyrrolidone, average mol. weight 10 000), 0.45% w/v NaCl, 0.1 % w/v Ca-propionate, optionally 0.2 % v/v ProClin 300, 0.05 % w/v 8-anilino-l -naphthalene sulfonic acid (ANS), 0.025 % penicillin, 0.025 % streptomycin, adjusted to pH 7.35). Incubate for 45 min at room temperature in the dark. Wash 5 times with a magnetic washer with PIWP. Add 100 mΐ TMB ONE (obtained from Kem-En-Tec Diagnostics, Denmark). Incubate for 15 min at room temperature in the dark while shaking. Stop reaction by adding 100 mΐ stop solution (5% v/v H2SO4). Absorbance was measured in a microtiter plate reader (Powerwave 320, Bio-Tek) at 450 nm and at 690 nm (background scatter from the beads), from which the difference was calculated (OD450-690 nm). Fig. 5 shows the result of the experiment. The positive serum shows a clearly distinguishable signal from the negative serum and the no-antibody -well. The positive control confirms the presence of antigen in the well.

Example 3: Expression of a stabilized MHC containing one disulfide bond

Fig. 7 shows the arrangement of expression vectors used in this example, vector 1 coding for the alpha chain of MHC I and vector 2 coding for p2-microglobulin. From N-terminus to C- terminus, the nucleic acid construct vector 1 contains directly adjacent one another the coding sequence of a secretion signal peptide (SP) directly linked to the coding sequence of the alpha chain (alpha chain of MHC class I), directly linked to a tag sequence (VX), herein comprising the tags V5 tag (V5), and HislO (Hisio). Vector 2 contains a signal peptide (SP), directly linked to the P2-microglobulin (b2-ih) coding sequence.

In this example the MHC I alleles HLA-A*02:01, HLA-A*34:01, HLA-A*68:01, HLA- B* 14:02 and HLA-C*07:02 were chosen for comparison of expression levels of different Cystein mutants. For each Cystein mutant one expression vector was created by primer mutagenesis. Cystein residues on alpha chain are A236C, D119C, G120C, and I23C. Cystein residues on p2-microglobulin are I1C, N24C, H31C, and L54C.

The A236C mutant of the alpha chain was expressed together with the N24C mutant of b2- microglobulin, corresponding to pair No. 5 of Cystein residues, and the D119C mutant of the alpha chain was expressed together with the II C mutant of b2-hήop¾1oI>u1ΐh as a comparative example, and the G120C mutant of the alpha chain was expressed together with the H31C mutant of b2-hήop¾1oI>u1ΐh, corresponding to pair No. 1 of Cystein residues, and the G120C mutant of the alpha chain was expressed together with the II C mutant of b2-microglobulin, corresponding to pair No. 2 of Cystein residues, and the 123 C mutant of the alpha chain was expressed together with the L54C mutant of b2-hήop¾1oI>u1ΐh, corresponding to pair No. 9 of Cystein residues.

The expression vectors did not contain a coding sequence for a transmembrane domain, and no coding sequence for a cognate antigenic peptide.

For expression HEK293 cells were transfected with an expression vector coding for the alpha chain of MHC I and an additional expression vector coding for b2-hήop¾1oI>u1ΐh using a transfection reagent (Xfect™, Takara Bio Inc.). The cells were seeded in 24-well-plates, each well containing 1 ml of cell culture medium. The cells were cultivated under cell culture conditions in DMEM-F12 medium with 2% FCS at 37 °C in a 5% CO2 atmosphere. On day 3, culture supernatant was removed and protein expression was analysed by an ELISA.

Fig. 8 shows the results of the ELISA, in which culture supernatant of cells expressing MHC I molecules was immobilized on a BBM.1 or PR1F11F11 (anti-V5 tag) antibody coated surface and detection was by W6/32 or anti-His antibody. The anti-His tag antibody indicates MHC I protein concentration (right hand columns), the W6/32 antibody indicates native MHC I conformation (left hand columns). For comparison, supernatant of a HEK293 cell expressing the wild-type alpha chain without transmembrane domain the wild-type of p2-microglobulin (WT/WT) were included.

