EP4348255A1 - Peptide screen - Google Patents

Peptide screen

Info

Publication number
EP4348255A1
EP4348255A1 EP22729763.7A EP22729763A EP4348255A1 EP 4348255 A1 EP4348255 A1 EP 4348255A1 EP 22729763 A EP22729763 A EP 22729763A EP 4348255 A1 EP4348255 A1 EP 4348255A1
Authority
EP
European Patent Office
Prior art keywords
peptide
hla
interest
cell
complex
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
EP22729763.7A
Other languages
German (de)
French (fr)
Inventor
Andrew Mcmichael
Geraldine GILLESPIE
Lucy WALTERS
Max QUASTEL
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Oxford University Innovation Ltd
Original Assignee
Oxford University Innovation Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from GBGB2107821.7A external-priority patent/GB202107821D0/en
Priority claimed from GBGB2119143.2A external-priority patent/GB202119143D0/en
Application filed by Oxford University Innovation Ltd filed Critical Oxford University Innovation Ltd
Publication of EP4348255A1 publication Critical patent/EP4348255A1/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/53—Immunoassay; Biospecific binding assay; Materials therefor
    • G01N33/569—Immunoassay; Biospecific binding assay; Materials therefor for microorganisms, e.g. protozoa, bacteria, viruses
    • G01N33/56966—Animal cells
    • G01N33/56977—HLA or MHC typing
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/5005—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells
    • G01N33/5008—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics
    • G01N33/5044—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics involving specific cell types
    • G01N33/5047—Cells of the immune system
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/68—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving proteins, peptides or amino acids
    • G01N33/6878—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving proteins, peptides or amino acids in eptitope analysis
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2333/00—Assays involving biological materials from specific organisms or of a specific nature
    • G01N2333/435—Assays involving biological materials from specific organisms or of a specific nature from animals; from humans
    • G01N2333/705—Assays involving receptors, cell surface antigens or cell surface determinants
    • G01N2333/70503—Immunoglobulin superfamily, e.g. VCAMs, PECAM, LFA-3
    • G01N2333/70539—MHC-molecules, e.g. HLA-molecules

Definitions

  • the present invention relates to methods of identifying a peptide which is capable of forming an HLA-E:peptide complex and being recognised by a T-cell and or B-cell, and methods of identifying one or more T-cell or B-cell which recognises an HLA-E:peptide complex.
  • HLA-E is a non-polymorphic HLA class I molecule. There are two major alleles in the population differing only in one amino acid at position 107 which is outside the peptide binding groove (Strong et al., Correlating differential expression, peptide affinities, crystal structures, and thermal stabilities. J Biol Chem. 2003;278(7):5082-90).
  • the primary function of HLA-E is to bind a peptide usually termed ‘VL9’ which is derived from the signal peptide of classical HLA class I A, B, C molecules and HLA-G, but not HLA-E.
  • the peptide has the sequence VMAPRTLVL, VMAPRTVLL, VMAPRTLLL, VMAPRTLIL, or VMAPRTLFL.
  • the HLA-VL9 complex in turn binds to the NKG2A- CD94 inhibitory or NKG2C-CD94 activating receptors on natural killer cells and a subset of T cells (Braud et al., HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C. Nature. 1998 ;391 (6669) : 795 -9) .
  • HLA-E HLA-E
  • its functionally conserved murine and rhesus orthologues, Qa-1 and Mamu-E is to regulate NK-cell activity via MHC class I signal peptide presentation
  • specific contexts appear to permit diversified peptide binding repertoires and MHC-E-restricted CD8+ T-cell priming in vivo (Wu et al., 2018) (Vance et al., 1998).
  • HLA-E-restricted pathogen peptides include a number of physiologically relevant epitopes such as the immunodominant SIV derived supertope, RL9SIV (RMYNPTNIL)(SEQ ID NO: 1).
  • HLA-E is non-polymorphic and HLA-E restricted responses to pathogens have thus far been poorly characterised, further identification of T cell responses to peptide antigen bound to HLA-E may be useful for immunotherapies which could be applicable universally in the population due to a lack of HLA-E genetic polymorphism; the HLA-E locus encodes two non-synonymous functional allelic variants HLA-E*01:01 and HLA-E*01:03 that differ by a single Arg or Gly amino acid residue, respectively, at position 107, resulting in largely overlapping peptide binding repertoires (O’Callaghan et al., 1998) (Strong et al., 2003).
  • Minimal allelic polymorphism and shared peptide binding repertoires position HLA-E as a particularly attractive restriction element for T cell-targeted vaccination strategies which could potentially offer universal, MHC class I allotype-unrestricted protection.
  • TCRs and/or monoclonal antibodies specific for HLA-E in complex with a peptide antigen can be generated and used therapeutically as cytotoxic reagents, or such antibodies and TCRs could manipulated as receptors, including chimeric receptors, which are transfected or transduced into effector cells to induce immune responses against the peptide antigen. Therefore, the generation of antibodies or T cells and B- cells which recognise HLA-E bound to peptide antigens derived from a cancer, pathogen or even autoantigens has considerable therapeutic potential.
  • a multistep process is necessary to identify peptides that bind to HLA- E, then identify whether such HLA-E:peptide complexes can be recognised by a T-cell, such as a CD8+ T-cell, and/or a B-cell, which may induce an immune response such as a CD8+ T cell response or a B lymphocyte-mediated antibody response, in vitro or in vivo in humans or in animal models.
  • a T-cell such as a CD8+ T-cell
  • B-cell which may induce an immune response such as a CD8+ T cell response or a B lymphocyte-mediated antibody response
  • a method of identifying a peptide which is capable of forming an HLA-E:peptide complex and being recognised by a T-cell and/or B-cell comprising the steps of:
  • step (b) Stratifying the peptide of interest into one of the following groups based on its relative binding determined in step (a):
  • a peptide of interest stratified into group (ii) in step (b) may be further stratified into one of the following groups based on its relative binding determined in step (a):
  • the method may further comprise performing one or more step to improve the stability and/or level of binding of the HLA-E:peptide complex comprising the peptide of interest.
  • the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may be performed for step (c) of the method.
  • the reference peptide referred to herein may be a VL9 peptide (VMAPRT[V/L][L/V/I/F]L), such as VMAPRTLVL (SEQ ID NO: 2), VMAPRTVLL (SEQ ID NO: 3), VMAPRTLLL (SEQ ID NO: 4), VMAPRTLIL (SEQ ID NO: 5), or VMAPRTLFL (SEQ ID NO: 6).
  • VMAPRTLVL SEQ ID NO: 2
  • VMAPRTVLL SEQ ID NO: 3
  • VMAPRTLLL SEQ ID NO: 4
  • VMAPRTLIL SEQ ID NO: 5
  • VMAPRTLFL SEQ ID NO: 6
  • a HLA-E:peptide complex referred to herein refers to a complex comprising or consisting of a peptide, HLA-E heavy chain and B2 microglobulin.
  • the reference HLA- E:peptide complex referred to herein may refer to a complex comprising or consisting of a VL9 peptide, HLA-E and B2 microglobulin.
  • the HLA-E may be HLA-E*01:01 or HLA-E*01:03.
  • the HLA-E may be mutated, for example at the residues described herein.
  • the HLA-E may refer to human HLA-E or a homologue from a non-human species, such as mouse or Rhesus monkey.
  • step (a) of the method allows the level of binding of a peptide of interest to HLA-E in an HLA-E:peptide complex to be assessed, relative to that of a reference peptide such as VL9, known to bind HLA-E strongly in an HLA-E:peptide complex.
  • the peptide of interest can be given a numerical value relative to VL9 (where VL9 binding is 100%), and this allows the stratification of peptides of interest into different groups.
  • the level of binding may be defined as the percentage of HLA-E:peptide complexes formed with the peptide of interest, versus complexes formed with the reference peptide, VL9.
  • Such stratification can be used to determine whether one or more step needs to be performed to improve the stability of the HLA-E:peptide complex comprising the peptide of interest, and if so, the nature of the step or steps required for a peptide of interest stratified into a given group.
  • the inventors have shown that many immunodominant peptides which bind to HLA-E, particularly peptides derived from pathogens and cancer derived epitopes, bind surprisingly weakly in an ELISA when compared to VL9. Biological investigation of such peptides in the pursuit of therapeutics such as T-cells and antibodies targeting such pathogens and cancers, and indeed infected or malignant cells, is difficult as current techniques rely on biological functional assays which utilise stable HLA-E:peptide complexes and multimers of such complexes.
  • the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may not need to be performed.
  • a peptide of interest which binds to HLA-E in a HLA-E:peptide complex at more than about 70% of the level of binding relative to the reference peptide will be stable enough to progress to biological functional biological assays without further stabilisation.
  • One or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may still be performed for a peptide of interest stratified into group (i), if desired.
  • step (b) If in the assay of step (b) the peptide of interest is stratified into group (ii), one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may be performed.
  • the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may comprise one or more of:
  • HLA-E Introducing one or more crosslink between amino acids in HLA-E;
  • D Introducing one or more mutation in HLA-E, such as one mutation, two mutations or three mutations, to increase the stability of the binding of a peptide of interest in the HLA-E:peptide complex; or
  • HLA-E in any of (A)-(D) may refer to human HLA-E, which may be or be derived from HLA-E*0101/ E*0103
  • WSDSAQGSESHSL (SEQ ID NO: 8), which are the amino acid sequences of HLA- E*0101 and E*0103 respectively without an N-terminal signal sequence.
  • Mamu-E may optionally be mutated to enhance peptide binding, stability or reactivity with monoclonal antibodies such as the anti -HLA-E antibody 3D 12.
  • Mamu-E comprises an almost identical peptide binding groove to HLA-E. However, Mamu-E possesses structural differences outside of the peptide binding groove which result in a more stable complex, and allowing the binding of more peptides.
  • Mamu-E may be used as an alternative to HLA-E in the Single Chain Trimer assay described herein. This enables the identification of yet further peptides which may be used as epitopes with HLA-E.
  • Mamu-E may comprise one or more of the following mutations: P57S, E79R, and/or G150A.
  • the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may comprise:
  • the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may comprise one or more of:
  • the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may comprise:
  • (A) may be performed in the presence of a molar excess of the peptide of interest, such as at IOOmM or more, to ensure stability of the complex.
  • the sensitive reference peptide may be UV-sensitive.
  • the UV-sensitive reference peptide may be a UV- sensitive VL9 peptide, such as VMAP(J*)TLVL (SEQ ID NO: 9), where J* is 3-amino- 3-(2-nitrophenyl)-propionic acid.
  • (A) may be performed in the presence or absence of UV light.
  • the crosslinking one or more amino acid of the peptide of interest to one or more amino acid of HLA-E may comprise: mutating one or more residues of HLA-E to a cysteine or lysine, and optionally substituting one or more residues of the peptide of interest, to a cysteine, lysine or synthetic amino acid, such that a crosslink is capable of being formed between the peptide of interest and HLA-E. This may involve the use of an additional small molecule to bridge the amino acids to be crosslinked. This may involve the formation of a salt bridge between oppositely charged amino acids of the peptide and HLA-E.
  • the substituting one or more residues of the peptide of interest may comprise substituting the residue in the first or second position of the peptide of interest to a cysteine, a homocysteine, or a synthetic amino acid comprising a free sulphydryl group, such that a crosslink in the form of a disulphide bond can be formed between a mutant amino acid in the HLA-E heavy chain and the amino acid at the first or second position in the peptide.
  • the crosslinking one or more amino acid of the peptide of interest to one or more amino acid of HLA-E may comprise mutating the tyrosine at position 84 of HLA-E to a cysteine, adding a glycine and cysteine to the carboxy terminus of the peptide of interest, and forming a disulphide bond between the cysteine at position 84 of HLA-E and the cysteine added to the carboxy terminus of the peptide of interest.
  • the crosslinking one or more amino acid of the peptide of interest to one or more amino acid of HLA-E may comprise mutating the methionine at position 45 of HLA-E to a cysteine, substituting the amino at position two of the peptide of interest to a cysteine, a homocysteine, or a synthetic amino acid that displays a free sulphydryl group (preferably at the end of a side chain of preferred length), and forming a disulphide bond between the cysteine at position 45 of HLA-E and the cysteine, homocysteine, or a synthetic amino acid at position two of the peptide of interest.
  • the synthetic amino acid comprising a free sulphydryl group may be a homocysteine analogue, (2S)-2-amino-5-sulfanylpentanoic acid or (2S)-2-amino-6 sulfanylhexanoic acid.
  • the one or more crosslink between amino acids in HLA-E may be introduced by mutating the tyrosine at position 84 of HLA-E to a cysteine, mutating the alanine at position 139 of HLA-E to a cysteine, and forming a disulphide bond between the two cysteine residues at position 84 and 139 of HLA-E.
  • This crosslink may improve the binding of a peptide of interest in the HLA-E:peptide complex, demonstrated by increased Tm ( Figure 8).
  • the one or more mutation in HLA-E may comprise or consist of one or more of, such as one of, two of, or all of, mutating Histidine at position 99 to Tyrosine, mutating Phenylalanine at position 116 to Tyrosine or mutating Serine at position 147 to Tryptophan.
  • These mutations close/alter different HLA-E binding pockets which are used optimally by the signal peptide VL9. This is in contrast to many low affinity pathogen or cancer-derived peptides that bind to HLA-E.
  • These mutations enhance the binding of the pathogen or cancer-derived peptides, demonstrated by increased melting temperature (Tm) of the protein ( Figures 9,11 and 12).
  • the one or more mutation in HLA-E may comprise swapping the alpha-3 domain of HLA-A3 with that of HLA-E.
  • the invention provides a mutant HLA-E heavy chain comprising one or more mutation which permits the formation of a HLA-E:peptide complex with increased stability when compared to the complex without the mutant HLA-E heavy chain.
  • the complex may further comprise b2 microglobulin.
  • the mutant HLA-E may comprise a mutation at Histidine at position 99 to Tyrosine.
  • the mutant HLA-E may comprise a mutation at Phenylalanine at position 116 to Tyrosine.
  • the mutant HLA-E may comprise a mutation at Serine at position 147 to Tryptophan.
  • the mutant HLA-E may comprise a mutation at Histidine at position 99 to Tyrosine and a mutation at Phenylalanine at position 116 to Tyrosine.
  • the mutant HLA-E may comprise a mutation at Histidine at position 99 to Tyrosine and a mutation at Serine at position 147 to Tryptophan.
  • the mutant HLA-E may comprise a mutation at Histidine at position 99 to Tyrosine, a mutation at phenylalanine at position 116 to Tyrosine, and a mutation at Serine at position 147 to Tryptophan.
  • the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest referred to herein may also be utilised in any assay referred to herein.
  • the peptide of interest may be identified using one or more amino acid sequence prediction process.
  • a peptide of interest may be identified as comprising a HLA-E restricted sequence motif (Walters et al., 2020 EJI).
  • a peptide of interest may be identified using one or more publicly available programme such as NetMHC, or using an algorithm.
  • a peptide of interest may be identified by using one or more of the following search parameters: peptide is a nonamer; peptide comprises one or more proline between positions 3 and 7, peptide comprises a restricted set of amino acids at positions 2 and 9, such as an M, L, V A, Q, or F at position 2, and/or a L, F, I, V or M at position 9.
  • the method may further comprise, before step (a), performing an initial screening assay to determine whether a HLA-E:peptide complex comprising the peptide of interest is capable of being expressed on a cell surface, wherein if the HLA-E:peptide complex comprising the peptide of interest is determined to be capable of being expressed on a cell surface, the method proceeds to step (a).
  • the initial assay may comprise transfecting into a cell, such as HEK293T cells, a nucleic acid encoding a polypeptide comprising the peptide of interest, B2microglobulin and HLA-E heavy chain, in that order.
  • a polypeptide is also referred to herein as a
  • the nucleic acid may also encode a linker sequence between each of the peptide of interest, B2microglobulin and HLA-E heavy chain, preferably in that order.
  • the initial assay may comprise performing an experimental technique to detect the presence of the peptide of interest bound to HLA-E at the surface of the cell which expresses the SCT.
  • Many experimental techniques are known to the skilled person in the art, such as flow cytometry, and microscopy such as confocal microscopy.
  • the experimental technique may utilise a conformation-specific antibody, which recognises a correctly folded HLA-E:peptide complex, for example which can be expressed at the cell surface. Such an antibody may recognise a domain in the a3 region of HLA-E.
  • Such an initial screening assay allows the skilled person to quickly, cheaply and reliably assess whether a peptide of interest can be presented on the cell surface by HLA-E. If a peptide of interest is not expressed at the cell surface bound to HLA-E, then the method may terminate at this stage for that peptide of interest. If a peptide of interest is expressed at the cell surface bound to HLA-E, the method may proceed.
  • the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest referred to herein may also be utilised in the initial screening assay.
  • the assay of step (a) may be an Enzyme-linked immunosorbent assay (ELISA).
  • ELISA Enzyme-linked immunosorbent assay
  • the ELISA may be a sandwich ELISA.
  • the sandwich ELISA assay may utilise an anti HLA-E antibody and an anti B2 microglobulin antibody.
  • the ELISA may be used to semi-quantitatively determine the level of binding of a peptide of interest in a HLA-E:peptide complex, relative to the level of binding of a reference peptide to HLA-E in a reference HLA-E complex.
  • the level of binding of a reference peptide to HLA-E in a reference HLA-E complex may be taken as 100%.
  • the reference peptide may be VL9, described above.
  • the VL9 may be VMAPRTLLL.
  • the HLA-E:peptide complex comprising the peptide of interest used in the ELISA may be formed using a peptide exchange method, in which a reference HLA-E:peptide complex comprising a UV-sensitive reference peptide is first formed, and the reference peptide is subsequently exchanged with the peptide of interest under conditions such that an HLA-E:peptide complex comprising the peptide of interest is formed.
  • the HLA-E heavy chain may be expressed in in E.coli, before being refolded by mixing with B2microglobulin and a UV-labile VL9 peptide, such as VMAP(J*)TLVL where J is 3-amino-3-(2-nitrophenyl)-propionic acid, a synthetic amino acid that is sensitive to UV light.
  • VMAP(J*)TLVL where J is 3-amino-3-(2-nitrophenyl)-propionic acid
  • This protein complex may then be purified, for example by Fast Protein Liquid Chromatography (FPLC) size exclusion, and the protein complex of the expected mass may be collected and analysed.
  • the complex which is either fresh or previously frozen and thawed, may then be mixed with a 100-fold molar excess of the peptide, and optionally exposed to UV light.
  • the ELISA may use the methodology described in Walters et al., EJI, 2020.
  • the assay of step (a) may comprise or consist of performing Thermal melt (Tm) analysis by Differential Scanning Fluorometry (DSF), or Small Angle X-ray Scattering of the material after flowing through the size exclusion column (SEC-SAXS), which purifies the refolded HLA-E after the peptide exchange.
  • Tm Thermal melt
  • SEC-SAXS Small Angle X-ray Scattering of the material after flowing through the size exclusion column
  • Such steps may be performed on fresh or previously frozen material that has been thawed.
  • the inventors have determined that the level of binding of a peptide of interest relative to VL9 binding to HLA-E in the ELISA assay has a strong correlation with the thermal stability of the HLA-E: peptide complex and its overall structural conformation in solution ( Figure 4).
  • binding in the ELISA assay and/or thermal melt determination is predictive of structural stability.
  • the method may further comprise performing one or more further assay after step (b) to determine the biophysical and/or biochemical characteristics of the HLA-E:peptide complex comprising the peptide of interest.
  • the one or more further assays may comprise Native PAGE, Thermal melt (Tm) analysis, and/or) Size Exclusion Chromatography- Small Angle X-ray Scatter (SEC-SAXS) .
  • SAXS small angle X ray scatter
  • one or more biological assay may be performed.
  • the one or more biological assay may use of multimers of the HLA-E:peptide complex comprising the peptide of interest.
  • the multimer may be a tetramer or a pentamer.
  • a multimer described herein may comprise streptavidin or a microbead and may comprise a label such as a fluorochrome.
  • the one or more biological assay may be a FACS assay.
  • the one or more biological assay may comprise introducing the peptide of interest into a co-culture of T-cells and monocytes, in the presence of IL-2 and IL-15.
  • the peptide of interest may then bind to HLA-E on the monocytes and be recognised by one or more T-cell.
  • T-cell and/or B-cell (or the corresponding TCR/BCR) identified in such a biological assay may then be selected, isolated, sequenced and/or cloned.
  • the selected/isolated T-cell and/or B-cell may be expanded.
  • one or more identified T cell may be sorted using a multimer, and seeded at less than one cell per well into multiple wells of a tissue culture plate, and cultured with irradiated allogeneic peripheral blood mononuclear cells (PBMC) as feeders together with IL-2 and phytohemagglutinin (PHA).
  • PBMC peripheral blood mononuclear cells
  • PHA phytohemagglutinin
  • step (c) of the assay it may be also determined whether an HLA-E:peptide complex comprising the peptide of interest is capable of stimulating an immune response in one or more T-cell and/or B-cell.
  • one or more biological assay may be performed.
  • the one or more biological assay may identify whether an HLA-E:peptide complex comprising the peptide of interest is capable of inducing an increase in IFN-g, TNF-a, CD107a/b and/or CD137 expression in T-cells.
  • numerous other activation markers can be used to determine whether an immune response is stimulated in a T-cell.
  • the one or more biological assay may comprise the use of multimers of the HLA- E:peptide complex comprising the peptide of interest to determine whether an immune response in a T-cell and/or B-cell can be stimulated.
  • the multimer may be a tetramer or a pentamer or a dextramer.
  • a multimer described herein may comprise streptavidin or a microbead and may comprise a label such as a fluorochrome.
  • the one or more biological assay may comprise introducing the peptide of interest into a co-culture of T-cells and monocytes, in the presence of IL-2 and IL-15.
  • the peptide of interest may then bind to HLA-E on the monocytes and stimulate an immune response in one or more T-cell.
  • One more T-cell identified as being stimulated in such biological assays may then be selected, isolated, sequenced and/or cloned.
  • the selected/isolated T-cell may be expanded.
  • one or more identified T cell may be sorted using a multimer, and seeded at less than one cell per well into multiple wells of a tissue culture plate, and cultured with irradiated allogeneic peripheral blood mononuclear cells (PBMC) as feeders together with IL-2 and phytohemagglutinin (PHA).
  • PBMC peripheral blood mononuclear cells
  • PHA phytohemagglutinin
  • one or more biological assay may be performed.
  • the one or more biological assay may identify whether an HLA-E:peptide complex comprising the peptide of interest is capable of being bound by a BCR, and/or whether this stimulates the secretion of antibody.
  • the HLA-E:peptide complex comprising the peptide of interest, or a multimer thereof, is capable of stimulating an immune response in one or more T- cell and or B-cell, this is indicative of the T-cell and/or B-cell recognising the HLA- E:peptide complex comprising the peptide of interest.
  • a method of identifying one or more T-cell or B- cell which recognises an HLA-E:peptide complex comprising a peptide of interest comprising:
  • An immune response may be stimulated in a T-cell and/or B-cell upon recognition of the HLA-E:peptide complex comprising the peptide of interest.
  • the assay to identify a T-cell or B-cell which recognises the HLA-E:peptide complex comprising the peptide of interest may comprise isolating T-cells and/or B-cells from a sample of blood or tissue/biological fluid from the site of infection or a tumour, of a healthy subject or from a subject who has been diagnosed with or is suspected of having cancer and/or an infection.
  • One or more T-cell or B-cell which recognise the HLA- E:peptide complex comprising the peptide of interest may then be selected.
  • the subject may be a mammal.
  • the subject may be a human.
  • the method may further comprise performing an assay to determine whether a T-cell or B-cell which is identified by any of the methods described above, is capable of recognising and/or being activated by an abnormal cell.
  • An abnormal cell may be a cell which is infected with a pathogen, a tumour cell, or a cell which is undergoing a stress response, and which comprises the peptide of interest.
  • the stress response may be to one or more of: metabolic stress, physical stress, apoptotic stress, pH stress, osmotic stress, DNA Damage stress, heat stress, hypoxic stress or toxin stress.
  • an identified T-cell or T-cell clone may be exposed to an abnormal cell and an assay performed to determine whether the T-cell is activated.
  • Presence or level of activation markers such as one or more of CD137, CD107a/b, TNF alpha, or interferon-gamma may be used to determine whether the T-cell is activated by contact with the abnormal cell.
  • This step confirms that the peptide of interest is a biologically relevant epitope which is presented on an abnormal cell by HLA-E.
  • This step also confirms that a multimer used for the T cell or B-cell selection displays the same antigen/epitope as is present on the abnormal cell; that is, where one or more step to improve the stability of the HLA- E:peptide complex comprising the peptide of interest has been performed, a resulting T-cell or B-cell identified is also capable of recognising and/or being stimulated by a corresponding natural HLA-E:peptide complex, in which no step to improve the stability has been performed.
  • This step therefore excludes T cells and B-cells that recognise and/or respond to altered forms of the HLA-E (or indeed Mamu-E) multimers which result from on chemical, biochemical or mutational steps used in the one or more step to improve the stability described herein.
  • T cells or B cells, or indeed their TCR or BCR respectively may then be developed as a therapeutic reagent.
  • a resulting TCR and/or BCR may be sequenced and cloned.
  • the nucleic acid encoding a TCR can then be used to generate soluble TCR or can be transfected or transduced into live cells such that they then express the receptor.
  • nucleic acid such as DNA, cDNA or RNA
  • TCRs or antibodies See for example Robbins et al., (2008) J Immunol. 180: 6116-6131.
  • Any method described herein may further comprise generating one or more T-cell or TCR, or antibody such as a monoclonal antibody, which recognises the HLA-E:peptide complex comprising the peptide of interest.
  • the skilled person will know of various methods to generate such an antibody, including immunising an animal such as a mouse such that antibodies which recognise the HLA-E:peptide complex comprising the peptide of interest are generated, or by using the Milstein method.
  • Figure 1 - shows Blue native (BN) gels signatures for pathogen peptide versus canonical leader peptide-loaded HLA-E complexes.
  • 10pg of pre-refolded HLA-E- 2m-UV-sensitive VL9 (VMAP(J*)TLVL) material was incubated with 12M excess of test peptides (VL9, Mtb44, IL9, RL9H, RL9S and BLZF1) or the no added peptide buffer control (Eno pep) for 3 hours on ice prior to Blue Native-PAGETM Novex 4-16% Bis-Tris gel evaluation.
  • Figure 2 - shows size exclusion chromatography-coupled small angle x-ray scattering of peptide-HLA-E complexes A.
  • Ab initio molecular envelope models displayed as small ‘dots’ represent the average conformational protein state adopted in solution and were generated from SEC-SAXS data using the DAMMIF and DAMAVER ATSAS packages in conjunction with ScAtter (Franke & Svergun, 2009) (Volkov & Svergun, 2003).
  • the top row of molecular envelope ‘dot’ models correspond to SEC- SAXS runs where HLA-E complexes were injected onto the HPLC column in the presence of 120mM excess peptide but no excess peptide was added to the HPLC elution buffer.
  • the bottom row correspond to SEC-SAXS runs where 120mM excess peptide was present in both the HPLC injection and elution buffers to ensure continual presence of excess ligand.
  • the purple-shaded molecular envelope ‘dot’ models represent canonical VL9 (VMAPRTVLL)-refolded HLA-E.
  • VMAPRTVLL canonical VL9
  • the structural coordinates of HLA-E in complex with VL9 (VMAPRTVLL) from 1MHE are shaded green and were superimposed onto SEC-SAXS-based molecular envelopes using the SUPCOMB package of ATSAS (O’Callaghan et al., 1998).
  • the neighbouring SEC- SAXS molecular envelope ‘dot’ models represent HLA-E refolded with Mtb44 (RLPAKAPLL) (SEQ ID NO: 10), onto which the previously published HLA-E- Mtb44 structural coordinates (shaded purple) were aligned (Walters et al., 2018).
  • the blue-shaded SEC-SAXS molecular envelope ‘dot’ models represents IL9 (IMYNYPAML)(SEQ ID NO: l l)-refolded HLA-E onto which the HLA-E-IL9 structural coordinates (shaded orange) were aligned.
  • the magenta-shaded SEC-SAXS molecular envelope ‘dot’ models correspond to Mtbl4 (RMAATAQVL)(SEQ ID NO: 12)-refolded HLA-E, with the HLA-E-Mtbl4 structural coordinates superimposed via SUPCOMB.
  • the neighbouring SEC-SAXS molecular envelope ‘dot’ models shaded yellow represent HLA-E refolded with RL9HIV (RMYSPTSIL) (SEQ ID NO: 13), onto which the previously published HLA-E-RL9HIV structural coordinates (shaded pink) were aligned (Walters et al., 2018).
  • Each row represents a SEC-SAXS experiment conducted in the absence or presence of 120mM excess peptide in the HPLC elution buffer.
  • the radius of gyration (Rg) and maximum dimension (dmax), both measured in A, are specified.
  • the volumes of SEC-SAXS molecular envelope ‘dot’ models (A3) displayed above in part A, are also denoted along with the % change in molecular envelope volume following the addition of excess peptide to the HPLC elution buffer.
  • C. (i), (iii), (v), (vii), (ix) & (xi) Log 10 scattering intensity plots for HLA-E SEC-SAXS experiments.
  • Figure 3 - shows size exclusion chromatography-coupled small angle x-ray scattering for peptide-HLA-A2 complexes.
  • A. (i) & (iii) Log 10 scattering intensity plots for HLA-A*02:01 SEC-SAXS experiments. Plotted on the X-axis is the scattering vector, q, measured in A-l, which for small angles is proportional to the scattering angle Q.
  • the scattered intensity, I(q) is plotted on the Y-axis with a log scale. Scattering intensity curves for HLA-A2 refolds run in the absence or presence of 120mM excess peptide in the HPLC elution buffer, are plotted together for reference.
  • Figure 4 - demonstrates that molecular envelope elongation negatively correlates with HLA-E thermal stability and peptide binding signals
  • a positive control VL9 leader peptide (VMAPRTVLL) in addition to a non binding negative control peptide derived from HIV Gag (QAISPRTLN) (SEQ ID NO: 14) are included.
  • the ‘ELISA rank’ column reflects normalised and previously published peptide binding signals which are expressed as percentages of the positive control VL9 signal (Walters et al. 2020).
  • HLA-E- 2M-peptide complex thermal stability is indicated by its melting temperature (°C) in the presence of 12M excess peptide.
  • the SEC-SAXS-obtained dmax value, measured in A, reflects the maximum dimension across the HLA-E complex in solution whereas the ‘Molecular Envelope VoT corresponds to the volume in A3 of the DAMMINF ab initio molecular envelope models presented in Figure 2.
  • the ‘no rescue’ negative control shaded in dark grey reflects a peptide exchange reaction conducted in the absence of excess test peptide.
  • VMAPRTVLL positive control VL9 leader peptide
  • the positive control VL9 leader peptide (VMAPRTVLL) corresponds to a peptide exchange reaction of 120mM excess peptide.
  • Figure 5 - shows structural characterisation of Mycobacterial peptide binding to HLA-E A.
  • Peptide residues are labelled and an electron density map contoured to 1 sigma is displayed in grey mesh overlaying the peptide (ii) Ribbon representation of the Mtbl4 (RMAATAQVL) peptide backbone shaded in hot pink from the HLA-E-Mtbl4 structure determined in this study with a superimposed VL9 (VMAPRTVLL) leader peptide backbone from the previously published HLA-E-VL9 structure, 1MHE, in violet.
  • VMAPRTVLL superimposed VL9 leader peptide backbone from the previously published HLA-E-VL9 structure, 1MHE, in violet.
  • the HLA-E binding groove is omitted and the peptide N and C termini labelled for clarity.
  • the distance (1.9 A) separating the position 7 Ca atom of the IL9 peptides from molecules 1 and 2 of the asymmetric unit is denoted (iv) Superimposed peptide backbones are depicted in ribbon-form with the HLA-E peptide binding groove omitted and N and C termini labelled for clarity.
  • the IL9 (IMYNYPAML) peptide backbones from molecules 1 and 2 of the asymmetric unit are displayed in blue.
  • the VL9 (VMAPRTVLL) leader peptide backbone from the previously published HLA-E-VL9 structure, 1MHE, is also displayed and shaded violet.
  • the associated Hepatitis C virus-derived nonamer (CINGVCWTV)(SEQ ID NO: 15) is shown in pink ribbon with the solvent-exposed position 7 side chain in stick-form projecting away from the E pocket. Visible sections of the HLA-A2 peptide binding groove including the a2 helix and b-sheet floor are shown in grey cartoon (ii) - (v) PyMol visualisation of the peptide position 7 anchor side chain-accommodating E pocket of HLA-E. Side chains of secondary E pocket-forming residues are depicted as grey sticks (Phe-116, Trp-133, Ser-147, Glu-152, Gln-156) with remaining visible regions of the peptide binding groove shown in grey cartoon.
  • VL9 (VMAPRTVLL) peptide mainchain is displayed as purple ribbon with the position 7 side chain in purple stick-form projecting downward into the secondary E pocket (PDB ID: 1MHE, O’Callaghan et al. 1998).
  • PDB ID: 1MHE, O’Callaghan et al. 1998 illustrate differential positioning at position 7 of HLA-E-bound pathogen-derived peptides relative to VL9 from 1MHE.
  • Mtb-derived IL9 IYNYPAML
  • HIV Gag- derived RL9HIV RYSPTSIL
  • Mtb-derived Mtbl4 RAATAQVL
  • peptide backbones are depicted as blue (iii), yellow (iv) and magenta (v) ribbons, respectively.
  • the position 7 side chains - Ala-7 of IL9, Ser-7 of RL9HIV and Gln-7 of Mtbl4 - are shown in blue (iii), yellow (iv) and magenta (v) stick-form, respectively.
  • VL9 peptide main chain from 1MHE is shown in purple ribbon and the distance between the aligned peptide position 7 Ca atoms is indicated by dashed lines - 2 A separates the position 7 Ca atoms of VL9 and IL9, 3.4 A separates the position 7 Ca atoms of VL9 and RL9HIV and 1.1 A separates the Ca atoms of VL9 and Mtbl4.
  • water-mediated intra- inter-chain hydrogen bonds are depicted as magenta dashed lines with the coordinated H20 molecule visualised as a cyan-shaded sphere.
  • a total of 4 hydrogen bonds are shown which indirectly link the Mtbl4 position 7 Gin side chain to the Ser-147 side chain and Ser-143 main chain of the HLA-E a2 -helix in addition to the position 8 main chain of the Mtbl4 peptide.
  • the Val-8 main chain of the Mtbl4 peptide and Ser-143 main chain of HLA-E are labelled ‘MC’ and depicted as magenta and grey sticks, respectively.
  • Tm values are listed for the canonical HLA-B7-derived VL9 leader peptide (VMAPRTVLL) and the Mtb-derived Mtbl4 peptide (RMAATAQVL) in addition to corresponding position 7 variant peptides in which a non-canonical polar Gin was introduced at position 7 in VL9 (VMAPRTQVL) (SEQ ID NO: 16) or a canonical hydrophobic Val was introduced at position 7 in Mtbl4 (RMAATAVVL) (SEQ ID NO: 17).
  • VMAPRTQVL VL9
  • RMAATAQVL Mtb-derived Mtbl4 peptide
  • VMAPRTVLL wild-type VL9 leader peptide
  • the Mtb-derived Mtbl4 peptide (RMAATAQVL) which also contains a non-canonical position 7 Gin was included for reference.
  • D. (i) & (iii) Log 10 scattering intensity plots for HLA-E SEC- SAXS experiments. Plotted on the X-axis is the scattering vector, q, measured in ⁇ -1, which for small angles is proportional to the scattering angle Q. The scattered intensity, I(q), is plotted on the Y-axis with a log scale.
  • Superimposed scattering intensity curves for peptide-HLA-E refolds are colour-coded according to the corresponding figure legend (ii) & (iv) Normalised Kratky plots with superimposed curves from HLA-E SEC-SAXS experiments.
  • Superimposed modulated Gaussian curves are colour-coded according to figure legends in log 10 intensity plots. Plotted on the X-axis of the normalised Kratky plot is the scattering vector multiplied by the radius of gyration. On the Y-axis, the scattering intensity, I(q), is divided by the experiment’s 1(0) and multiplied by (q*Rg)2.
  • Figure 7 - demonstrates that distinct that structural motifs emerge in the absence of HLA-E-associated VL9 leader peptide A.
  • SA short-arm of the a2 helix
  • LA long-arm labelled ‘LA’
  • residue positions denoted The a2 helix from the VL9 (VMAPRTVLL)-associated HLA-E structure, 1MHE, is shaded grey.
  • HLA-E-Mtb44 HLA-E-RL9HIV
  • RYSPTSIL HLA-E-RL9HIV
  • RAATAQVL HLA-E-Mtbl4
  • IMYNYPAML HLA-E-IL9
  • HLA- E-associated peptides Ca backbones of superimposed HLA- E-associated peptides are depicted as ribbons.
  • the canonical VL9 leader peptide (VMAPRTVLL) peptide from 1MHE is shaded grey.
  • Pathogen-derived HLA-E- associated peptides are coloured green (Mtb44 - RLPAKAPLL), yellow (RL9HIV - RMYSPTSIL), magenta (Mtbl4 - RMAATAQVL) and blue (IL9 - IMYNYPAML).
  • Mtb44 - RLPAKAPLL yellow
  • RL9HIV - RMYSPTSIL yellow
  • magenta Mtbl4 - RMAATAQVL
  • blue IL9 - IMYNYPAML
  • Position 5 Ca atoms are circled with the maximum distance separating the VL9 peptide Arg-5 Ca from pathogen-derived peptide position 5 Ca atoms, denoted. 2.3 A separates the position 5 Ca atoms of VL9 and RL9HIV, 1.9 A separates the position 5 Ca atoms of VL9 and IL9, 1.9 A separates the Ca atoms of V9 and Mtbl4 and 1.5 A separates the position 5 Ca atoms of VL9 and Mtb44. C.
  • the HLA-E a2 helix is shown in grey cartoon with superimposed pathogen-derived peptide backbones, RL9HIV (RMYSPTSIL), IL9 (IMYNYPAML) and Mtbl4 (RMAATAQVL) in yellow, blue and magenta ribbon, respectively.
  • the superimposed VL9 leader peptide (VMAPRTVLL) peptide backbone from 1MHE is shown in grey ribbon.
  • Ser-147, Glu- 152 and Gln-156 side chains of the HLA-E a2 helix are shown in stick-form and are colour-coded according the corresponding peptide.
  • the 2 salt bridges connecting Glu- 152 and the Arg-5 side chain of the VL9 peptide are shown as grey dashed lines.
  • a hydrogen bond between the Gln-156 side chain and the main chain Oxygen of VL9 Arg- 5 is also depicted as grey dashed lines (ii)
  • the HLA-E a2 helix is shown in grey cartoon with the Mtb-derived peptide Mtb44 (RLPAKAPLL) peptide backbone (green) superimposed to that of VL9 (VMAPRTVLL) from 1MHE (grey).
  • the HLA-E a2 helix is shown in grey cartoon with the superimposed pathogen-derived peptides, RL9HIV (RMYSPTSIL) and IL9 (IMYNYPAML), in yellow and blue ribbon, respectively.
  • the Tyr-3 side chain of RL9HIV and IL9 is shown in yellow and blue stick-form, respectively.
  • HLA-E a2 helix Glu-152 side chains from the HLA-E-RL9HIV and HLA-E-IL9 structures are shown as yellow and blue sticks, respectively, with corresponding hydrogen bonds depicted as yellow/blue dashed lines.
  • HLA-E-VL9 VMAPRTVLL
  • VMAPRTVLL Glu-152 side chain from the a2 helix of HLA-E-VL9
  • VMAPRTVLL VMAPRTVLL
  • E Table detailing 7 previously published non-receptor-bound HLA-E structures in addition to the 2 novel HLA-E structures presented in this study.
  • HLA-E-associated peptide IDs, organisms of origin and amino acid sequences are specified along with the corresponding HLA-E allelic variant, PDB accession code (AC) and reference.
  • Figure 8 Demonstrates thermal gain of HLA-EC84-C139 over canonical HLA-E when incubated with 100M excess peptide.
  • IOUM of pre-refolded HLA-E and HLA- EC84-C139 material was incubated with 100M excess test peptides (P1-P9) for 30 minutes at room temperature prior to thermal melt analysis using a Prometheus NT.48 Series Differential Scanning Fluorimetry instrument. Test samples were split between two Prometheus NT.48 Series nanoDSF Grade Standard Capillaries and a ramp rate of 1 °C/min from 20 °C to 95 °C was applied.
  • the ratio for fluorescence emission at 330 nm and 350 nm was used to derive the thermal melt of unfolding (Tm). Shown are the relative Tm data for canonical HLA-E (left column) and HLA-EC84-C139 (right column) datasets, where the corresponding no-peptide control Tm data for canonical HLA-E and HLA-EC84-C139 have been subtracted, respectively. The numbers plotted above the red bars denote the equivalent Tm loss/gains obtained for the HLA-EC84- C139 variant over canonical HLA-E.
  • Figure 9 Demonstrates thermal gain of HLA-EW147over canonical HLA-E when incubated with 100M excess peptide.
  • IOuM of pre-refolded HLA-E or HLA- EW147material was incubated with 100M excess test peptides (P1-P9) for 30 minutes at room temperature prior to thermal melt analysis using a Prometheus NT.48 Series Differential Scanning Fluorimetry instrument. Test samples were split between two Prometheus NT.48 Series nanoDSF Grade Standard Capillaries and a ramp rate of 1 °C/min from 20 °C to 95 °C was applied.
  • the ratio for fluorescence emission at 330 nm and 350 nm was used to derive the thermal melt of unfolding (Tm). Shown are the relative control Tm gains for canonical HLA-E (left column) and HLA-EW147 (right column) datasets, where the corresponding no-peptide control Tm data for canonical HLA-E and HLA-EW147 have been subtracted, respectively. The numbers plotted above the red bars denote the equivalent Tm loss/gains obtained for the HLA- EW147variant over canonical HLA-E.
  • Figure 10 - demonstrates that pathogen-derived epitopes drive suboptimal HLA-E complex formation, despite previously reported immunogenicity in vivo A.
  • Rg radius of gyration
  • the x-axis denotes HPLC-eluted x-ray exposed frames across the HLA-E-VL9 protein peak, whereas the Y-axis corresponds to the SAXS scattering intensity signal (ii)
  • Ab initio molecular envelope model displayed as small purple ‘dots’ representing the average conformational state of the leading peak fraction of HPLC-eluted HLA-E refolded with the canonical VL9 leader peptide.