The result depicted in Fig. 8 shows that the comparative MHC I mutant, containing a Cystein in amino acid No. 119 of the alpha chain and a Cystein in amino acid No. 1 of the b2- microglobulin for HLA-A*02:01, for HLA-A*34:01, and for HLA-A*68:01 did not express an MHC I molecule. For the MHC I molecules of the invention, both expression and confirmation of the natural MHC I conformation by the conformation-specific antibody W6/32 was found.

Further comparative exemplary combinations contained the Q96C mutation in the alpha chain and the W60C mutation in the P2-microglobulin, and the Q96C mutation in the alpha chain and the H31C mutation in the p2-microglobulin. These comparative MHC I molecules were not expressed in a natural conformation as determined by the W6/32 antibody ELISA.

Example 4: Expression of a stabilized MHC containing one disulfide bond

For this example the MHC I alleles HLA-A*34:01, HLA-B* 14:02, and HLA-C*07:02 were selected for comparison of expression levels of different Cystein mutants. Cystein residues on alpha chain are A236C, G120C, H192C, and R35C. Cystein residues on p2-microglobulin are N24C, H31C, D98C, and L54C.

The A236C mutant of the alpha chain was expressed together with the N24C mutant of b2- microglobulin, corresponding to pair No. 5 of Cystein residues, and the G120C mutant of the alpha chain was expressed together with the H31C mutant of b2-ihΐop¾1o1>u1ΐh,

corresponding to pair No. 1 of Cystein residues, and the R35C mutant of the alpha chain was expressed together with the D53C mutant of b2-ihΐop¾1o1>u1ΐh, corresponding to pair No. 3 of Cystein residues. The 192C mutant of the alpha chain was expressed together with the 98C mutant of b2-ihΐop¾1o1>u1ΐh, corresponding to pair No. 12 of Cystein residues.

For expression, HEK293 cells were transfected with an expression vector coding for the alpha chain of MHC I and an additional expression vector coding for b2-ihΐop¾1o1>u1ΐh using a transfection reagent (Xfect™, Takara Bio Inc.). HLA-A*34:01, HLA-B* 14:02, and HLA- C*07:02 were selected as alleles for the alpha chain. The cells were seeded in 24-well-plates, each well containing 1 ml of cell culture medium. The cells were cultivated under cell culture conditions in DMEM-F12 medium with 2% FCS at 37 °C in a 5% CO2 atmosphere. On day 3, culture supernatant was removed and protein expression was analysed by ELISA.

Fig. 9 shows the results of the ELISA, in which culture supernatant of cells expressing MHC I molecules was immobilized on a BBM.1 or PR1F11F11 (anti-V5 tag) antibody coated surface and detection was by W6/32 indicating native MHC I conformation (left hand columns) or anti-His antibody indicating MHC I protein concentration (right hand columns). For comparison, supernatant of a HEK293 cell expressing the wild-type alpha chain without transmembrane domain the wild-type of p2-microglobulin (WT/WT) were included.

The result shows that these MHC I molecules were expressed in natural MHC I conformation as determined by W6/32 ELISA.

Certain samples of cell supernatant were further buffered into PBS (7 mM NaHiPCri, 150 mM NaCl, pH 7.4) and purified by IMAC. Purified samples were run under non-reducing conditions on an SDS-PAGE with subsequent Western blotting onto a PVDF membrane. For detection the BBM.l antibody that is specific for the p2-microglobulin, conjugated to horse radish peroxidase (HRP) for detection, was used.

The result is shown in Fig. 10, indicating that these exemplary MHC I molecules of the invention were expressed, running at similar apparent sizes in the non-reducing SDS-PAGE, the size indicating that a single disulfide bond was formed between the alpha chain and b2- microglobulin.

The examples show that the MHC I molecules of the invention, which lack a covalently bound cognate antigenic peptide and have no transmembrane domain, can be expressed in cell culture and have a native MHC I conformation.