  • This molecular envelope ‘dot’ model corresponds to a SEC-SAXS run where an HLA-E*01:03 refold was injected onto the HPLC column in the presence of 120mM excess VL9 peptide but no excess peptide was added to the HPLC elution buffer.
  • This molecular envelope ‘dot’ model corresponds to a SEC-SAXS run where an HLA-E*01:03 refold was injected onto the HPLC column in the presence of 120mM excess IL9 peptide but no excess peptide was added to the HPLC elution buffer.
  • the structural coordinates for HLA-E-IL9 are aligned to the SAXS-generated molecular envelope and displayed in orange cartoon-form (iii) Same as B. (ii), but for tailing peak fraction (iv) Log 10 scattering intensity plot for HLA-E SEC-SAXS. Plotted on the X-axis is the scattering vector, q, measured in A-l, which for small angles is proportional to the scattering angle Q.
  • the scattered intensity, I(q), is plotted on the Y-axis with a log scale.
  • Scattering intensity curves for HLA-E refolds run in the absence of 120mM excess peptide in the HPLC elution buffer are plotted for the leading and tailing peak fractions of IL9-refolded HLA-E and are colour- coded according to the corresponding figure legend (v) Normalised Kratky plot with superimposed curves corresponding to SEC-SAXS leading and tailing peak fractions for IL9-refolded HLA-E in the absence of 120mM excess peptide in the HPLC elution buffer, colour-coded according to the figure legend in B. (iv).
  • This molecular envelope ‘dot’ model corresponds to a SEC-SAXS run where an HLA-E*01:03 refold was injected onto the HPLC column in the presence of 120mM excess RL9HIV peptide but no excess peptide was added to the HPLC elution buffer.
  • the previously published structural coordinates for HLA-E-RL9HIV (PDB AC: 6GL1) are aligned to the SAXS-generated molecular envelope and displayed in pink cartoon-form (iii) Same as C. (ii), but for tailing peak fraction (iv) LoglO scattering intensity plot for HLA-E SEC-SAXS.
  • the scattered intensity, I(q), is plotted on the Y- axis with a log scale.
  • Scattering intensity curves for HLA-E refolds run in the absence of 120mM excess peptide in the HPLC elution buffer are plotted for the leading and tailing peak fractions of RL9HIV-refolded HLA-E and are colour-coded according to the corresponding figure legend (v) Normalised Kratky plot with superimposed curves corresponding to SEC-SAXS leading and tailing peak fractions for RL9HIV-refolded HLA-E in the absence of 120mM excess peptide in the HPLC elution buffer, colour- coded according to the figure legend in C. (iv). Plotted on the X-axis is the scattering vector multiplied by the radius of gyration.
  • FIG. 11 Thermal gain of HLA-E H99Y , HLA-E f116Y and HLA-E S147W over canonical HLA-E when incubated with 10M excess peptide lOuM of pre-refolded HLA-E and HLA-E H99Y (A), HLA-E F116Y (B), or HLA-E S147W (C) material was incubated with 10M excess test peptides (from panel pA to pG) for 30 minutes at room temperature prior to thermal melt analysis using a Prometheus NT.48 Series Differential Scanning Fluorimetry instrument.
  • Test samples were split between two Prometheus NT.48 Series nanoDSF Grade Standard Capillaries and a ramp rate of 1 °C/min from 20 °C to 95 °C was applied. The ratio for fluorescence emission at 330 nm and 350 nm was used to derive the thermal melt of unfolding (Tm). Shown are the relative control Tm gains for canonical HLA-E (left columns) and HLA-E H99Y , HLA- E f116Y or HLA-E S147W (right columns) datasets, where the corresponding no-peptide control Tm data for canonical HLA-E and HLA-E H99Y , HLA-E f116Y or HLA-ES147W have been subtracted, respectively. The numbers plotted above the right-hand columns denote the equivalent Tm gains obtained for the HLA-E H99Y , HLA-E f116Y orHLA-E sl47W variants over canonical HLA-E, respectively.
  • Figure 12 Thermal stability and Blue Native gel analysis of canonical HLA-E and versus single, double and triple HLA-E mutants refolded with RL9HIV peptide.
  • Example 1 Distinct blue native gel signatures for VL9- versus pathogen epitope- associated HLA-E
  • Example 2 SEC-SAXS reveals striking peptide-dependent differences in the HLA- E conformational ensemble Given that multiple distinct gel band signatures were identified in BNG analyses indicative of sample non-uniformity, size exclusion chromatography-coupled small angle x-ray scattering (SEC-SAXS) was used to further probe potential discrepancies in the protein conformational ensemble for canonical VL9 versus pathogen peptide- associated HLA-E. Individual HLA-E*01:03 protein refolds were assembled for each HLA-E-restricted pathogen-derived peptide present in the aforementioned ‘intermediate’ binding panel in addition to the positive control VL9 peptide, VMAPRTVLL.
  • SEC-SAXS size exclusion chromatography-coupled small angle x-ray scattering
  • SAXS The resolution limits of SAXS are inferior relative to those of x-ray crystallography making it challenging to decipher local, fine-tuned structural movement with SAXS data alone.
  • SAXS is capable of detecting dynamic protein folded states and large conformational adjustments of proteins in solution (Kikhney and Svergun, 2015).
  • HLA-E-restricted epitopes Hansen et al., 2016; Walters et al., EJI, 2020
  • SAXS immunodominant SIV epitope
  • EKQRESREK immunodominant SIV epitope
  • Tm DSF-generated melting temperatures
  • the negative control peptide (HIV Gag, QAISPRTLN) generated a comparable Tm (31.8°C) to the peptide-free-HLA-E background (32.0°C) whereas Mtb44 (RLPAKAPLL) generated a comparable Tm (50.6°C) to the positive control VL9-incubated HLA-E (49.4°C).
  • the remaining ‘intermediate’ binding peptides generated Tm values ranging from 35.2°C (RL9SIV, RMYNPTNIL) to 40.7°C (IL9, IMYNYPAML).
  • ELISA signals for the positive control VL9 (VMAPRTVLL) and Mtb44 (RLPAKAPLL) peptides remained significantly higher than those generated by the ‘intermediate’ pathogen-derived peptides IL9 (IMYNYPAML), Mtbl4 (RMAATAQVL), RL9HIV (RMYSPTSIL), BZLF1 (SQAPLPCVL) and RL9SIV (RMYNPTNIL), even when such ‘intermediate’ binding pathogen peptides were supplied at 1.2mM concentrations.
  • Such observations not only demonstrate the tiny selectivity of HLA-E for its canonical VL9 leader peptide but also reveal the suboptimal nature of diverse pathogen-derived epitope sampling by HLA-E.
  • Tm values and normalised ELISA signals exhibited strong negative correlations with measuremets obtained via SEC-SAXS analyses including the maximal linear dimension (as indicated by dmax values measured in A) and molecular envelope volume (as indicated by ab initio model volume measured in A3) of HLA-E complexes in solution ( Figure 4, B, i-iv).
  • Tm and ELISA signals increase, the average conformation of the protein ensemble in solution contracts and more closely aligns to previously obtained HLA-E crystallographic coordinates.
  • Tm thermal stability of peptide-HLA-E complexes
  • HLA-E*01:03 heavy chain and b2M light chain were crystallised in complex with the HLA-E-restricted Mycobacterial epitopes, IL9 (IMYNYPAML) and Mtbl4 (RMAATAQVL) (McMurtrey et al., 2017a) (Joosten et al., 2010).
  • HLA-E-IL9 (IMYNYPAML) crystals diffracted to 1.7 A whereas diffraction data was collected to 2.05 A for HLA-E-Mtbl4 (RMAATAQVL).
  • RMAATAQVL HLA-E-IL9
  • Mtbl4 (RMAATAQVL) adopted the classical conformation also adopted by canonical VL9, with buried termini and a solvent-exposed central kink at residues 4 & 5 ( Figure 5, A).
  • the less constrained central kinked region of Mtbl4 displayed the greatest movement relative to VL9 with 1.6 and 1.7 A separating Ca atoms at positions 4 and 5, respectively, resulting in the central portion of the Mtbl4 backbone leaning closer toward the HLA-E al helix.
  • the alternative IL9 peptide configuration features a position 7 Ca atom that projects 2.0 A further toward solvent relative to VL9 (VMAPRTVLL) and a position 7 secondary anchor side chain which projects away from its corresponding E pocket in the HLA-E groove.
  • HLA-E-IL9 INPYNYPAML
  • HLA- E-Mtbl4 RMAATAQVL
  • RYSPTSIL HLA-E-RL9HIV
  • the secondary E pocket of HLA-E comprises a deep hydrophobic recess with a more defined pocket-like aspect in contrast to classical MHC class la molecules which contain a large, highly conserved E pocket-occluding Trp side chain at position 147 ( Figure 6, A, i).
  • the E pocket of HLA-E classically accommodates a downward-projecting medium sized hydrophobic side chain at position 7 of the bound peptide, such as the highly conserved Val or Leu present in VL9 variants ( Figure 6, A, ii).
  • the unusual conformational dimorphism of the IL9 (IMYNYPAML) peptide primarily involves position 7 with the position 7 Ca atom projecting 2 A further toward solvent relative to canonical VL9 in one of the two observed peptide configurations in the asymmetric unit.
  • HLA-E-associated peptides including IL9 commonly participate in crystal packing interfaces, such structural polymorphism may reflect weak tethering of the small position 7 Ala side chain of IL9 to the deep secondary E pocket of HLA-E, in turn rendering it susceptible to crystal packing-induced repositioning.
  • a polar, non-canonical Gin at position 7 of Mtbl4 (RMAATAQVL) is buried within the hydrophobic E pocket and forms water- mediated hydrogen bonds to Ser-143 and Ser-147 of the HLA-E a2 helix with resulting entropic penalties ( Figure 6, A, V).
  • Combined mutagenesis and ELISA-based HLA-E peptide binding or thermal melt assays support a major contribution of the E pocket to HLA-E complex stability; the introduction of a canonical position 7 Val in place of the polar Gin in Mtbl4 resulted in a 36% increase in the ELISA-based binding signal and a 3.6°C increase in thermal stability (Figure 6, B & C).
  • VMAPRTVLL a polar Gin at position 7 of VL9
  • Example 6 Unique structural features distinguish canonical VL9 versus pathogen peptide-associated HLA-E Comparative structural analyses revealed a distinct configuration located in the a2 -helical kink region which distinguishes VL9-associated versus non-VL9-bound HLA-E ( Figure 7, A).
  • Mtb44 was the only pathogen peptide which exhibited comparable binding to VL9 in previous ELISA-based screens in addition to being the only pathogen peptide with partially preserved position 5 hydrogen bonding, perhaps eluding to the importance of centrally- positioned intermolecular bonds for HLA-E complex stability (Walters, McMichael and Gillespie, 2020).
  • HLA-E complex stability Wang, McMichael and Gillespie, 2020.
  • the loss of the hydrogen bond between Gln-156 and the VL9 position 5 main chain results in Gln-156 side chain repositioning of up to 2.2 A ( Figure 7, C, i).
  • the Glu-152 side chain in pathogen peptide-bound HLA-E structures projects up to 4 A further toward the N-terminus of the peptide binding groove relative to Glu-152 in VL9-bound HLA-E, which in turn facilitates novel inter chain hydrogen bonding with the position 3 Tyr side chains present in IL9 (IMYNYPAML) and RL9HIV (RMYSPTSIL) peptides ( Figure 7, D).
  • IL9 IL9
  • RYSPTSIL RL9HIV
  • Figure 7, D A conserved position 3 Tyr side chain orientation with clear electron density is adopted by both peptides, in which the superposed position 3 Tyr side chains project toward the HLA-E a2 -helix and form a novel hydrogen bond with the re-orientated Glu-152.
  • Glu-152 side chain orientation present in HLA-E-VL9 structures prohibits such hydrogen bond formation as it is positioned 5.5 A from the position 3 Tyr side chain present in the RL9HIV (RMYSPTSIL) or IL9 (IMYNYPAML) peptides.
  • RYSPTSIL RL9HIV
  • IL9 IL9
  • canonical VL9 peptides contain a highly conserved Ala at position 3 which projects downward into the shallow secondary D pocket, these data help redefine what can be stably accommodated at the secondary anchor position 3 of non-VL9 HLA-E binding peptides.
  • compensatory, non-pocket-based peptide- HLA-E hydrogen bonding can permit secondary anchor side chain accommodation when a lack of shape complementarity for the corresponding pocket prohibits peptide side- chain occupancy.
  • Example 7 - single mutations or a combination of H99Y, F116Y and S147W mutations dramatically improve the binding of the RL9HIV peptide to HLA-E
  • Pathogen-derived peptide HLA-E complexes including RL9SIV RMYNPTNIL, RL9HIV RMYSPTSIL, BZLF1 SQAPLPCVL, Mtbl4 RMAATAQVL and IL9 IMYNYPAML yielded conformationally heterogeneous populations and elongated molecular envelopes with average dimensions exceeding those of superimposed crystallographic coordinates.
  • crystal structures of IL9-, Mtbl4- and RL9HIV-bound HLA-E likely represent a subpopulation of the heterogeneous conformational ensemble observed in SEC-SAXS analyses which is also signified by the faint compact band observed in blue native gel analyses.
  • differential local structural features were identified and may be linked to complex instability and elongated molecular envelopes observed in solution.
  • IL9, Mtbl4 and RL9HIV peptides contain the canonical primary anchor Met at position 2 and Leu at position 9 which optimally occupy the primary B and F pockets, respectively.
  • Trp- 147 is highly conserved among classical MHC class I molecules with its bulky side chain largely occluding the E pocket whereas the smaller Ser-147 present in HLA-E and its murine and rhesus counterparts, Qa-1 and Mamu-E, results in a deep hydrophobic recess with a discrete pocket-like nature (O’Callaghan et al., 1998).
  • Another structural feature which distinguishes pathogen peptide- versus canonical VL9- associated HLA-E is a differentially positioned a2 helical kink region.
  • Glu-152 is a common TCR-interacting residue and Glu-152 to Ala mutation results in a > 10-fold reduction in CD94/NKG2A binding REF, its exclusive repositioning in structures of HLA-E lacking associated leader peptide could signal loss of VL9 to both HLA-E-restricted CD8+ T cells and NK cells, respectively, in an innate-like manner that is not wholly dictated by peptide-specific interactions (Sullivan et al., 2007). Accordingly, semi-invariant Qa-1 -restricted lymphocytes with a common Va recognised non-Qdm peptides presented by Qa-1 in a TAP-independent manner (Doorduijn et al., 2018).
  • Innate-like T cell recognition has previously been reported for other unconventionally-restricted subsets such as semi-invariant natural killer T (iNKT) cells restricted by CD Id and mucosal-associated invariant T (MAIT) cells restricted by MR1 (Cotton et al., 2018). Further, a number of structural studies have demonstrated multiple distinct modes of CD 1 -restricted TCR recognition which transcend the epitope/antigen-presenting-molecule co-recognition paradigm with minimal direct contact between the TCR and CD 1 -associated lipid antigen (Birkinshaw et al., 2015) (Wun et al., 2018).
  • canonical VL9 peptides contain a highly conserved Ala at position 3 which projects into the shallow secondary D pocket, Glu-152 repositioning and resultant hydrogen bond formation with the position 3 Tyr side chain of HLA-E-bound pathogen peptides, IL9 and RL9HIV, redefines what can be stably accommodated at the secondary anchor position 3 of non-VL9 HLA-E binding peptides - such distinct intermolecular hydrogen bonding appears to compensate for a lack of anchor side chain shape complementarity with its corresponding D pocket.
  • a VL9-prevalent MHC- E-presented ligandome in healthy cells would in turn maintain inhibitory NK cell interactions through CD94/NKG2A engagement, whereas an abundance of alternative HLA-E-restricted peptides with comparable or higher binding affinity than VL9 could conceivably disrupt this immunoregulatory interaction. Consistent with this, the inventors have demonstrated that more diverse peptide sampling by HLA-E, in settings where MHC class I trafficking pathways are disrupted - such as in Mtb infection or following RhCMV 68-1 vaccination - is largely suboptimal.
  • This non-optimal HLA-E peptide binding repertoire yields unstable, heterogeneous protein populations which at the structural level, share altered conformational adjustments relative to HLA-E-VL9 involving distinct a2-helical kink configurations that are likely to impact TCR and CD94/NKG2 class receptor recognition.
  • the inventors have demonstrated an optimised process to identify peptide epitopes presented by HLA-E on normal and abnormal cells and recognised by specific T cells and/or B-cells. Similarly, optimised methods to generate stable HLA-E-peptide complexes that can be used to make multimers to detect antigen specific T and B cells are demonstrated.
  • Methods are described to identify which peptides bind well, moderately or poorly to HLA-E, and then apply biochemical methods to enhance peptide binding to HLA-E, such as mutational methods to enhance peptide binding to HLA-E, chemical cross linking to enhance peptide binding, use of mutated Mamu-E to enhance peptide binding and to allow use of Mamu-E multimers as surrogates for HLA-E-peptide multimers. Additionally, the inventors demonstrate efficient validation of the epitopes identified by selecting, cloning and testing T cells/ B-cells specific for HLA-E-peptide complex presented on abnormal cells, thereby validating the therapeutic potential of those identified peptides, otherwise overlooked in conventional screening.
  • Peptides were purchased as lyophilised powder at >85% purity from Genscript USA prior to reconstitution in DMSO (200mM) and storage at -80 °C.
  • a UV-labile peptide based on the HLA-B leader peptide (VMAPRTLVL) with a 3-amino-3-(2-nitrophenyl)- propionic acid residue (J residue) substitution at position 5 was synthesised by Dris Elatmioui at LUMC The Netherlands, for use in peptide binding assays.
  • HLA-E*01:03 protein refolds were assembled in the traditional macro-refolding buffer for MHC class I molecules comprising 100 mM Tris pH8.0, 400mM L-arginine monohydrochloride, 2mM EDTA, 5mM reduced glutathione and 0.5mM oxidised Glutathione, prepared in MiliQ water.
  • 2-Microglobulin in Urea-Mes was initially refolded for 30 min at 4°C at a final concentration of 2 mM.
  • Test peptide was subsequently added to the refold at a concentration of 30-60 pM followed by HLA- E*01:03 heavy chain which was pulsed into the refolding buffer to reach a final concentration of 1 pM.
  • Refolds were subject to a 72 hour incubation period at 4°C prior to filtration through 1.0 pM cellular nitrate membranes to ensure the removal of aggregated material. Refolds were concentrated by a VivaFlow 50R system and VivaSpin Turbo Ultrafiltration centrifugal devices, both with 10 kDa molecular weight cut-offs. Refolded and concentrated material was used at a concentration of lOmg/mL in SEC-SAXS experiments without subsequent chromatographic separation.
  • Refolded and concentrated material intended for ELISA-based HLA-E peptide binding assays, DSF or crystallisation screening was subject to subsequent fast protein liquid chromatography (FPLC) size separation on an AKTA Start System using a Superdex S75 16/60 column.
  • FPLC fast protein liquid chromatography
  • HLA-E protein complex peaks were eluted into 20 mM Tris pH8, 100 mM NaCl and discriminated from non-associated b2M and large misfolded aggregates via elution profile visualisation by UV absorbance at 280 mAU.
  • FPLC- purified protein peaks fractions were combined and concentrated to a desired concentration for subsequent experiments using 10 kDa cut-off VivaSpin Turbo Ultrafiltration centrifugal devices - the final protein concentration was obtained by measurement of the absorbance at 280nm using a NanoDrop ND-1000 Spectrophotometer.
  • the composition of eluted protein samples was also analysed by non-reducing SDS-PAGE electrophoresis on NuPAGETM 12% Bis-Tris protein gels to demonstrate the presence of non-aggregated HLA-E heavy chain and b2M.
  • HLA-E- 2m complexes previously refolded with the UV-sensitive VL9 peptide were incubated in the presence of molar excess test peptide and evaluated via the Blue Native-PAGETM Novex Bis-Tris gel system (life technologies) (Walters et al., 2018b).
  • pre-refolded and purified HLA-E in complex with the UV-sensitive peptide was incubated at RT in the presence of 12 M excess test peptide prior to the addition of 3 pL 4x Native-PAGETM Sample Buffer per 10 pg (10 pL) of sample.
  • HLA-E- 2M-peptide refolds were assembled in the L-Arginine-Tris macro refolding buffer according to the protein refolding method detailed above and incubated for 72hrs at 4°C prior to concentration with the VivaFlow 50R system with a lOkDa molecular weight cut-off (Sartorius).
  • HPLC high-performance liquid chromatography
  • HPLC elution buffers corresponded exactly to the L- Arginine Tris pH 8 macro-refolding injection buffer minus the protein components - although 60pM or 120pM excess peptide was added to the elution buffer for certain SEC-SAXS experiments.
  • SEC-SAXS data were collected at Diamond Light Source Beamline B21 and images were taken every 3 seconds of X-ray-exposed HPLC-purified material over the course of a 32 minute elution period. Scattering data were circularly integrated prior to buffer subtraction followed by Guinier fitting and pairwise distribution function calculations, which were performed in the SAXS-dedicated software, ScAtter, developed by Robert Rambo (Franke et al., 2017).
  • Peptide exchange ELISA-based HLA-E peptide binding assays were conducted according to a previously published method that was developed and optimised by the inventors. Peptide exchange micro-reactions were assembled in the traditional macro refolding buffer for MHC class I molecules comprising 100 mM Tris pH8.0, 400 mM 1- arginine monohydrochloride, 2 mM EDTA, 5 mM reduced glutathione, and 0.5 mM oxidized Glutathione, prepared in MiliQ water.
  • Peptide exchange reactions were subsequently interrogated by sandwich ELISA.
  • 96- well ELISA plates were coated in lOmg/mL 3D12, an anti-human HLA-E capture antibody prior to a 12 h incubation period at 4°C.
  • ELISA plate wells were blocked with 300 pL of 2% IgG-free BSA for 2 h at RT. Blocked wells were washed five times in 0.05% Tween-based ELISA wash buffer (BioLegend) followed by a single wash in PBS prior to the addition of 50 pL of peptide exchange reaction diluted 1: 100 in 2% BSA to each well.
  • ELISA plates containing peptide exchange reaction samples were incubated for 1 h at RT and subsequently washed in 0.05% Tween-based ELISA wash buffer and PBS.
  • a polyclonal anti-human b2M HRP- conjugated IgG detection antibody (ThermoFisher Scientific) was diluted 1:2500 in 2% BSA and 50pL added to each ELISA well.
  • ELISA plates were incubated in the dark for 30 min prior to wash steps in a 0.05% Tween-based ELISA wash buffer and PBS.
  • TMB 3,3',5,5'-tetramethyl benzidine
  • Pearson product-moment correlation coefficients were calculated for DSF-determined melting temperatures and normalised ELISA-based peptide binding signals to establish the congruity between these techniques. Additionally, linear correlations were identified between melting temperatures or normalised ELISA-based signals and various SEC-SAXS parameters including the maximal dimensions of the average conformation in solution (dmax) or the volume of the DAMMIF ab initio molecular envelope model (A3).
  • Crystals were cryopreserved in 25% glycerol by Dr. Karl Harlos of Oxford University and diffraction data were collected at the Diamond Light Source, beamlines i04 (HLA-E-Mtbl4 structure) and i03 (HLA-E-IL9 structure). Diffraction data were auto-indexed by Xia2 DIALS. Since the outer shell CCl/2 exceeded the minimum threshold (>0.3) for both datasets, no reflections were excluded from downstream analysis. A more conservative data truncation approach according to the Rmerge and I/sigma cut-offs has been shown to result in the elimination of useful data which would otherwise have contributed to model quality (Karplus and Diederichs, 2012).
  • thermostability of canonically refolded HEA-E-b2ih peptide complexes and C terminus extended peptides with a cysteine refolded with HLA-E containing a tyrosine to cysteine mutation was determined by heat-induced fluorescent dye incorporation, using the commercially available Protein Thermal Shift Dye KitTM (Applied Biosystems). 5 pg of test HEA-E-b2ih complexes was aliquoted into 0.1 mL MicroAmp Fast Optical 96-well plates containing pre-mixed Protein Thermal Shift Dye and Protein Thermal Shift Buffer.
  • Sample buffer (either PBS or Tris pH8, 100 mM NaCL) was added to achieve a final volume of 20 pL.
  • Control samples reconstituted with buffer were prepared to monitor background fluorescent signal. Both samples and controls were set up in quadruplicate.
  • Thermal-driven dye incorporation was measured on an Applied Biosystem Real-Time 7500 Fast PCR System. Data was collected over a temperature ramp ranging from 25 to 95 °C, with 1 °C intervals. Melt curve data were analysed using Protein thermal Shift Software vl.3, and median Derivative Tm values (°C) are reported.
  • Excitation power was pre-adjusted to obtain between 8000 and 20,000 Raw Fluorescence Units for fluorescence emission at 330 nm and 350 nm.
  • a thermal ramp ranging from 20 °C to 95 °C, at a rate of 1 °C/min, was applied.
  • Automated thermal melt data calling was generated by the analysis software within PR.ThermControl, (version 2.1.5) software.

Abstract

The invention relates to methods of identifying a peptide which is capable of forming an HLA-E:peptide complex and being recognised by a T-cell and or B-cell, and methods of improving the stability of the HLA-E:peptide complex. Further, the invention relates to methods of identifying one or more T-cell or B-cell which recognises an HLA-E:peptide complex.

Description

PEPTIDE SCREEN
FIELD
The present invention relates to methods of identifying a peptide which is capable of forming an HLA-E:peptide complex and being recognised by a T-cell and or B-cell, and methods of identifying one or more T-cell or B-cell which recognises an HLA-E:peptide complex.
BACKGROUND HLA-E is a non-polymorphic HLA class I molecule. There are two major alleles in the population differing only in one amino acid at position 107 which is outside the peptide binding groove (Strong et al., Correlating differential expression, peptide affinities, crystal structures, and thermal stabilities. J Biol Chem. 2003;278(7):5082-90). The primary function of HLA-E is to bind a peptide usually termed ‘VL9’ which is derived from the signal peptide of classical HLA class I A, B, C molecules and HLA-G, but not HLA-E. The peptide has the sequence VMAPRTLVL, VMAPRTVLL, VMAPRTLLL, VMAPRTLIL, or VMAPRTLFL. The HLA-VL9 complex in turn binds to the NKG2A- CD94 inhibitory or NKG2C-CD94 activating receptors on natural killer cells and a subset of T cells (Braud et al., HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C. Nature. 1998 ;391 (6669) : 795 -9) .
Although the well-established primary role of HLA-E and its functionally conserved murine and rhesus orthologues, Qa-1 and Mamu-E, is to regulate NK-cell activity via MHC class I signal peptide presentation, specific contexts appear to permit diversified peptide binding repertoires and MHC-E-restricted CD8+ T-cell priming in vivo (Wu et al., 2018) (Vance et al., 1998). An alternative functional role for HLA-E and its rhesus counterpart, Mamu-E, as CD8+ T cell restriction elements has been demonstrated in settings of Mycobacteria infection or exposure and SIV/HIV recombinant CMV vector vaccination, respectively (Hansen et al., 2016) (McMurtrey et al., 2017b) (Prezzemolo et al., 2018). Notably, Mamu-E-restricted CD8+ T cell responses confer unprecedented sterile immune protection against SIV challenge in around 50% of vaccinated macaques (Malouli et al., 2020) (Hansen et al., 2016). Sequence-diverse peptide epitopes with non-canonical primary and secondary anchor residues derived from SIV, HIV and Mtb were subsequently identified in these settings, the majority of which demonstrated unusually low HLA-E binding (<10% of VL9 binding) for immunogenic epitopes in an ELISA-based screening assay (Walters, McMichael and Gillespie, 2020). This ELISA- based assay also identified a modest number of HLA-E-restricted pathogen-derived epitopes which demonstrated HLA-E binding with normalised signals >20% of the positive control VL9 signal. Notably, such HLA-E-restricted pathogen peptides include a number of physiologically relevant epitopes such as the immunodominant SIV derived supertope, RL9SIV (RMYNPTNIL)(SEQ ID NO: 1).
Thus, although HLA-E is non-polymorphic and HLA-E restricted responses to pathogens have thus far been poorly characterised, further identification of T cell responses to peptide antigen bound to HLA-E may be useful for immunotherapies which could be applicable universally in the population due to a lack of HLA-E genetic polymorphism; the HLA-E locus encodes two non-synonymous functional allelic variants HLA-E*01:01 and HLA-E*01:03 that differ by a single Arg or Gly amino acid residue, respectively, at position 107, resulting in largely overlapping peptide binding repertoires (O’Callaghan et al., 1998) (Strong et al., 2003). Minimal allelic polymorphism and shared peptide binding repertoires position HLA-E as a particularly attractive restriction element for T cell-targeted vaccination strategies which could potentially offer universal, MHC class I allotype-unrestricted protection.
For example, TCRs and/or monoclonal antibodies specific for HLA-E in complex with a peptide antigen can be generated and used therapeutically as cytotoxic reagents, or such antibodies and TCRs could manipulated as receptors, including chimeric receptors, which are transfected or transduced into effector cells to induce immune responses against the peptide antigen. Therefore, the generation of antibodies or T cells and B- cells which recognise HLA-E bound to peptide antigens derived from a cancer, pathogen or even autoantigens has considerable therapeutic potential.
To achieve this, a multistep process is necessary to identify peptides that bind to HLA- E, then identify whether such HLA-E:peptide complexes can be recognised by a T-cell, such as a CD8+ T-cell, and/or a B-cell, which may induce an immune response such as a CD8+ T cell response or a B lymphocyte-mediated antibody response, in vitro or in vivo in humans or in animal models.
As many potential peptides which are capable of binding HLA-E and initiating an immune response bind much weaker relative to the canonical VL9 peptide, as discussed above and in Walters, McMichael and Gillespie, 2020, many potentially therapeutically relevant peptides may not bind to HLA-E in a manner stable enough to undertake the multistep process described above, and may be bypassed in known selection processes. It is therefore an aim of the invention to provide improved methods of identifying peptides which are capable of forming an HLA-E:peptide complex and being recognised by a T-cell and/or B-cell.
SUMMARY
Accordingly, in an aspect there is provided a method of identifying a peptide which is capable of forming an HLA-E:peptide complex and being recognised by a T-cell and/or B-cell, comprising the steps of:
(a) Performing an assay to determine the level of binding of a peptide of interest in a HLA-E:peptide complex, relative to the level of binding of a reference peptide to
HLA-E in a reference HLA-E complex;
(b) Stratifying the peptide of interest into one of the following groups based on its relative binding determined in step (a):
(i) More than about 70% of the level of binding to HLA-E relative to the level of binding of the reference peptide;
(ii) Less than about 70% of the level of binding to HLA-E relative to the level of binding of the reference peptide; and
(c) Performing an assay to determine whether a HLA-E:peptide complex comprising the peptide of interest is recognised by a T-cell and/or B-cell.
A peptide of interest stratified into group (ii) in step (b) may be further stratified into one of the following groups based on its relative binding determined in step (a):
(al) Between about 35% to about 70% of the level of binding to HLA-E relative to the level of binding of the reference peptide; (bl) Between about 15% to about 35% of the level of binding to HLA-E relative to the level of binding of the reference peptide; or
(cl) Under 15% of the level of binding to HLA-E relative to the level of binding of the reference peptide.
The method may further comprise performing one or more step to improve the stability and/or level of binding of the HLA-E:peptide complex comprising the peptide of interest. The one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may be performed for step (c) of the method.
The reference peptide referred to herein may be a VL9 peptide (VMAPRT[V/L][L/V/I/F]L), such as VMAPRTLVL (SEQ ID NO: 2), VMAPRTVLL (SEQ ID NO: 3), VMAPRTLLL (SEQ ID NO: 4), VMAPRTLIL (SEQ ID NO: 5), or VMAPRTLFL (SEQ ID NO: 6). The reference peptide may be VMAPRTLLL.
A HLA-E:peptide complex referred to herein refers to a complex comprising or consisting of a peptide, HLA-E heavy chain and B2 microglobulin. The reference HLA- E:peptide complex referred to herein may refer to a complex comprising or consisting of a VL9 peptide, HLA-E and B2 microglobulin. The HLA-E may be HLA-E*01:01 or HLA-E*01:03. The HLA-E may be mutated, for example at the residues described herein. The HLA-E may refer to human HLA-E or a homologue from a non-human species, such as mouse or Rhesus monkey.
The assay of step (a) of the method allows the level of binding of a peptide of interest to HLA-E in an HLA-E:peptide complex to be assessed, relative to that of a reference peptide such as VL9, known to bind HLA-E strongly in an HLA-E:peptide complex. The peptide of interest can be given a numerical value relative to VL9 (where VL9 binding is 100%), and this allows the stratification of peptides of interest into different groups. The level of binding may be defined as the percentage of HLA-E:peptide complexes formed with the peptide of interest, versus complexes formed with the reference peptide, VL9.
Such stratification can be used to determine whether one or more step needs to be performed to improve the stability of the HLA-E:peptide complex comprising the peptide of interest, and if so, the nature of the step or steps required for a peptide of interest stratified into a given group.
The inventors have shown that many immunodominant peptides which bind to HLA-E, particularly peptides derived from pathogens and cancer derived epitopes, bind surprisingly weakly in an ELISA when compared to VL9. Biological investigation of such peptides in the pursuit of therapeutics such as T-cells and antibodies targeting such pathogens and cancers, and indeed infected or malignant cells, is difficult as current techniques rely on biological functional assays which utilise stable HLA-E:peptide complexes and multimers of such complexes. The result to date is that the majority of pathogen and cancer-derived peptides which have the capacity to be presented on a cell surface by HLA-E to be recognised by a T-cell and/or B-cell, to induce an immune response, have largely been bypassed, for example in high-throughput studies. This is at least in part due to a lack of knowledge that such peptides and HLA-E itself can be manipulated, depending on their binding capacity relative to a canonical VL9 peptide, to allow the formation of more stable HLA-E:peptide complexes on which biological assays can be reliably performed, for the downstream identification novel therapeutics which recognise infected or cancerous cells presenting such peptides. The invention therefore permits the appropriate action to be taken with a given peptide of interest to stabilise its binding to HLA-E in a HLA-E:peptide complex, such that further study can be undertaken.
Methods of improving binding stability
If in the assay of step (b) the peptide of interest is stratified into group (i), the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may not need to be performed. A peptide of interest which binds to HLA-E in a HLA-E:peptide complex at more than about 70% of the level of binding relative to the reference peptide will be stable enough to progress to biological functional biological assays without further stabilisation. One or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may still be performed for a peptide of interest stratified into group (i), if desired.
If in the assay of step (b) the peptide of interest is stratified into group (ii), one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may be performed.
The one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may comprise one or more of:
(A) First forming a HLA-E:peptide complex using a UV-sensitive reference peptide and subsequently exchanging this with the peptide of interest such that an HLA- E:peptide complex comprising the peptide of interest is formed;
(B) Crosslinking one or more amino acid of the peptide of interest to one or more amino acid of HLA-E;
(C) Introducing one or more crosslink between amino acids in HLA-E; (D) Introducing one or more mutation in HLA-E, such as one mutation, two mutations or three mutations, to increase the stability of the binding of a peptide of interest in the HLA-E:peptide complex; or
(E) Forming a complex of the peptide of interest with Mamu-E.
Any reference to HLA-E in any of (A)-(D) may refer to human HLA-E, which may be or be derived from HLA-E*0101/ E*0103
(GSHSLKYFHTSVSRPGRGEPRFISVGYVDDTQFVRFDNDAASPRMVPRAPWMEQ EGSEYWDRETRSARDTAQIFRVNLRTLRGYYNQSEAGSHTLQWMHGCELGPDG RFLRGYEQFAYDGKDYLTLNEDLRSWTAVDTAAQISEQKSNDASEAEHQRAYLE DTCVEWLHKYLEKGKETLLHLEPPKTHVTHHPISDHEATLRCWALGFYPAEITLT WQQDGEGHTQDTELVETRPAGDGTFQKWAAVVVPSGEEQRYTCHVQHEGLPEP VTLRWKPASQPTIPIV GIIAGLVLLGS VV SGAVVAAVIWRKKS SGGKGGSY SKAE WSDSAQGSESHSL (SEQ ID NO: 7); or GSHSLKYFHTSVSRPGRGEPRFISVGYVDDTQFVRFDNDAASPRMVPRAPWMEQ EGSEYWDRETRSARDTAQIFRVNLRTLRGYYNQSEAGSHTLQWMHGCELGPDR RFLRGYEQFAYDGKDYLTLNEDLRSWTAVDTAAQISEQKSNDASEAEHQRAYLE DTCVEWLHKYLEKGKETLLHLEPPKTHVTHHPISDHEATLRCWALGFYPAEITLT WQQDGEGHTQDTELVETRPAGDGTFQKWAAVVVPSGEEQRYTCHVQHEGLPEP VTLRWKPASQPTIPIV GIIAGLVLLGS VV SGAVVAAVIWRKKS SGGKGGSY SKAE
WSDSAQGSESHSL (SEQ ID NO: 8), which are the amino acid sequences of HLA- E*0101 and E*0103 respectively without an N-terminal signal sequence.
Mamu-E may optionally be mutated to enhance peptide binding, stability or reactivity with monoclonal antibodies such as the anti -HLA-E antibody 3D 12. Mamu-E comprises an almost identical peptide binding groove to HLA-E. However, Mamu-E possesses structural differences outside of the peptide binding groove which result in a more stable complex, and allowing the binding of more peptides. Mamu-E may be used as an alternative to HLA-E in the Single Chain Trimer assay described herein. This enables the identification of yet further peptides which may be used as epitopes with HLA-E. Mamu-E may comprise one or more of the following mutations: P57S, E79R, and/or G150A. These residues interact with TCRs, and such a Mamu-E:peptide complex may be used to generate T cells, TCRs, B-cells and BCRs (including antibodies) for therapeutic development. If in the assay of step (b) the peptide of interest is stratified into group (al), the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may comprise:
(A) First forming a HLA-E:peptide complex using a UV-sensitive reference peptide and subsequently exchanging this with the peptide of interest such that an HLA- E:peptide complex comprising the peptide of interest is formed.
If in the assay of step (b) the peptide of interest is stratified into group (bl), the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may comprise one or more of:
(C) Introducing one or more crosslink between amino acids in HLA-E; and/or
(D) Introducing one or more mutation in HLA-E, such as one mutation, two mutations or three mutations, to increase the stability of the binding of a peptide of interest in the HLA-E:peptide complex.
If in the assay of step (b) the peptide of interest is stratified into group (cl), the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest may comprise:
(B) Crosslinking one or more amino acid of the peptide of interest to one or more amino acid of HLA-E.
(A) may be performed in the presence of a molar excess of the peptide of interest, such as at IOOmM or more, to ensure stability of the complex. In (A), the sensitive reference peptide may be UV-sensitive. The UV-sensitive reference peptide may be a UV- sensitive VL9 peptide, such as VMAP(J*)TLVL (SEQ ID NO: 9), where J* is 3-amino- 3-(2-nitrophenyl)-propionic acid. (A) may be performed in the presence or absence of UV light.
In (B) the crosslinking one or more amino acid of the peptide of interest to one or more amino acid of HLA-E may comprise: mutating one or more residues of HLA-E to a cysteine or lysine, and optionally substituting one or more residues of the peptide of interest, to a cysteine, lysine or synthetic amino acid, such that a crosslink is capable of being formed between the peptide of interest and HLA-E. This may involve the use of an additional small molecule to bridge the amino acids to be crosslinked. This may involve the formation of a salt bridge between oppositely charged amino acids of the peptide and HLA-E.
The substituting one or more residues of the peptide of interest may comprise substituting the residue in the first or second position of the peptide of interest to a cysteine, a homocysteine, or a synthetic amino acid comprising a free sulphydryl group, such that a crosslink in the form of a disulphide bond can be formed between a mutant amino acid in the HLA-E heavy chain and the amino acid at the first or second position in the peptide.
The crosslinking one or more amino acid of the peptide of interest to one or more amino acid of HLA-E may comprise mutating the tyrosine at position 84 of HLA-E to a cysteine, adding a glycine and cysteine to the carboxy terminus of the peptide of interest, and forming a disulphide bond between the cysteine at position 84 of HLA-E and the cysteine added to the carboxy terminus of the peptide of interest.
The crosslinking one or more amino acid of the peptide of interest to one or more amino acid of HLA-E may comprise mutating the methionine at position 45 of HLA-E to a cysteine, substituting the amino at position two of the peptide of interest to a cysteine, a homocysteine, or a synthetic amino acid that displays a free sulphydryl group (preferably at the end of a side chain of preferred length), and forming a disulphide bond between the cysteine at position 45 of HLA-E and the cysteine, homocysteine, or a synthetic amino acid at position two of the peptide of interest.
The synthetic amino acid comprising a free sulphydryl group may be a homocysteine analogue, (2S)-2-amino-5-sulfanylpentanoic acid or (2S)-2-amino-6 sulfanylhexanoic acid.
In (C) the one or more crosslink between amino acids in HLA-E may be introduced by mutating the tyrosine at position 84 of HLA-E to a cysteine, mutating the alanine at position 139 of HLA-E to a cysteine, and forming a disulphide bond between the two cysteine residues at position 84 and 139 of HLA-E. This crosslink may improve the binding of a peptide of interest in the HLA-E:peptide complex, demonstrated by increased Tm (Figure 8).
In (D) the one or more mutation in HLA-E may comprise or consist of one or more of, such as one of, two of, or all of, mutating Histidine at position 99 to Tyrosine, mutating Phenylalanine at position 116 to Tyrosine or mutating Serine at position 147 to Tryptophan. These mutations close/alter different HLA-E binding pockets which are used optimally by the signal peptide VL9. This is in contrast to many low affinity pathogen or cancer-derived peptides that bind to HLA-E. These mutations enhance the binding of the pathogen or cancer-derived peptides, demonstrated by increased melting temperature (Tm) of the protein (Figures 9,11 and 12).
In (D), the one or more mutation in HLA-E may comprise swapping the alpha-3 domain of HLA-A3 with that of HLA-E.
Accordingly, in another aspect, the invention provides a mutant HLA-E heavy chain comprising one or more mutation which permits the formation of a HLA-E:peptide complex with increased stability when compared to the complex without the mutant HLA-E heavy chain. The complex may further comprise b2 microglobulin. The mutant HLA-E may comprise a mutation at Histidine at position 99 to Tyrosine. The mutant HLA-E may comprise a mutation at Phenylalanine at position 116 to Tyrosine. The mutant HLA-E may comprise a mutation at Serine at position 147 to Tryptophan. The mutant HLA-E may comprise a mutation at Histidine at position 99 to Tyrosine and a mutation at Phenylalanine at position 116 to Tyrosine. The mutant HLA-E may comprise a mutation at Histidine at position 99 to Tyrosine and a mutation at Serine at position 147 to Tryptophan. The mutant HLA-E may comprise a mutation at Histidine at position 99 to Tyrosine, a mutation at phenylalanine at position 116 to Tyrosine, and a mutation at Serine at position 147 to Tryptophan.
The one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest referred to herein may also be utilised in any assay referred to herein.
Peptide Prediction
The peptide of interest may be identified using one or more amino acid sequence prediction process. A peptide of interest may be identified as comprising a HLA-E restricted sequence motif (Walters et al., 2020 EJI). A peptide of interest may be identified using one or more publicly available programme such as NetMHC, or using an algorithm. A peptide of interest may be identified by using one or more of the following search parameters: peptide is a nonamer; peptide comprises one or more proline between positions 3 and 7, peptide comprises a restricted set of amino acids at positions 2 and 9, such as an M, L, V A, Q, or F at position 2, and/or a L, F, I, V or M at position 9. Single Chain Trimer
The method may further comprise, before step (a), performing an initial screening assay to determine whether a HLA-E:peptide complex comprising the peptide of interest is capable of being expressed on a cell surface, wherein if the HLA-E:peptide complex comprising the peptide of interest is determined to be capable of being expressed on a cell surface, the method proceeds to step (a).
The initial assay may comprise transfecting into a cell, such as HEK293T cells, a nucleic acid encoding a polypeptide comprising the peptide of interest, B2microglobulin and HLA-E heavy chain, in that order. Such a polypeptide is also referred to herein as a
Single Chain Trimer (SCT). The nucleic acid may also encode a linker sequence between each of the peptide of interest, B2microglobulin and HLA-E heavy chain, preferably in that order. After expression of the SCT, the initial assay may comprise performing an experimental technique to detect the presence of the peptide of interest bound to HLA-E at the surface of the cell which expresses the SCT. Many experimental techniques are known to the skilled person in the art, such as flow cytometry, and microscopy such as confocal microscopy. The experimental technique may utilise a conformation-specific antibody, which recognises a correctly folded HLA-E:peptide complex, for example which can be expressed at the cell surface. Such an antibody may recognise a domain in the a3 region of HLA-E.
Such an initial screening assay allows the skilled person to quickly, cheaply and reliably assess whether a peptide of interest can be presented on the cell surface by HLA-E. If a peptide of interest is not expressed at the cell surface bound to HLA-E, then the method may terminate at this stage for that peptide of interest. If a peptide of interest is expressed at the cell surface bound to HLA-E, the method may proceed. The one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest referred to herein may also be utilised in the initial screening assay. ELISA
In the method, the assay of step (a) may be an Enzyme-linked immunosorbent assay (ELISA). The ELISA may be a sandwich ELISA. The sandwich ELISA assay may utilise an anti HLA-E antibody and an anti B2 microglobulin antibody.
The ELISA may be used to semi-quantitatively determine the level of binding of a peptide of interest in a HLA-E:peptide complex, relative to the level of binding of a reference peptide to HLA-E in a reference HLA-E complex. The level of binding of a reference peptide to HLA-E in a reference HLA-E complex may be taken as 100%. The reference peptide may be VL9, described above. The VL9 may be VMAPRTLLL.
The HLA-E:peptide complex comprising the peptide of interest used in the ELISA may be formed using a peptide exchange method, in which a reference HLA-E:peptide complex comprising a UV-sensitive reference peptide is first formed, and the reference peptide is subsequently exchanged with the peptide of interest under conditions such that an HLA-E:peptide complex comprising the peptide of interest is formed.
For example, the HLA-E heavy chain may be expressed in in E.coli, before being refolded by mixing with B2microglobulin and a UV-labile VL9 peptide, such as VMAP(J*)TLVL where J is 3-amino-3-(2-nitrophenyl)-propionic acid, a synthetic amino acid that is sensitive to UV light. This protein complex may then be purified, for example by Fast Protein Liquid Chromatography (FPLC) size exclusion, and the protein complex of the expected mass may be collected and analysed. The complex, which is either fresh or previously frozen and thawed, may then be mixed with a 100-fold molar excess of the peptide, and optionally exposed to UV light.
The ELISA may use the methodology described in Walters et al., EJI, 2020.
Alternatively, or additionally the assay of step (a) may comprise or consist of performing Thermal melt (Tm) analysis by Differential Scanning Fluorometry (DSF), or Small Angle X-ray Scattering of the material after flowing through the size exclusion column (SEC-SAXS), which purifies the refolded HLA-E after the peptide exchange. Such steps may be performed on fresh or previously frozen material that has been thawed. The inventors have determined that the level of binding of a peptide of interest relative to VL9 binding to HLA-E in the ELISA assay has a strong correlation with the thermal stability of the HLA-E: peptide complex and its overall structural conformation in solution (Figure 4). Thus, binding in the ELISA assay and/or thermal melt determination is predictive of structural stability.
Biophysical Quality Control
The method may further comprise performing one or more further assay after step (b) to determine the biophysical and/or biochemical characteristics of the HLA-E:peptide complex comprising the peptide of interest. The one or more further assays may comprise Native PAGE, Thermal melt (Tm) analysis, and/or) Size Exclusion Chromatography- Small Angle X-ray Scatter (SEC-SAXS) .
From these data it is possible to predict the structural homogeneity and stability of the protein, which may be confirmed by small angle X ray scatter (SAXS).
Determining whether the peptide is capable of being recognised by a T-cell and/or B-cell
In step (c) of the assay, to determine whether an HLA-E:peptide complex comprising the peptide of interest is recognised by a T-cell and/or B-cell, one or more biological assay may be performed. The one or more biological assay may use of multimers of the HLA-E:peptide complex comprising the peptide of interest. The multimer may be a tetramer or a pentamer. A multimer described herein may comprise streptavidin or a microbead and may comprise a label such as a fluorochrome.
The one or more biological assay may be a FACS assay.
The one or more biological assay may comprise introducing the peptide of interest into a co-culture of T-cells and monocytes, in the presence of IL-2 and IL-15. The peptide of interest may then bind to HLA-E on the monocytes and be recognised by one or more T-cell.
One more T-cell and/or B-cell (or the corresponding TCR/BCR) identified in such a biological assay may then be selected, isolated, sequenced and/or cloned. The selected/isolated T-cell and/or B-cell may be expanded. For example, one or more identified T cell may be sorted using a multimer, and seeded at less than one cell per well into multiple wells of a tissue culture plate, and cultured with irradiated allogeneic peripheral blood mononuclear cells (PBMC) as feeders together with IL-2 and phytohemagglutinin (PHA).
Determining whether the peptide is capable of stimulating an immune response
In step (c) of the assay, it may be also determined whether an HLA-E:peptide complex comprising the peptide of interest is capable of stimulating an immune response in one or more T-cell and/or B-cell. To achieve this, one or more biological assay may be performed. The one or more biological assay may identify whether an HLA-E:peptide complex comprising the peptide of interest is capable of inducing an increase in IFN-g, TNF-a, CD107a/b and/or CD137 expression in T-cells. The skilled person will appreciate that numerous other activation markers can be used to determine whether an immune response is stimulated in a T-cell.
The one or more biological assay may comprise the use of multimers of the HLA- E:peptide complex comprising the peptide of interest to determine whether an immune response in a T-cell and/or B-cell can be stimulated. The multimer may be a tetramer or a pentamer or a dextramer. A multimer described herein may comprise streptavidin or a microbead and may comprise a label such as a fluorochrome.
The one or more biological assay may comprise introducing the peptide of interest into a co-culture of T-cells and monocytes, in the presence of IL-2 and IL-15. The peptide of interest may then bind to HLA-E on the monocytes and stimulate an immune response in one or more T-cell.
One more T-cell identified as being stimulated in such biological assays may then be selected, isolated, sequenced and/or cloned. The selected/isolated T-cell may be expanded. For example, one or more identified T cell may be sorted using a multimer, and seeded at less than one cell per well into multiple wells of a tissue culture plate, and cultured with irradiated allogeneic peripheral blood mononuclear cells (PBMC) as feeders together with IL-2 and phytohemagglutinin (PHA).
To determine whether an HLA-E:peptide complex comprising the peptide of interest, or a multimer thereof, is capable of stimulating an immune response in one or more B- cells, one or more biological assay may be performed. The one or more biological assay may identify whether an HLA-E:peptide complex comprising the peptide of interest is capable of being bound by a BCR, and/or whether this stimulates the secretion of antibody.
If it is determined that the HLA-E:peptide complex comprising the peptide of interest, or a multimer thereof, is capable of stimulating an immune response in one or more T- cell and or B-cell, this is indicative of the T-cell and/or B-cell recognising the HLA- E:peptide complex comprising the peptide of interest.
In another aspect, there is provided a method of identifying one or more T-cell or B- cell which recognises an HLA-E:peptide complex comprising a peptide of interest, comprising:
(a) performing the method of the first aspect to identify a peptide which is capable of forming an HLA-E:peptide complex and being recognised by a T-cell and/or B-cell; and
(b) performing an assay to identify one or more T-cell or B-cell which recognises the HLA-E:peptide complex.
An immune response, as described above, may be stimulated in a T-cell and/or B-cell upon recognition of the HLA-E:peptide complex comprising the peptide of interest.
The assay to identify a T-cell or B-cell which recognises the HLA-E:peptide complex comprising the peptide of interest may comprise isolating T-cells and/or B-cells from a sample of blood or tissue/biological fluid from the site of infection or a tumour, of a healthy subject or from a subject who has been diagnosed with or is suspected of having cancer and/or an infection. One or more T-cell or B-cell which recognise the HLA- E:peptide complex comprising the peptide of interest may then be selected.
The subject may be a mammal. The subject may be a human.
Validation that the T-cell/ B-cell recognises abnormal cells
The method may further comprise performing an assay to determine whether a T-cell or B-cell which is identified by any of the methods described above, is capable of recognising and/or being activated by an abnormal cell. An abnormal cell may be a cell which is infected with a pathogen, a tumour cell, or a cell which is undergoing a stress response, and which comprises the peptide of interest. The stress response may be to one or more of: metabolic stress, physical stress, apoptotic stress, pH stress, osmotic stress, DNA Damage stress, heat stress, hypoxic stress or toxin stress.
For example, an identified T-cell or T-cell clone may be exposed to an abnormal cell and an assay performed to determine whether the T-cell is activated. Presence or level of activation markers such as one or more of CD137, CD107a/b, TNF alpha, or interferon-gamma may be used to determine whether the T-cell is activated by contact with the abnormal cell.
This step confirms that the peptide of interest is a biologically relevant epitope which is presented on an abnormal cell by HLA-E. This step also confirms that a multimer used for the T cell or B-cell selection displays the same antigen/epitope as is present on the abnormal cell; that is, where one or more step to improve the stability of the HLA- E:peptide complex comprising the peptide of interest has been performed, a resulting T-cell or B-cell identified is also capable of recognising and/or being stimulated by a corresponding natural HLA-E:peptide complex, in which no step to improve the stability has been performed. This step therefore excludes T cells and B-cells that recognise and/or respond to altered forms of the HLA-E (or indeed Mamu-E) multimers which result from on chemical, biochemical or mutational steps used in the one or more step to improve the stability described herein. Such T cells or B cells, or indeed their TCR or BCR respectively, may then be developed as a therapeutic reagent.
A resulting TCR and/or BCR may be sequenced and cloned. The nucleic acid encoding a TCR can then be used to generate soluble TCR or can be transfected or transduced into live cells such that they then express the receptor.
There are a number of methods suitable for the transfection or transduction of cells with nucleic acid (such as DNA, cDNA or RNA) encoding the TCRs or antibodies (see for example Robbins et al., (2008) J Immunol. 180: 6116-6131).
Any method described herein may further comprise generating one or more T-cell or TCR, or antibody such as a monoclonal antibody, which recognises the HLA-E:peptide complex comprising the peptide of interest. The skilled person will know of various methods to generate such an antibody, including immunising an animal such as a mouse such that antibodies which recognise the HLA-E:peptide complex comprising the peptide of interest are generated, or by using the Milstein method.
The skilled person will appreciate that preferred features of any one embodiment and/or aspect of the invention may be applied to all other embodiments and/or aspects of the invention.
The invention will now be described, by way of example only, by the following figures and examples.
BRIEF DESCRIPTION OF THE FIGURES
Figure 1 - shows Blue native (BN) gels signatures for pathogen peptide versus canonical leader peptide-loaded HLA-E complexes. 10pg of pre-refolded HLA-E- 2m-UV-sensitive VL9 (VMAP(J*)TLVL) material was incubated with 12M excess of test peptides (VL9, Mtb44, IL9, RL9H, RL9S and BLZF1) or the no added peptide buffer control (Eno pep) for 3 hours on ice prior to Blue Native-PAGE™ Novex 4-16% Bis-Tris gel evaluation. Gel electrophoresis was performed at 150 Volts for 2 h at RT, over a current range of 15-16 to 2-4 mAmps. Gels were subsequently rinsed in MilliQ water (3 times), stained for 2-3 h in SimplyBlue™ Safe Stain at RT, with final de- staining steps performed in MilliQ water over a period of 24 hour prior to imaging. The NativeMark™ Protein Standard 66kDa band is denoted for reference. The blue arrows denote both Compact (Cf) and Diffuse (Df) HLA-E gel forms.
Figure 2 - shows size exclusion chromatography-coupled small angle x-ray scattering of peptide-HLA-E complexes A. Ab initio molecular envelope models displayed as small ‘dots’ represent the average conformational protein state adopted in solution and were generated from SEC-SAXS data using the DAMMIF and DAMAVER ATSAS packages in conjunction with ScAtter (Franke & Svergun, 2009) (Volkov & Svergun, 2003). The top row of molecular envelope ‘dot’ models correspond to SEC- SAXS runs where HLA-E complexes were injected onto the HPLC column in the presence of 120mM excess peptide but no excess peptide was added to the HPLC elution buffer. By contrast, the bottom row correspond to SEC-SAXS runs where 120mM excess peptide was present in both the HPLC injection and elution buffers to ensure continual presence of excess ligand. From left to right, the purple-shaded molecular envelope ‘dot’ models represent canonical VL9 (VMAPRTVLL)-refolded HLA-E. The structural coordinates of HLA-E in complex with VL9 (VMAPRTVLL) from 1MHE are shaded green and were superimposed onto SEC-SAXS-based molecular envelopes using the SUPCOMB package of ATSAS (O’Callaghan et al., 1998). The neighbouring SEC- SAXS molecular envelope ‘dot’ models, shaded green, represent HLA-E refolded with Mtb44 (RLPAKAPLL) (SEQ ID NO: 10), onto which the previously published HLA-E- Mtb44 structural coordinates (shaded purple) were aligned (Walters et al., 2018). Next, the blue-shaded SEC-SAXS molecular envelope ‘dot’ models represents IL9 (IMYNYPAML)(SEQ ID NO: l l)-refolded HLA-E onto which the HLA-E-IL9 structural coordinates (shaded orange) were aligned. The magenta-shaded SEC-SAXS molecular envelope ‘dot’ models correspond to Mtbl4 (RMAATAQVL)(SEQ ID NO: 12)-refolded HLA-E, with the HLA-E-Mtbl4 structural coordinates superimposed via SUPCOMB. The neighbouring SEC-SAXS molecular envelope ‘dot’ models shaded yellow represent HLA-E refolded with RL9HIV (RMYSPTSIL) (SEQ ID NO: 13), onto which the previously published HLA-E-RL9HIV structural coordinates (shaded pink) were aligned (Walters et al., 2018). Finally, the blue-shaded SEC-SAXS molecular envelope ‘dot’ models furthest to the right, represent HLA-E refolded with the RL9SIV (RMYSPTSIL) ‘supertope’. Although a crystal structure has not been determined for HLA-E-RL9SIV, the structural coordinates for HLA-E-RL9HIV were aligned to the SEC-SAXS molecular envelope for reference. B. Table detailing the HLA-E-restricted ‘intermediate’-binding pathogen peptides tested via SEC-SAXS including the VL9 (VMAPRTVLL) positive control peptide. Each row represents a SEC-SAXS experiment conducted in the absence or presence of 120mM excess peptide in the HPLC elution buffer. The radius of gyration (Rg) and maximum dimension (dmax), both measured in A, are specified. The volumes of SEC-SAXS molecular envelope ‘dot’ models (A3) displayed above in part A, are also denoted along with the % change in molecular envelope volume following the addition of excess peptide to the HPLC elution buffer. C. (i), (iii), (v), (vii), (ix) & (xi) Log 10 scattering intensity plots for HLA-E SEC-SAXS experiments. Plotted on the X-axis is the scattering vector, q, measured in A-l, which for small angles is proportional to the scattering angle Q. The scattered intensity, I(q), is plotted on the Y-axis with a log scale. Scattering intensity curves for HLA-E refolds run in the absence or presence of 120mM excess peptide in the HPLC elution buffer, are plotted together for reference. Peptide sequences are specified in the corresponding figure legend (ii), (iv), (vi), (viii), (x) & (xii) Normalised Kratky plots with superimposed curves corresponding to HLA-E SEC-SAXS experiments in the absence or presence of 120mM excess peptide. Modulated Gaussian curves are colour-coded according to figure legends in adjacent log 10 intensity plots. Plotted on the X-axis of the normalised Kratky plot is the scattering vector multiplied by the radius of gyration. On the Y-axis the scattering intensity I(q) is divided by the experiment’s 1(0) and multiplied by (q*Rg)2. The units on the X- and Y-axes are chosen such that the peak of the modulated Gaussian curve will always lie at q*Rg = 3 with a magnitude of 3-e-l, regardless of protein size and concentration when the Guinier’s approximation is obeyed
- this is true for globular and compact proteins. A shift from this peak thus signifies deviation from the Guinier’s approximation, indicating an increased degree of protein flexibility or conformational asymmetry.
Figure 3 - shows size exclusion chromatography-coupled small angle x-ray scattering for peptide-HLA-A2 complexes. A. (i) & (iii) Log 10 scattering intensity plots for HLA-A*02:01 SEC-SAXS experiments. Plotted on the X-axis is the scattering vector, q, measured in A-l, which for small angles is proportional to the scattering angle Q. The scattered intensity, I(q), is plotted on the Y-axis with a log scale. Scattering intensity curves for HLA-A2 refolds run in the absence or presence of 120mM excess peptide in the HPLC elution buffer, are plotted together for reference. Peptide sequences are specified in the corresponding figure legend (ii) & (iv) Normalised Kratky plots with superimposed curves corresponding to HLA-A2 SEC-SAXS experiments in the absence or presence of 120mM excess peptide. Modulated Gaussian curves are colour- coded according to figure legends in log 10 intensity plots. Plotted on the X-axis of the normalised Kratky plot is the scattering vector multiplied by the radius of gyration. On the Y-axis the scattering intensity I(q) is divided by the experiment’s 1(0) and multiplied by (q*Rg)2. The units on the X- and Y-axes are chosen such that the peak of the modulated Gaussian curve will always lie at q*Rg = 3 with a magnitude of 3-e-l, regardless of protein size and concentration when the Guinier’s approximation is obeyed
- this is true for globular and compact proteins. A shift from this peak thus signifies deviation from the Guinier’s approximation, indicating an increased degree of protein flexibility or conformational asymmetry. B. Table detailing the E1LA-A* 02: 01 -peptide refolds tested via SEC-SAXS. Each row represents a SEC-SAXS experiment conducted in the absence or presence of 120mM excess peptide in the HPLC elution buffer. The radius of gyration (Rg) and maximum dimension (dmax), both measured in A, are specified, along with previously reported melting temperatures (Tms) obtained by circular dichroism (Borbulevych et al., 2009).
Figure 4 - demonstrates that molecular envelope elongation negatively correlates with HLA-E thermal stability and peptide binding signals
A. Table detailing ELISA-based peptide binding signals, thermal stability measurements and size exclusion chromatography-coupled small angle x-ray scattering (SEC-SAXS) data for a panel of pathogen-derived HLA-E-restricted peptides. Further information including biological origin and corresponding references for each peptide are listed. A positive control VL9 leader peptide (VMAPRTVLL) in addition to a non binding negative control peptide derived from HIV Gag (QAISPRTLN) (SEQ ID NO: 14) are included. The ‘ELISA rank’ column reflects normalised and previously published peptide binding signals which are expressed as percentages of the positive control VL9 signal (Walters et al. 2020). HLA-E- 2M-peptide complex thermal stability is indicated by its melting temperature (°C) in the presence of 12M excess peptide. The SEC-SAXS-obtained dmax value, measured in A, reflects the maximum dimension across the HLA-E complex in solution whereas the ‘Molecular Envelope VoT corresponds to the volume in A3 of the DAMMINF ab initio molecular envelope models presented in Figure 2. B. (i) - (iv) Scatter graphs where normalised ELISA-based HLA- E peptide binding signals or melting temperatures, denoted in A, are plotted on the x- axes with SEC-SAXS-obtained dmax values or molecular envelope volumes, denoted in A, on the Y-axes. Pearson’s correlation coefficients (r) relating HLA-E peptide binding/thermal stability to SEC-SAXS-obtained indicators of HLA-E complex global dimensionality are denoted with corresponding ‘p’ values (v) Scatter graph with normalised ELISA-based HLA-E peptide binding signals and melting temperatures, denoted in A, plotted on the x and y axes, respectively. The Pearson’s correlation coefficient (r) is denoted with the corresponding ‘p’ value or ‘ns’ for non-significant correlations (vi) Figure legend for B (i) - (v). C. (i) - (vii) Bar charts representing ELISA-based HLA-E peptide titration assays in which the peptide concentration plotted on the X-axis was varied from 7.5mM to 1.2mM during the peptide exchange reaction prior to sampling by sandwich ELISA. ELISA-based average absorbance signals at 450nm are plotted on the Y-axis and the test peptide ID and sequence is denoted above each individual plot. The ‘no rescue’ negative control shaded in dark grey reflects a peptide exchange reaction conducted in the absence of excess test peptide. For (ii) to (vii), the positive control VL9 leader peptide (VMAPRTVLL), shaded dark red, corresponds to a peptide exchange reaction of 120mM excess peptide. Pearson’s correlation coefficients (r) for peptide concentration versus ELISA-based average absorbance signals are denoted with corresponding ‘p’ values or ‘ns’ for non-significant correlations. Error bars depict standard error of the mean. Biological repeats n=3, technical replica n=2.
Figure 5 - shows structural characterisation of Mycobacterial peptide binding to HLA-E A. (i) A side-on PyMol visualisation of the Mycobacterial peptide, Mtbl4 (RMAATAQVL), from molecule 1 of the asymmetric unit shown in hot pink stick-form with the HLA-E peptide binding groove omitted for clarity. Peptide residues are labelled and an electron density map contoured to 1 sigma is displayed in grey mesh overlaying the peptide (ii) Ribbon representation of the Mtbl4 (RMAATAQVL) peptide backbone shaded in hot pink from the HLA-E-Mtbl4 structure determined in this study with a superimposed VL9 (VMAPRTVLL) leader peptide backbone from the previously published HLA-E-VL9 structure, 1MHE, in violet. The HLA-E binding groove is omitted and the peptide N and C termini labelled for clarity. B (i) A side-on PyMol visualisation of the Mycobacterial peptide, IL9 (IMYNYPAML), from molecule 1 of the asymmetric unit shown in blue stick-form with the HLA-E peptide binding groove omitted for clarity (ii) A side-on PyMol visualisation of the Mycobacterial peptide, IL9 (IMYNYPAML), from molecule 2 of the asymmetric unit shown in blue stick-form with the HLA-E peptide binding groove omitted for clarity. In B (i) and (ii) peptide residues are labelled and an electron density map contoured to 1 sigma is displayed in grey mesh overlaying the peptide (iii) Superimposed peptide backbones are depicted with the HLA-E peptide binding groove omitted and N and C termini labelled for clarity. The IL9 (IMYNYPAML) peptide backbones from molecules 1 and 2 of the asymmetric unit are displayed in blue ribbon with the position 7 Ala side chains in blue stick-form. The distance (1.9 A) separating the position 7 Ca atom of the IL9 peptides from molecules 1 and 2 of the asymmetric unit is denoted (iv) Superimposed peptide backbones are depicted in ribbon-form with the HLA-E peptide binding groove omitted and N and C termini labelled for clarity. The IL9 (IMYNYPAML) peptide backbones from molecules 1 and 2 of the asymmetric unit are displayed in blue. The VL9 (VMAPRTVLL) leader peptide backbone from the previously published HLA-E-VL9 structure, 1MHE, is also displayed and shaded violet. The distance (2.0 A) separating the position 7 Ca atom of the IL9 peptide from molecule 1 of the asymmetric unit and the position 7 Ca atom of VL9 is denoted. Figure 6 - shows that suboptimal E pocket-based interactions compromise HLA-E complex stability A. (i) PyMol visualisation of the HLA-A2 E pocket (PDB ID: 3MRG, Reiser et al. 2014) with side chains of pocket-forming residues depicted as grey or green sticks (Tyr-116, Trp-133, Trp-147, Val-152, Leu-156). The associated Hepatitis C virus-derived nonamer (CINGVCWTV)(SEQ ID NO: 15) is shown in pink ribbon with the solvent-exposed position 7 side chain in stick-form projecting away from the E pocket. Visible sections of the HLA-A2 peptide binding groove including the a2 helix and b-sheet floor are shown in grey cartoon (ii) - (v) PyMol visualisation of the peptide position 7 anchor side chain-accommodating E pocket of HLA-E. Side chains of secondary E pocket-forming residues are depicted as grey sticks (Phe-116, Trp-133, Ser-147, Glu-152, Gln-156) with remaining visible regions of the peptide binding groove shown in grey cartoon. In (i) the VL9 (VMAPRTVLL) peptide mainchain is displayed as purple ribbon with the position 7 side chain in purple stick-form projecting downward into the secondary E pocket (PDB ID: 1MHE, O’Callaghan et al. 1998). (iii) - (v) illustrate differential positioning at position 7 of HLA-E-bound pathogen-derived peptides relative to VL9 from 1MHE. The Mtb-derived IL9 (IMYNYPAML), HIV Gag- derived RL9HIV (RMYSPTSIL) and Mtb-derived Mtbl4 (RMAATAQVL) peptide backbones are depicted as blue (iii), yellow (iv) and magenta (v) ribbons, respectively. The position 7 side chains - Ala-7 of IL9, Ser-7 of RL9HIV and Gln-7 of Mtbl4 - are shown in blue (iii), yellow (iv) and magenta (v) stick-form, respectively. The superimposed VL9 peptide main chain from 1MHE is shown in purple ribbon and the distance between the aligned peptide position 7 Ca atoms is indicated by dashed lines - 2 A separates the position 7 Ca atoms of VL9 and IL9, 3.4 A separates the position 7 Ca atoms of VL9 and RL9HIV and 1.1 A separates the Ca atoms of VL9 and Mtbl4. In (v), water-mediated intra- inter-chain hydrogen bonds are depicted as magenta dashed lines with the coordinated H20 molecule visualised as a cyan-shaded sphere. A total of 4 hydrogen bonds are shown which indirectly link the Mtbl4 position 7 Gin side chain to the Ser-147 side chain and Ser-143 main chain of the HLA-E a2 -helix in addition to the position 8 main chain of the Mtbl4 peptide. The Val-8 main chain of the Mtbl4 peptide and Ser-143 main chain of HLA-E are labelled ‘MC’ and depicted as magenta and grey sticks, respectively. B. Table of HLA-E- 2M-peptide thermal stability as indicated by melting temperature (Tm) in the presence of 12 M excess peptide measured in °C. Tm values are listed for the canonical HLA-B7-derived VL9 leader peptide (VMAPRTVLL) and the Mtb-derived Mtbl4 peptide (RMAATAQVL) in addition to corresponding position 7 variant peptides in which a non-canonical polar Gin was introduced at position 7 in VL9 (VMAPRTQVL) (SEQ ID NO: 16) or a canonical hydrophobic Val was introduced at position 7 in Mtbl4 (RMAATAVVL) (SEQ ID NO: 17). C. (i) Bar chart of an ELISA-based HLA-E peptide binding assay in which the wild- type Mtb-derived Mtbl4 peptide (RMAATAQVL) was compared to an Mtbl4 variant peptide containing a canonical position 7 Val. The VL9 leader peptide (VMAPRTVLL), shaded dark red, was included for reference as a positive control whereas the negative control, shaded grey, comprised a peptide-free ‘no rescue’ peptide exchange reaction prior to sandwich ELISA sampling. Test peptides are plotted on the X-axis, average absorbance readings at 450nm on the Y-axis and p values from unpaired t-tests are denoted above the corresponding bars - no star=P>0.05, *=P<0.05, **=P<0.01, ***=P<0.001, ****P=<0.0001. Error bars depict standard error of the mean. Biological repeats n=3, technical replica n=2. (ii) Bar chart as described in C (i), in which HLA- E binding capacity of the wild-type VL9 leader peptide (VMAPRTVLL) was compared to a variant peptide containing a non-canonical polar position 7 Gin. The Mtb-derived Mtbl4 peptide (RMAATAQVL) which also contains a non-canonical position 7 Gin was included for reference. D. (i) & (iii) Log 10 scattering intensity plots for HLA-E SEC- SAXS experiments. Plotted on the X-axis is the scattering vector, q, measured in Ά-1, which for small angles is proportional to the scattering angle Q. The scattered intensity, I(q), is plotted on the Y-axis with a log scale. Superimposed scattering intensity curves for peptide-HLA-E refolds are colour-coded according to the corresponding figure legend (ii) & (iv) Normalised Kratky plots with superimposed curves from HLA-E SEC-SAXS experiments. Superimposed modulated Gaussian curves are colour-coded according to figure legends in log 10 intensity plots. Plotted on the X-axis of the normalised Kratky plot is the scattering vector multiplied by the radius of gyration. On the Y-axis, the scattering intensity, I(q), is divided by the experiment’s 1(0) and multiplied by (q*Rg)2. The units on the X- and Y-axes are chosen such that the peak of the modulated Gaussian curve will always lie at q*Rg = 3 with a magnitude of 3-e-l, regardless of protein size and concentration when the Guinier’s approximation is obeyed - this is true for globular and compact proteins. A shift from this peak thus signifies deviation from the Guinier’s approximation, indicating an increased degree of protein flexibility or conformational asymmetry. E. Table detailing HLA-E-peptide refolds tested via SEC-SAXS. Peptide ID, sequence and origin are specified along with the radius of gyration (Rg) and maximum dimension (dmax), both measured in A. Figure 7 - demonstrates that distinct that structural motifs emerge in the absence of HLA-E-associated VL9 leader peptide A. (i) PyMol visualisation of superimposed HLA-E a2 helical kink regions depicted as lines with the short-arm of the a2 helix labelled ‘SA’, the long-arm labelled ‘LA’ and residue positions denoted. The a2 helix from the VL9 (VMAPRTVLL)-associated HLA-E structure, 1MHE, is shaded grey. The a2 helices from superimposed pathogen peptide-associated HLA-E molecules are depicted including HLA-E-Mtb44 (RLPAKAPLL) in green, HLA-E-RL9HIV (RMYSPTSIL) in yellow, HLA-E-Mtbl4 (RMAATAQVL) in magenta and HLA-E-IL9 (IMYNYPAML) in blue (ii) The maximum distance in A separating Ca atoms of a2 helix residues in the HLA-E-VL9 structure, 1MHE, versus corresponding Ca atoms in pathogen peptide-associated HLA-E structures. B. Ca backbones of superimposed HLA- E-associated peptides are depicted as ribbons. The canonical VL9 leader peptide (VMAPRTVLL) peptide from 1MHE is shaded grey. Pathogen-derived HLA-E- associated peptides are coloured green (Mtb44 - RLPAKAPLL), yellow (RL9HIV - RMYSPTSIL), magenta (Mtbl4 - RMAATAQVL) and blue (IL9 - IMYNYPAML). The position of the HLA-E al and a2 helices are indicated in addition to the N and C peptide termini. Position 5 Ca atoms are circled with the maximum distance separating the VL9 peptide Arg-5 Ca from pathogen-derived peptide position 5 Ca atoms, denoted. 2.3 A separates the position 5 Ca atoms of VL9 and RL9HIV, 1.9 A separates the position 5 Ca atoms of VL9 and IL9, 1.9 A separates the Ca atoms of V9 and Mtbl4 and 1.5 A separates the position 5 Ca atoms of VL9 and Mtb44. C. (i) The HLA-E a2 helix is shown in grey cartoon with superimposed pathogen-derived peptide backbones, RL9HIV (RMYSPTSIL), IL9 (IMYNYPAML) and Mtbl4 (RMAATAQVL) in yellow, blue and magenta ribbon, respectively. The superimposed VL9 leader peptide (VMAPRTVLL) peptide backbone from 1MHE is shown in grey ribbon. Ser-147, Glu- 152 and Gln-156 side chains of the HLA-E a2 helix are shown in stick-form and are colour-coded according the corresponding peptide. The 2 salt bridges connecting Glu- 152 and the Arg-5 side chain of the VL9 peptide are shown as grey dashed lines. A hydrogen bond between the Gln-156 side chain and the main chain Oxygen of VL9 Arg- 5 is also depicted as grey dashed lines (ii) The HLA-E a2 helix is shown in grey cartoon with the Mtb-derived peptide Mtb44 (RLPAKAPLL) peptide backbone (green) superimposed to that of VL9 (VMAPRTVLL) from 1MHE (grey). Ser-147, Glu-152 and Gln-156 side chains of the HLA-E a2 helix are shown in stick-form and are colour- coded according the corresponding peptide. In this PyMol visualisation, molecule 3 of the HLA-E-Mtb44 structure is shown, in which a hydrogen bond (green dashed lines) connects the HLA-E Gin- 156 side chain to the main chain Oxygen of Mtb44 Lys-5, which is depicted as green sticks. Note that this hydrogen bond is only present in 1 of the 4 molecules present in the asymmetric unit of the HLA-E-Mtb44 structure. D. The HLA-E a2 helix is shown in grey cartoon with the superimposed pathogen-derived peptides, RL9HIV (RMYSPTSIL) and IL9 (IMYNYPAML), in yellow and blue ribbon, respectively. The Tyr-3 side chain of RL9HIV and IL9 is shown in yellow and blue stick-form, respectively. HLA-E a2 helix Glu-152 side chains from the HLA-E-RL9HIV and HLA-E-IL9 structures are shown as yellow and blue sticks, respectively, with corresponding hydrogen bonds depicted as yellow/blue dashed lines. The Glu-152 side chain from the a2 helix of HLA-E-VL9 (VMAPRTVLL) in 1MHE is shown in grey stick-form for reference. E. Table detailing 7 previously published non-receptor-bound HLA-E structures in addition to the 2 novel HLA-E structures presented in this study. HLA-E-associated peptide IDs, organisms of origin and amino acid sequences are specified along with the corresponding HLA-E allelic variant, PDB accession code (AC) and reference.
Figure 8 - Demonstrates thermal gain of HLA-EC84-C139 over canonical HLA-E when incubated with 100M excess peptide. IOUM of pre-refolded HLA-E and HLA- EC84-C139 material was incubated with 100M excess test peptides (P1-P9) for 30 minutes at room temperature prior to thermal melt analysis using a Prometheus NT.48 Series Differential Scanning Fluorimetry instrument. Test samples were split between two Prometheus NT.48 Series nanoDSF Grade Standard Capillaries and a ramp rate of 1 °C/min from 20 °C to 95 °C was applied. The ratio for fluorescence emission at 330 nm and 350 nm was used to derive the thermal melt of unfolding (Tm). Shown are the relative Tm data for canonical HLA-E (left column) and HLA-EC84-C139 (right column) datasets, where the corresponding no-peptide control Tm data for canonical HLA-E and HLA-EC84-C139 have been subtracted, respectively. The numbers plotted above the red bars denote the equivalent Tm loss/gains obtained for the HLA-EC84- C139 variant over canonical HLA-E.
Figure 9 - Demonstrates thermal gain of HLA-EW147over canonical HLA-E when incubated with 100M excess peptide. IOuM of pre-refolded HLA-E or HLA- EW147material was incubated with 100M excess test peptides (P1-P9) for 30 minutes at room temperature prior to thermal melt analysis using a Prometheus NT.48 Series Differential Scanning Fluorimetry instrument. Test samples were split between two Prometheus NT.48 Series nanoDSF Grade Standard Capillaries and a ramp rate of 1 °C/min from 20 °C to 95 °C was applied. The ratio for fluorescence emission at 330 nm and 350 nm was used to derive the thermal melt of unfolding (Tm). Shown are the relative control Tm gains for canonical HLA-E (left column) and HLA-EW147 (right column) datasets, where the corresponding no-peptide control Tm data for canonical HLA-E and HLA-EW147 have been subtracted, respectively. The numbers plotted above the red bars denote the equivalent Tm loss/gains obtained for the HLA- EW147variant over canonical HLA-E.
Figure 10 - demonstrates that pathogen-derived epitopes drive suboptimal HLA-E complex formation, despite previously reported immunogenicity in vivo A. (i) Graph showing the radius of gyration (Rg) in A plotted as red dots across the SAXS curve, which corresponds to the HPLC elution trace, for VL9 (VMAPRTVLL)-refolded HLA- E* 01:03. The x-axis denotes HPLC-eluted x-ray exposed frames across the HLA-E-VL9 protein peak, whereas the Y-axis corresponds to the SAXS scattering intensity signal (ii) Ab initio molecular envelope model displayed as small purple ‘dots’ representing the average conformational state of the leading peak fraction of HPLC-eluted HLA-E refolded with the canonical VL9 leader peptide. This molecular envelope ‘dot’ model corresponds to a SEC-SAXS run where an HLA-E*01:03 refold was injected onto the HPLC column in the presence of 120mM excess VL9 peptide but no excess peptide was added to the HPLC elution buffer. The previously published structural coordinates for HLA-E-VL9 (VMAPRTVLL) (PDB AC: 1MHE) are aligned to the SAXS-generated molecular envelope and displayed as green cartoon (iii) Same as A. (ii), but for tailing peak fraction (iv) LoglO scattering intensity plot for HLA-E SEC-SAXS. Plotted on the X-axis is the scattering vector, q, measured in Ά-1, which for small angles is proportional to the scattering angle Q. The scattered intensity, I(q), is plotted on the Y- axis with a log scale. Scattering intensity curves for HLA-E refolds run in the absence of 120mM excess peptide in the HPLC elution buffer are plotted for the leading and tailing peak fractions of VL9-refolded HLA-E and are colour-coded according to the corresponding figure legend (v) Normalised Kratky plot with superimposed curves corresponding to SEC-SAXS leading and tailing peak fractions for VL9-refolded HLA- E in the absence of 120mM excess peptide in the HPLC elution buffer, colour-coded according to the figure legend in A. (iv). Plotted on the X-axis is the scattering vector multiplied by the radius of gyration. On the Y-axis the scattering intensity I(q) is divided by the experiment’s 1(0) and multiplied by (q*Rg)2. The units on the X- and Y- axes are chosen such that the peak of the modulated Gaussian curve will always lie at q*Rg = 3 with a magnitude of 3 e-l, regardless of protein size and concentration when the Guinier’s approximation is obeyed - this is true for globular and compact proteins. A shift from this peak thus signifies deviation from the Guinier’s approximation, indicating an increased degree of protein flexibility or conformational asymmetry (vi) Table detailing the HLA-E-VL9 refold tested via SEC-SAXS described in A (i)-(v). The radius of gyration (Rg) and maximum dimension (dmax), both measured in A, are specified. Volumes of SEC-SAXS molecular envelope ‘dot’ models (A3) displayed in A (ii) and (iii), are also denoted along with the % change in molecular envelope volume from the leading to tailing peak fraction.
B. (i) Graph showing the radius of gyration (Rg) in A plotted as red dots across the SAXS curve, which corresponds to the HPLC elution trace, for IL9 (IMYNYPAML)- refolded HLA-E*01:03. The x-axis denotes HPLC-eluted x-ray exposed frames across the HLA-E-IL9 protein peak, whereas the Y-axis corresponds to the SAXS scattering intensity signal (ii) Ab initio molecular envelope model displayed as small blue ‘dots’ representing the average conformational state of the leading peak fraction of HPLC- eluted HLA-E refolded with the IL9 peptide. This molecular envelope ‘dot’ model corresponds to a SEC-SAXS run where an HLA-E*01:03 refold was injected onto the HPLC column in the presence of 120mM excess IL9 peptide but no excess peptide was added to the HPLC elution buffer. The structural coordinates for HLA-E-IL9 are aligned to the SAXS-generated molecular envelope and displayed in orange cartoon-form (iii) Same as B. (ii), but for tailing peak fraction (iv) Log 10 scattering intensity plot for HLA-E SEC-SAXS. Plotted on the X-axis is the scattering vector, q, measured in A-l, which for small angles is proportional to the scattering angle Q. The scattered intensity, I(q), is plotted on the Y-axis with a log scale. Scattering intensity curves for HLA-E refolds run in the absence of 120mM excess peptide in the HPLC elution buffer are plotted for the leading and tailing peak fractions of IL9-refolded HLA-E and are colour- coded according to the corresponding figure legend (v) Normalised Kratky plot with superimposed curves corresponding to SEC-SAXS leading and tailing peak fractions for IL9-refolded HLA-E in the absence of 120mM excess peptide in the HPLC elution buffer, colour-coded according to the figure legend in B. (iv). Plotted on the X-axis is the scattering vector multiplied by the radius of gyration. On the Y-axis the scattering intensity I(q) is divided by the experiment’s 1(0) and multiplied by (q*Rg)2. (vi) Table detailing the HLA-E-IL9 refold tested via SEC-SAXS described in B (i)-(v). The radius of gyration (Rg) and maximum dimension (dmax), both measured in A, are specified. Volumes of SEC-SAXS molecular envelope ‘dot’ models (A3) displayed in B (ii) and (iii), are also denoted along with the % change in molecular envelope volume from the leading to tailing peak fraction.
C. (i) Graph showing the radius of gyration (Rg) in A plotted as red dots across the SAXS curve, which corresponds to the HPLC elution trace, for RL9HIV (RMYSPTSIL)- refolded HLA-E*01:03. The x-axis denotes HPLC-eluted x-ray exposed frames across the HLA-E-RL9HIV protein peak, whereas the Y-axis corresponds to the SAXS scattering intensity signal (ii) Ab initio molecular envelope model displayed as small yellow ‘dots’ representing the average conformational state of the leading peak fraction of HPLC-eluted HLA-E refolded with RL9HIV peptide. This molecular envelope ‘dot’ model corresponds to a SEC-SAXS run where an HLA-E*01:03 refold was injected onto the HPLC column in the presence of 120mM excess RL9HIV peptide but no excess peptide was added to the HPLC elution buffer. The previously published structural coordinates for HLA-E-RL9HIV (PDB AC: 6GL1) are aligned to the SAXS-generated molecular envelope and displayed in pink cartoon-form (iii) Same as C. (ii), but for tailing peak fraction (iv) LoglO scattering intensity plot for HLA-E SEC-SAXS. Plotted on the X-axis is the scattering vector, q, measured in A-l, which for small angles is proportional to the scattering angle Q. The scattered intensity, I(q), is plotted on the Y- axis with a log scale. Scattering intensity curves for HLA-E refolds run in the absence of 120mM excess peptide in the HPLC elution buffer are plotted for the leading and tailing peak fractions of RL9HIV-refolded HLA-E and are colour-coded according to the corresponding figure legend (v) Normalised Kratky plot with superimposed curves corresponding to SEC-SAXS leading and tailing peak fractions for RL9HIV-refolded HLA-E in the absence of 120mM excess peptide in the HPLC elution buffer, colour- coded according to the figure legend in C. (iv). Plotted on the X-axis is the scattering vector multiplied by the radius of gyration. On the Y-axis the scattering intensity I(q) is divided by the experiment’s 1(0) and multiplied by (q*Rg)2. (vi) Table detailing the HLA-E-RL9HIV refold tested via SEC-SAXS described in C (i)-(v). The radius of gyration (Rg) and maximum dimension (dmax), both measured in A, are specified. Volumes of SEC-SAXS molecular envelope ‘dot’ models (A3) displayed in C (ii) and (iii), are also denoted along with the % change in molecular envelope volume from the leading to tailing peak fractions. Figure 11 - Thermal gain of HLA-EH99Y, HLA-Ef116Y and HLA-ES147W over canonical HLA-E when incubated with 10M excess peptide lOuM of pre-refolded HLA-E and HLA-EH99Y (A), HLA-EF116Y (B), or HLA-ES147W(C) material was incubated with 10M excess test peptides (from panel pA to pG) for 30 minutes at room temperature prior to thermal melt analysis using a Prometheus NT.48 Series Differential Scanning Fluorimetry instrument. Test samples were split between two Prometheus NT.48 Series nanoDSF Grade Standard Capillaries and a ramp rate of 1 °C/min from 20 °C to 95 °C was applied. The ratio for fluorescence emission at 330 nm and 350 nm was used to derive the thermal melt of unfolding (Tm). Shown are the relative control Tm gains for canonical HLA-E (left columns) and HLA-EH99Y, HLA- Ef116Y or HLA-ES147W (right columns) datasets, where the corresponding no-peptide control Tm data for canonical HLA-E and HLA-EH99Y, HLA-Ef116Y or HLA-ES147W have been subtracted, respectively. The numbers plotted above the right-hand columns denote the equivalent Tm gains obtained for the HLA-EH99Y, HLA-Ef116Y orHLA-Esl47W variants over canonical HLA-E, respectively.
Figure 12 - Thermal stability and Blue Native gel analysis of canonical HLA-E and versus single, double and triple HLA-E mutants refolded with RL9HIV peptide.
Canonical HLA-E, pre-refolded with VL9 peptide (VMAPRTVLL) and RL9HIV (RMYSPTSIL), was compared to single (HLA-EH99Y, HLA-Ef116Y, HLA-Es147W), double (HLA-EH99Y+f116Y, HLA-EH99Y+s147W ) and one triple (HLA-EH99Y+F116Y+S147W) HLA-E mutant pre-refolded with RL9HIV peptide. For thermal melt analysis (i), 20m1 of 0.5 mg/ml individual HLA-E samples were split between two Prometheus NT.48 Series nanoDSF Grade Standard Capillaries and placed in the sample holder of a Prometheus Panta Series Differential Scanning Fluorimetry instrument. A ramp rate of 1 °C/min from 20 °C to 95 °C was applied. The ratio for fluorescence emission at 330 nm and 350 nm was used to derive the thermal melt of unfolding (Tm), which are reported (above individual bars) for the VL9 pre-refolded material (black bar) and the RL9HIV pre-refolded samples (canonical and mutant - all hashed bar). For Blue Native gel analysis (ii), 12pL of 0.5 mg/ml of HLA-E material was loaded into a NativePAGE 4-16% Bis-Tris 1.0mm x 10 well gel. Gel electrophoresis was performed at 150 volts for 2 hours at room temperature (RT), over a current range of 15-16 to 2-4mAmps. Gels were subsequently rinsed in MilliQ water and stained for 2-3 hours, at RT, in SimplyBlue™ SafeStain. Gel de-staining was performed in MilliQ water over a period of 24 hours prior to imaging. The NativeMark™ protein standard 66kDa band is denoted for reference. The arrows denote both Compact (Cf) and Diffuse (Df) HLA-E- peptide gel forms. The identity of lane samples 1-8 are provided in the accompanying Table. EXAMPLES
Example 1 - Distinct blue native gel signatures for VL9- versus pathogen epitope- associated HLA-E
A panel of pathogen-derived HLA-E-restricted epitopes which demonstrated HLA-E binding capacity above 20% of the positive control VL9 peptide in a previously published ELISA-based peptide binding assay (Walters et al., EJI, 2020), were evaluated using a previously established blue native gel system (Walters et al., Nat. Comm, 2018;9(1):3137). Peptide exchange reactions were conducted using HLA-E complexes pre-refolded with a UV-sensitive VL9 variant (VMAPJTVLL) in the presence of molar excess test peptide prior to sampling by blue native gel. Differential gel signatures were observed for the positive control VL9 leader peptide relative to the pathogen-derived peptides IL9, Mtbl4, RL9HIV, BZLF1 and RL9SIV. Where VL9- HLA-E generated a single compact band on the gel, HLA-E incubated in the presence of pathogen-derived peptide resulted in the emergence of a mainly diffuse gel band. Mtb44, the only pathogen-derived peptide to demonstrate a comparable HLA-E binding capacity to VL9 (97%) in previous ELISA-based screens, exclusively generated a compact gel band, similar to VL9.
Example 2 - SEC-SAXS reveals striking peptide-dependent differences in the HLA- E conformational ensemble Given that multiple distinct gel band signatures were identified in BNG analyses indicative of sample non-uniformity, size exclusion chromatography-coupled small angle x-ray scattering (SEC-SAXS) was used to further probe potential discrepancies in the protein conformational ensemble for canonical VL9 versus pathogen peptide- associated HLA-E. Individual HLA-E*01:03 protein refolds were assembled for each HLA-E-restricted pathogen-derived peptide present in the aforementioned ‘intermediate’ binding panel in addition to the positive control VL9 peptide, VMAPRTVLL. Scattering data from X-ray-exposed protein samples were collected during high-performance liquid chromatography (HPLC) elution, from which the radius of gyration (Rg) and maximum dimension (dmax) of protein complexes in solution were calculated, following subtraction of buffer-induced scattering. DAMMIF ab initio molecular envelope models were also generated for peptide-bound HLA-E which represent the average conformer in solution (Franke and Svergun, 2009) (Volkov and Svergun, 2003).
The resolution limits of SAXS are inferior relative to those of x-ray crystallography making it challenging to decipher local, fine-tuned structural movement with SAXS data alone. However, unlike the static, lattice-restrained snapshot afforded by x-ray crystallography, SAXS is capable of detecting dynamic protein folded states and large conformational adjustments of proteins in solution (Kikhney and Svergun, 2015). Surprisingly, pronounced and reproducible differences were observed of up to 68 A and 82,764 A3 in global dimensionality between canonical VL9-associated HLA-E versus HLA-E refolded with a number of ‘intermediate’ binding pathogen-derived epitopes including RL9HIV (RMYSPTSIL), RL9SIV (RMYNPTNIL), IL9 (IMYNYPAML), Mtbl4 (RMAATAQVL) and BZLF1 (SQAPLPCVL) (SEQ ID NO: 18) (Figure 2, A & B). Superposition of HLA-E crystallographic coordinates to SEC-SAXS-obtained DAMMIF molecular envelope models revealed strong alignment and similar global dimensions for VL9-refolded HLA-E (O’Callaghan et al., 1998) (Figure 2, A). However, a strikingly poor fit to the crystallographic coordinates was observed for molecular envelopes corresponding to HLA-E refolded with the pathogen-derived peptides, RL9HIV, RL9SIV, IL9, Mtbl4 and BZLF1. Consistent with higher dmax values, large protrusions in the molecular envelopes above and below the superimposed crystallographic coordinates were observed for HLA-E complexes refolded with these pathogen epitopes, indicative of elongated and partially unfolded protein species in solution. This finding was somewhat unexpected considering crystal structures of HLA- E in complex with RL9HIV, IL9 and Mtbl4 exhibited comparable global dimensions and superposed to structures of HLA-E-VL9.
The panel of pathogen-derived epitopes tested in SEC-SAXS analyses were among the strongest HLA-binding peptides identified in an extensive ELISA-based screening (Walters, McMichael and Gillespie, 2020). In addition to generating comparable ELISA-based peptide binding signals and blue native gel signatures to the positive control VL9 peptide, Mtb44 (RLPAKAPLL) also generated a more compact molecular envelope and comparable Rg and Dmax values to VL9-refolded HLA-E in SEC-SAXS analyses (Figure 2). Strikingly elongated molecular envelopes and higher Dmax values relative to VL9-refolded HLA-E were observed for Mtbl4 (RMAATAQVL), BZLF1 (SQAPLPCVL), RL9HIV (RMYAPTSIL) (SEQ ID NO: 19) and the immunodominant RL9SIV (RMYNPTNIL), indicating that these epitopes do not support the formation of uniformly compact, fully folded HLA-E complexes in solution. Further, a number of highly immunogenic HLA-E-restricted epitopes (Hansen et al., 2016; Walters et al., EJI, 2020) that exhibited weak or apparent non-binding in the ELISA-based method were also tested via SAXS, including the immunodominant SIV epitope, EK9 (EKQRESREK) (SEQ ID NO: 20). However, proportionally larger HPLC-eluted aggregate peaks and smaller protein peaks, owing to complex instability, yielded insufficient x-ray scattering for reliable downstream data processing.
SEC-SAXS analyses also revealed conformational homogeneity for VL9 and Mtb44 HLA-E refolds, as evidenced by similar Rg values across x-ray-exposed frames of the HPLC-eluted protein peaks (Figure 10). However, non-constant Rg values which increased across x-ray-exposed frames of the eluted protein peak were observed for HLA-E refolded with the pathogen-derived peptides, RL9HIV, RL9SIV, Mtbl4, IL9 and BZLF1, reflecting conformational heterogeneity among eluted HLA-E complexes. Accordingly, stratification of SEC-SAXS HPLC peaks into leading and tailing fractions confirmed sample heterogeneity as indicated by discrepancies in molecular envelope volumes and Rg or dmax values between peak fractions (Figure 10). Thus, these pathogen-derived epitopes did not afford adequate binding energy to drive conformationally homogeneous populations of stable, fully-folded HLA-E complexes in solution. Together, the heterogeneous conformational protein ensembles and elongated molecular envelopes imply that the majority of ‘intermediate’ strength HLA- E-restricted pathogen-derived epitopes drive suboptimal HLA-E complex formation, despite previously reported immunogenicity in vivo.
To evaluate whether this phenomenon was related to peptide binding affinity, SEC- SAXS was performed in the constant presence of excess peptide. The addition of molar excess peptide to the HPLC elution buffer transformed elongated molecular envelopes of pathogen epitope-refolded HLA-E into compact forms that resembled VL9-refolded HLA-E (Figure 2, A). Molecular envelope models for RL9HIV-, RL9SIV-, IL9- and Mtbl4-HLA-E injected and eluted in the presence of excess peptide exhibited improved alignment to previously published superimposed HLA-E crystal structures relative to the corresponding elongated forms generated in the absence of excess peptide in the elution buffer. Additionally, superimposed log 10 intensity plots which show the intensity of scattered x-rays over the small angle range, revealed differentially shaped curves in the absence or presence of excess peptide for Mtbl4, RL9HIV or RL9SIV (Figure 2, C). Dimensionless Kratky plots normalised for particle mass and concentration also demonstrated conformational discrepancies between protein populations in the presence versus absence of excess peptide; the peak of the modulated Gaussian curve shifted toward 3 (x-axis) with a magnitude of 3-e-l (y-axis) in the presence of excess Mtbl4, RL9HIV, RL9SIV or IL9 peptide, consistent with a more compact, globular protein state. By contrast, deviations from this peak maxima in the absence of excess peptide indicate a greater degree of particle flexibility or asymmetry. Unsurprisingly, the higher affinity peptides including VL9 and Mtb44 were less impacted by the addition of excess peptide, as evidenced by stronger alignment in the superimposed scattering intensity curves and Kratky plots (Figure 2, C, i-iv). In summary, throughout SEC-SAXS analyses, all pathogen epitope-HLA-E samples were injected onto the HPLC column in the presence of excess peptide and gave a distinct elution peak. However, for RL9HIV-, Mtbl4-, RL9SIV- and IL9-HLA-E, compact molecular envelopes were only generated when molar excess peptide was also present throughout the HPLC injection and elution process implicating weak peptide binding affinity and fast dissociation rates as key contributing factors to molecular envelope elongation.
For comparative purposes, SEC-SAXS experiments were also performed with HLA-A2 refolded with the well characterised Tellp (MLWGYLQYV)(SEQ ID NO: 21) and Tax (LLFGYPVYV)(SEQ ID NO: 22) epitopes derived from S.cerevisiae and HTLV-1, respectively (Figure 3). Similar to VL9-refolded HLA-E, the molecular envelope volumes, shape of the scattering intensity curves and positioning of the modulated Gaussian peak maxima on the dimensionless Kratky plots were unaffected by the addition of excess peptide. Further, the Rg and dmax values were comparable for Tellp- or Tax-refolded HLA-A2 and VL9-refolded HLA-E (Figure 3). Normalised Kratky plots revealed that HLA-A2 particles obey the Guinier’s approximation with peaks at 3 with a magnitude of 3-e-l, indicative of compact fully-folded globular protein (Figure 3, B, ii & iv). Example 3 - Correlations between SAXS data, HLA-E peptide binding signals and melting temperatures
DSF-generated melting temperatures (Tm) were obtained for previously purified peptide-free-HLA-E complexes incubated with HLA-E-restricted peptides from the ‘intermediate’ binding panel in addition to a positive (VL9, VMAPRTVLL) and negative (HIV Gag, QAISPRTLN) control peptide at a 12M excess peptide:protein ratio (Figure 4, A). The negative control peptide (HIV Gag, QAISPRTLN) generated a comparable Tm (31.8°C) to the peptide-free-HLA-E background (32.0°C) whereas Mtb44 (RLPAKAPLL) generated a comparable Tm (50.6°C) to the positive control VL9-incubated HLA-E (49.4°C). The remaining ‘intermediate’ binding peptides generated Tm values ranging from 35.2°C (RL9SIV, RMYNPTNIL) to 40.7°C (IL9, IMYNYPAML). A strikingly strong positive correlation (r = 0.98) was observed between HLA-E complex melting temperatures and previously obtained %VL9 binding scores from ELISA-based screens, highlighting the strong agreement between the two independent techniques (Figure 4, B). Considering immunogenic classical MHC class I-restricted peptide epitopes typically exhibit melting temperatures between 52 and 65°C, these ‘intermediate’ binding HLA-E-restricted pathogen epitopes sample an unusually low space on the pMHC thermal stability continuum. Peptide titration assays were also conducted using the previously optimised ELISA- based HLA-E peptide screening system (Figure 4, C) (Walters, McMichael and Gillespie, 2020). Weak correlations were identified between peptide concentration and ELISA signals for the two strongest binding peptides, VL9 (r = 0.43) and Mtb44 (r = 0.04) highlighting the capacity of these peptides to support HLA-E complex formation at low concentrations (Figure 4, C, i & ii). Stronger positive correlations were observed for the remaining ‘intermediate’ pathogen-derived peptides, including IL9 - IMYNYPAML (r = 0.59), Mtbl4 - RMAATAQVL (r = 0.71), RL9HIV - RMYSPTSIL (r = 0.83), BZLF1 - SQAPLPCVL (r = 0.73) and RL9SIV - RMYNPTNIL (r = 0.90) illustrating the peptide-dependent nature of ELISA signals and the weak binding affinity of these HLA-E-restricted pathogen epitopes (Figure 4, C, iii-vii). Notably, ELISA signals for the positive control VL9 (VMAPRTVLL) and Mtb44 (RLPAKAPLL) peptides remained significantly higher than those generated by the ‘intermediate’ pathogen-derived peptides IL9 (IMYNYPAML), Mtbl4 (RMAATAQVL), RL9HIV (RMYSPTSIL), BZLF1 (SQAPLPCVL) and RL9SIV (RMYNPTNIL), even when such ‘intermediate’ binding pathogen peptides were supplied at 1.2mM concentrations. Such observations not only demonstrate the exquisite selectivity of HLA-E for its canonical VL9 leader peptide but also reveal the suboptimal nature of diverse pathogen-derived epitope sampling by HLA-E.
Both Tm values and normalised ELISA signals exhibited strong negative correlations with measuremets obtained via SEC-SAXS analyses including the maximal linear dimension (as indicated by dmax values measured in A) and molecular envelope volume (as indicated by ab initio model volume measured in A3) of HLA-E complexes in solution (Figure 4, B, i-iv). Thus, as Tm and ELISA signals increase, the average conformation of the protein ensemble in solution contracts and more closely aligns to previously obtained HLA-E crystallographic coordinates. By contrast, lower thermal stability of peptide-HLA-E complexes (Tm of 40.7°C and below) correlates with extended average conformers.
Example 4 - Structural determination of Mycobacterial epitope-associated HLA-E
The extracellular domain of the HLA-E*01:03 heavy chain and b2M light chain were crystallised in complex with the HLA-E-restricted Mycobacterial epitopes, IL9 (IMYNYPAML) and Mtbl4 (RMAATAQVL) (McMurtrey et al., 2017a) (Joosten et al., 2010). HLA-E-IL9 (IMYNYPAML) crystals diffracted to 1.7 A whereas diffraction data was collected to 2.05 A for HLA-E-Mtbl4 (RMAATAQVL). Four non-crystallographic symmetry-related molecules were present in each asymmetric unit and packing occurred in the C2 space group. Clear, unambiguous electron density enabled manual model building of the two Mycobacterial epitopes into the HLA-E peptide binding groove. Mtbl4 (RMAATAQVL) adopted the classical conformation also adopted by canonical VL9, with buried termini and a solvent-exposed central kink at residues 4 & 5 (Figure 5, A). The less constrained central kinked region of Mtbl4 displayed the greatest movement relative to VL9 with 1.6 and 1.7 A separating Ca atoms at positions 4 and 5, respectively, resulting in the central portion of the Mtbl4 backbone leaning closer toward the HLA-E al helix.
Unusually, conformational peptide dimorphism was observed for IL9 (IMYNYPAML) - this peptide adopted two distinct configurations in different non-crystallographic symmetry-related molecules of the asymmetric unit (Figure 5, B). One such configuration, adopted in molecules 2 and 3, closely aligned to canonical VL9 peptides. However, an alternative configuration was adopted in molecules 1 and 4 featuring 1.9 A positional adjustments of the position 7 IL9 peptide backbone and divergent position 7 Ala side chain trajectories, differing by up to 2.2 A relative to molecules 2 and 3 (Figure 5, B, iii & iv). In brief, the alternative IL9 peptide configuration features a position 7 Ca atom that projects 2.0 A further toward solvent relative to VL9 (VMAPRTVLL) and a position 7 secondary anchor side chain which projects away from its corresponding E pocket in the HLA-E groove.
Example 5 -Comparative structural analyses of HLA-E reveal a crucial role for the secondary E pocket in complex stability
A comparative structural analysis of HLA-E revealed that disrupted E pocket occupancy featured exclusively among structures whose corresponding SEC-SAXS-obtained molecular envelopes appeared elongated: including HLA-E-IL9 (IMYNYPAML), HLA- E-Mtbl4 (RMAATAQVL) and the previously published HLA-E-RL9HIV (RMYSPTSIL) structure (Walters et al., 2018a). Canonical primary anchor residues, position 2 Met and position 9 Leu, are present in IL9 (IMYNYPAML), Mtbl4 (RMAATAQVL) and RL9HIV (RMYSPTSIL) and project downward into their respective primary B and F pockets similar to that observed for HLA-E-VL9 (O’Callaghan et al., 1998). Therefore, additional unmet peptide binding criteria such as suboptimal E pocket-based interactions most likely contribute substantially to the conformational heterogeneity and complex instability detected by SEC-SAXS. The secondary E pocket of HLA-E comprises a deep hydrophobic recess with a more defined pocket-like aspect in contrast to classical MHC class la molecules which contain a large, highly conserved E pocket-occluding Trp side chain at position 147 (Figure 6, A, i). The E pocket of HLA-E classically accommodates a downward-projecting medium sized hydrophobic side chain at position 7 of the bound peptide, such as the highly conserved Val or Leu present in VL9 variants (Figure 6, A, ii). As previously described, the unusual conformational dimorphism of the IL9 (IMYNYPAML) peptide primarily involves position 7 with the position 7 Ca atom projecting 2 A further toward solvent relative to canonical VL9 in one of the two observed peptide configurations in the asymmetric unit. Given that HLA-E-associated peptides including IL9 commonly participate in crystal packing interfaces, such structural polymorphism may reflect weak tethering of the small position 7 Ala side chain of IL9 to the deep secondary E pocket of HLA-E, in turn rendering it susceptible to crystal packing-induced repositioning. Similarly, an unoccupied E pocket is present in the previously published HLA-E- RL9HIV structure, with the position 7 Ca projecting 3.4 A further toward solvent relative to canonical VL9 (Figure 6, A, iv). The resulting disruptions to stabilising E pocket-based-interactions and the introduction of potentially destabilising E pocket- based cavities thus ensue for HLA-E-IL9 (IMYNYPAML) and HLA-E-RL9HIV (RMYSPTSIL) (Xu et al., 1998). Suboptimal E pocket occupancy is also observed for Mtbl4 (RMAATAQVL)-associated HLA-E. A polar, non-canonical Gin at position 7 of Mtbl4 (RMAATAQVL) is buried within the hydrophobic E pocket and forms water- mediated hydrogen bonds to Ser-143 and Ser-147 of the HLA-E a2 helix with resulting entropic penalties (Figure 6, A, V). Combined mutagenesis and ELISA-based HLA-E peptide binding or thermal melt assays support a major contribution of the E pocket to HLA-E complex stability; the introduction of a canonical position 7 Val in place of the polar Gin in Mtbl4 resulted in a 36% increase in the ELISA-based binding signal and a 3.6°C increase in thermal stability (Figure 6, B & C). By contrast, the introduction of a polar Gin at position 7 of VL9 (VMAPRTVLL) resulted in a 23% reduction in the ELISA-based binding signal, with an additional 6.6°C drop in thermal stability of VL9 P7-Gln-loaded HLA-E complexes. Analogously, the introduction of a canonical Val at position 7 of Mtbl4 in SEC-SAXS analyses resulted in a smaller molecular envelope volume, a 6.8 A decrease in the Rg, a 23 A decrease in the dmax and a shifted maxima of the modulated Gaussian curve on the normalised Kratky plot relative to wild-type Mtbl4, indicative of a more globular and compact average conformer (Figure 6, D & E). By contrast, SEC-SAXS analyses of VL9 P7-Gln resulted in an increased Rg and dmax in addition to a shifted curve on the normalised Kratky plot indicative of a greater degree of conformational flexibility or asymmetry relative to wild-type VL9, further validating the importance of E pocket occupancy for optimal complex stability.
Example 6 - Unique structural features distinguish canonical VL9 versus pathogen peptide-associated HLA-E Comparative structural analyses revealed a distinct configuration located in the a2 -helical kink region which distinguishes VL9-associated versus non-VL9-bound HLA-E (Figure 7, A). Both inter-molecular hydrogen bonding and salt bridge formation between the HLA-E a2 helix and centre of the associated peptide is observed in all structures of HLA-E bound to VL9 leader peptide variants; not only does Gin- 156 of the HLA-E a2 helix form a hydrogen bond with the position 5 mainchain Oxygen of VL9, but the VL9 position 5 Arg side chain forms two salt bridges with Glu-152 of the HLA-E a2 helix. In contrast, repositioning of the position 5 peptide mainchain up to 2.3 A further toward the al helix, relative to VL9, is unanimously observed among structures of HLA-E bound to pathogen-derived peptides (Figure 7, B). These peptide mainchain rearrangements in addition to the absence of a position 5 Arg side chain in all the pathogen epitopes tested here (Mtb44 - RLPAKAPLL, RL9HIV - RMYSPTSIL, IL9 - IMYNYPAML and Mtbl4 - RMAATAQVL), result in the loss of centrally positioned, intermolecular bonds (Figure 7, C). This, in turn, appears to disrupt the a2 helical geometry and side chain configurations adopted in HLA-E-VL9 structures. Consequently, positional adjustment of the a2 helical mainchain (up to 2.4 A) and side chains (up to 4 A) between HLA-E residues Ala-139-Gln-156 ensues in structures of HLA-E in complex with pathogen epitopes (Figure 7, A). In 1 of 4 molecules of the asymmetric unit for Mtb44 (RLPAKAPLL)-associated HLA-E, the hydrogen bond connecting Gin- 156 to the position 5 main chain Oxygen of the HLA-E-associated peptide is preserved (Figure 7, C, ii). As a result, Gln-156 positioning in HLA-E-Mtb44 is not disrupted and aligns to the corresponding Gln-156 position adopted in HLA-E-VL9 structures. Notably, Mtb44 was the only pathogen peptide which exhibited comparable binding to VL9 in previous ELISA-based screens in addition to being the only pathogen peptide with partially preserved position 5 hydrogen bonding, perhaps eluding to the importance of centrally- positioned intermolecular bonds for HLA-E complex stability (Walters, McMichael and Gillespie, 2020). In remaining pathogen epitope-associated HLA-E structures the loss of the hydrogen bond between Gln-156 and the VL9 position 5 main chain results in Gln-156 side chain repositioning of up to 2.2 A (Figure 7, C, i). Further, the absence of VL9 position 5 Arg - HLA-E Glu-152 salt bridge formation results in pronounced Glu- 152 reorientation of up to 4 A in all structures of HLA-E in complex with pathogen- derived peptides, including Mtb44. The distinct Glu-152 orientation which emerges in HLA-E structures lacking associated VL9 variant peptides exhibits strong alignment among all structures of HLA-E bound to pathogen-derived peptide and permits the emergence of an alternative peptide binding motif in non-VL9-associated HLA-E molecules. Specifically, the Glu-152 side chain in pathogen peptide-bound HLA-E structures projects up to 4 A further toward the N-terminus of the peptide binding groove relative to Glu-152 in VL9-bound HLA-E, which in turn facilitates novel inter chain hydrogen bonding with the position 3 Tyr side chains present in IL9 (IMYNYPAML) and RL9HIV (RMYSPTSIL) peptides (Figure 7, D). A conserved position 3 Tyr side chain orientation with clear electron density is adopted by both peptides, in which the superposed position 3 Tyr side chains project toward the HLA-E a2 -helix and form a novel hydrogen bond with the re-orientated Glu-152. By contrast, the Glu-152 side chain orientation present in HLA-E-VL9 structures prohibits such hydrogen bond formation as it is positioned 5.5 A from the position 3 Tyr side chain present in the RL9HIV (RMYSPTSIL) or IL9 (IMYNYPAML) peptides. Given that canonical VL9 peptides contain a highly conserved Ala at position 3 which projects downward into the shallow secondary D pocket, these data help redefine what can be stably accommodated at the secondary anchor position 3 of non-VL9 HLA-E binding peptides. It also provides an example where compensatory, non-pocket-based peptide- HLA-E hydrogen bonding can permit secondary anchor side chain accommodation when a lack of shape complementarity for the corresponding pocket prohibits peptide side- chain occupancy.
Example 7 - single mutations or a combination of H99Y, F116Y and S147W mutations dramatically improve the binding of the RL9HIV peptide to HLA-E
The thermal stability of canonical HLA-E compared to HLA-E with individual mutations of H99Y, F116Y and S147W was assessed (Fig 11). Multiple test peptides can be seen to have an increased thermal gain when each of the single mutations are made, denoting an increase in the stability of binding between the peptide to HLA-E. The thermal stability of canonical HLA-E compared to HLA-E with a combination of H99Y, F116Y and S147W mutations, including single, double and triple mutations, was then assessed (Fig. 12). Fig. 12A demonstrates that single mutants improve the stability of binding between the HIV-1 derived RL9 peptide to HLA-E, which is improved further when different double mutants are used, and further still to a Tm which is almost identical to the thermal melt of canonical peptide VL9 bound to HLA-E when a triple mutant is used. DISCUSSION
In light of recent findings where MHC-E-restricted CD8+ T cells conferred sterile immune protection against SIV challenge in rhesus macaques vaccinated with recombinant RhCMV 68-1 vectors (Malouli et al., 2020), a structural and biophysical characterisation was conducted for HLA-E complexes associated with pathogen-derived epitopes versus canonical VL9 leader peptide. The panel of MHC-E-restricted epitopes evaluated in this study comprised a selection of the strongest binding pathogen-derived peptides previously identified in extensive ELISA-based screens (Walters, McMichael and Gillespie, 2020). SEC-SAXS analyses revealed marked differences in the conformational ensemble between VL9 versus pathogen-derived peptide-refolded HLA- E. Pathogen-derived peptide HLA-E complexes (including RL9SIV RMYNPTNIL, RL9HIV RMYSPTSIL, BZLF1 SQAPLPCVL, Mtbl4 RMAATAQVL and IL9 IMYNYPAML) yielded conformationally heterogeneous populations and elongated molecular envelopes with average dimensions exceeding those of superimposed crystallographic coordinates. This extended configuration featuring large protrusions surpassing the boundaries of superimposed x-ray structures likely reflects complex instability, a degree of unfolding and is reminiscent of conformationally flexible, partially peptide-loaded intermediate transition states. Accordingly, HLA-E complex stability and peptide binding strength, as indicated by melting temperatures and ELISA- based assay signals, respectively, strongly negatively correlated with the extent of molecular envelope elongation. By contrast, canonical VL9 peptide-associated HLA-E yielded conformationally homogeneous protein ensembles with compact molecular envelopes which did not project significantly beyond superposed HLA-E crystallographic coordinates, indicative of stable, fully-folded VL9 peptide-loaded HLA-E. Notably, the capacity of molar excess peptide to transform extended molecular envelopes into compact forms that resemble VL9-refolded HLA-E, implicates weak peptide binding affinity and fast dissociation rates as key contributing factors to molecular envelope elongation. These constitute unusual features for highly immunogenic epitopes such as the immunodominant RL9SIV (RMYNPTNIL) vaccine- identified ‘supertope’ which elicited Mamu-E-restricted CD8+ T cell responses in all vaccinated macaques. Notably, the majority of immunogenic MHC-E-restricted epitopes generated exceptionally weak peptide binding signals in ELISA screens (<10%VL9) such as the second immunodominant RhCMV 68-1 vaccine-identified ‘supertope’, EK9 (EKQRESREK), in addition to the immunodominant HLA-E- restricted Mycobacterial epitope, EK11 (EIEVDDDLIQK) (SEQ ID NO: 23). However, small HPLC protein elution peaks generated by HLA-E refolded with such weak binding peptides in SEC-SAXS experiments induced insufficient x-ray scattering for reliable downstream data processing (Walters, McMichael and Gillespie, 2020). Thus, in the absence of molar excess peptide, a small number of ‘intermediate’ binding HLA-E- restricted epitopes identified from extensive ELISA-based screening methods provide insufficient binding energy to drive homogeneous populations of fully-folded stable protein complexes whereas the majority of weak binding immunogenic epitopes drive insufficient protein refolding for experimental probing. These observations demonstrate that immunogenicity of MHC-E-restricted epitopes does not easily equate to optimal complex formation in vitro. Such findings are at odds with the positive correlation which frequently relates immunogenicity to complex stability among classical MHC class la molecules and raises the question of whether MHC-E-restricted CD8+ T cells might be mechanistically distinct from their classically-restricted counterparts (Harndahl et al., 2012). One possibility is that unconventionally MHC-E restricted CD8+ T cells are capable of recognising alternative structural conformers that arise in the absence of VL9.
Structural determination of HLA-E in complex with the ‘intermediate’ binding Mycobacterial peptides, IL9 (IMYNYPAML), Mtbl4 (RMAATAQVL), in addition to the previously published HLA-E-RL9HIV (RMYSPTSIL) structure, generated superposable crystallographic coordinates to VL9-bound HLA-E structures with comparable global dimensions - unlike the elongated corresponding molecular envelopes detected in SEC-SAXS analyses (Walters et al., 2018a). Crystallographic analyses were unable to detect such conformationally plastic and elongated forms present in solution due to preferential incorporation of the lowest-energy native state into the crystal and subsequent lattice-imposed dynamical restraints. Thus, crystal structures of IL9-, Mtbl4- and RL9HIV-bound HLA-E likely represent a subpopulation of the heterogeneous conformational ensemble observed in SEC-SAXS analyses which is also signified by the faint compact band observed in blue native gel analyses. Despite indistinguishable global architectures for pathogen peptide-HLA-E structures following VL9-HLA-E superposition, differential local structural features were identified and may be linked to complex instability and elongated molecular envelopes observed in solution. IL9, Mtbl4 and RL9HIV peptides contain the canonical primary anchor Met at position 2 and Leu at position 9 which optimally occupy the primary B and F pockets, respectively. However, disrupted occupancy of the large, secondary E pocket features in all three of these pathogen peptide-HLA-E structures which contrasts the optimal E pocket interactions formed between HLA-E and canonical VL9. A key role for the E pocket in HLA-E complex stability was subsequently confirmed, with the introduction of a hydrophobic position 7 Val into Mtbl4 resulting in increased thermal stability and a reduction in the extent of molecular envelope elongation. The requirement for optimal E pocket occupancy in addition to the primary B and F pockets for HLA-E is in alignment with its unique composition relative to classical MHC class I molecules; Trp- 147 is highly conserved among classical MHC class I molecules with its bulky side chain largely occluding the E pocket whereas the smaller Ser-147 present in HLA-E and its murine and rhesus counterparts, Qa-1 and Mamu-E, results in a deep hydrophobic recess with a discrete pocket-like nature (O’Callaghan et al., 1998). It is possible that the presence of a more pronounced E pocket in HLA-E serves to impose additional peptide binding criteria which must be satisfied for optimal complex stability, in turn limiting the diversity of presented peptides and promoting a VL9-dominated ligandome to maintain inhibitory CD94/NKG2A interactions with NK cells in the steady state.
Another structural feature which distinguishes pathogen peptide- versus canonical VL9- associated HLA-E is a differentially positioned a2 helical kink region. A distinct structural configuration, primarily involving Glu-152, arises in pathogen peptide- associated HLA-E complexes and could conceivably facilitate immune discrimination between VL9- versus non-VL9-bound HLA-E in the steady and stressed states, respectively. Considering Glu-152 is a common TCR-interacting residue and Glu-152 to Ala mutation results in a > 10-fold reduction in CD94/NKG2A binding REF, its exclusive repositioning in structures of HLA-E lacking associated leader peptide could signal loss of VL9 to both HLA-E-restricted CD8+ T cells and NK cells, respectively, in an innate-like manner that is not wholly dictated by peptide-specific interactions (Sullivan et al., 2007). Accordingly, semi-invariant Qa-1 -restricted lymphocytes with a common Va recognised non-Qdm peptides presented by Qa-1 in a TAP-independent manner (Doorduijn et al., 2018). Innate-like T cell recognition has previously been reported for other unconventionally-restricted subsets such as semi-invariant natural killer T (iNKT) cells restricted by CD Id and mucosal-associated invariant T (MAIT) cells restricted by MR1 (Cotton et al., 2018). Further, a number of structural studies have demonstrated multiple distinct modes of CD 1 -restricted TCR recognition which transcend the epitope/antigen-presenting-molecule co-recognition paradigm with minimal direct contact between the TCR and CD 1 -associated lipid antigen (Birkinshaw et al., 2015) (Wun et al., 2018). Whether specific TCR a or b chains employ invariant- modes of recognition to bind HLA-E in complex with pathogen-derived peptides warrants exploration. The distinct Glu-152 orientation shared among pathogen peptide- bound HLA-E structures also evolves the permitted HLA-E peptide binding motif. Although, canonical VL9 peptides contain a highly conserved Ala at position 3 which projects into the shallow secondary D pocket, Glu-152 repositioning and resultant hydrogen bond formation with the position 3 Tyr side chain of HLA-E-bound pathogen peptides, IL9 and RL9HIV, redefines what can be stably accommodated at the secondary anchor position 3 of non-VL9 HLA-E binding peptides - such distinct intermolecular hydrogen bonding appears to compensate for a lack of anchor side chain shape complementarity with its corresponding D pocket. An identical out-of-pocket binding mechanism with conserved intermolecular hydrogen bonding was previously reported for H2-Kb and a position 3 Tyr-containing octameric VSV (vesicular stomatitis virus) nucleoprotein-derived peptide (Fremont et al., 1992), with Glu-152 to Ala mutation in H2-Kb abrogating the preferential accommodation of position 3 Tyr in the associated peptide (Van Bleek and Nathenson, 1991). Further, the presence of a Glu at position 152 in H2-Kb and all known HLA-E, Qa-1 and Mamu-E alleles is unusual with Ala or Val being present at position 152 in the majority of MHC class I molecules. Thus, this mechanism may signify a unique mode of binding that can be exploited by HLA-E to broaden its peptide binding repertoire in the absence of VL9.
The results presented here demonstrate that the stringent peptide binding criteria imposed by HLA-E for optimal complex formation, including primary and secondary pocket occupancy in addition to central hydrogen bonding and salt bridge formation, are exclusively satisfied in HLA-E structures in complex with VL9 leader peptide. These numerous and restrictive peptide binding requirements throughout the length of the nonamer are likely driven by NK-cell evolutionary forces to maintain a narrow peptide binding repertoire dominated by VL9 peptide variants. A VL9-prevalent MHC- E-presented ligandome in healthy cells would in turn maintain inhibitory NK cell interactions through CD94/NKG2A engagement, whereas an abundance of alternative HLA-E-restricted peptides with comparable or higher binding affinity than VL9 could conceivably disrupt this immunoregulatory interaction. Consistent with this, the inventors have demonstrated that more diverse peptide sampling by HLA-E, in settings where MHC class I trafficking pathways are disrupted - such as in Mtb infection or following RhCMV 68-1 vaccination - is largely suboptimal. This non-optimal HLA-E peptide binding repertoire yields unstable, heterogeneous protein populations which at the structural level, share altered conformational adjustments relative to HLA-E-VL9 involving distinct a2-helical kink configurations that are likely to impact TCR and CD94/NKG2 class receptor recognition.
In summary, the inventors have demonstrated an optimised process to identify peptide epitopes presented by HLA-E on normal and abnormal cells and recognised by specific T cells and/or B-cells. Similarly, optimised methods to generate stable HLA-E-peptide complexes that can be used to make multimers to detect antigen specific T and B cells are demonstrated. Methods are described to identify which peptides bind well, moderately or poorly to HLA-E, and then apply biochemical methods to enhance peptide binding to HLA-E, such as mutational methods to enhance peptide binding to HLA-E, chemical cross linking to enhance peptide binding, use of mutated Mamu-E to enhance peptide binding and to allow use of Mamu-E multimers as surrogates for HLA-E-peptide multimers. Additionally, the inventors demonstrate efficient validation of the epitopes identified by selecting, cloning and testing T cells/ B-cells specific for HLA-E-peptide complex presented on abnormal cells, thereby validating the therapeutic potential of those identified peptides, otherwise overlooked in conventional screening.
MATERIALS AND METHODS Peptide synthesis
Peptides were purchased as lyophilised powder at >85% purity from Genscript USA prior to reconstitution in DMSO (200mM) and storage at -80 °C. A UV-labile peptide based on the HLA-B leader peptide (VMAPRTLVL) with a 3-amino-3-(2-nitrophenyl)- propionic acid residue (J residue) substitution at position 5 was synthesised by Dris Elatmioui at LUMC The Netherlands, for use in peptide binding assays.
Protein refolding
HLA-E*01:03 protein refolds were assembled in the traditional macro-refolding buffer for MHC class I molecules comprising 100 mM Tris pH8.0, 400mM L-arginine monohydrochloride, 2mM EDTA, 5mM reduced glutathione and 0.5mM oxidised Glutathione, prepared in MiliQ water. 2-Microglobulin in Urea-Mes, was initially refolded for 30 min at 4°C at a final concentration of 2 mM. Test peptide was subsequently added to the refold at a concentration of 30-60 pM followed by HLA- E*01:03 heavy chain which was pulsed into the refolding buffer to reach a final concentration of 1 pM. Refolds were subject to a 72 hour incubation period at 4°C prior to filtration through 1.0 pM cellular nitrate membranes to ensure the removal of aggregated material. Refolds were concentrated by a VivaFlow 50R system and VivaSpin Turbo Ultrafiltration centrifugal devices, both with 10 kDa molecular weight cut-offs. Refolded and concentrated material was used at a concentration of lOmg/mL in SEC-SAXS experiments without subsequent chromatographic separation.
Refolded and concentrated material intended for ELISA-based HLA-E peptide binding assays, DSF or crystallisation screening was subject to subsequent fast protein liquid chromatography (FPLC) size separation on an AKTA Start System using a Superdex S75 16/60 column. HLA-E protein complex peaks were eluted into 20 mM Tris pH8, 100 mM NaCl and discriminated from non-associated b2M and large misfolded aggregates via elution profile visualisation by UV absorbance at 280 mAU. FPLC- purified protein peaks fractions were combined and concentrated to a desired concentration for subsequent experiments using 10 kDa cut-off VivaSpin Turbo Ultrafiltration centrifugal devices - the final protein concentration was obtained by measurement of the absorbance at 280nm using a NanoDrop ND-1000 Spectrophotometer. The composition of eluted protein samples was also analysed by non-reducing SDS-PAGE electrophoresis on NuPAGE™ 12% Bis-Tris protein gels to demonstrate the presence of non-aggregated HLA-E heavy chain and b2M.
Blue native gels
HLA-E- 2m complexes previously refolded with the UV-sensitive VL9 peptide (VMAPJTVLL) were incubated in the presence of molar excess test peptide and evaluated via the Blue Native-PAGE™ Novex Bis-Tris gel system (life technologies) (Walters et al., 2018b). In brief, pre-refolded and purified HLA-E in complex with the UV-sensitive peptide was incubated at RT in the presence of 12 M excess test peptide prior to the addition of 3 pL 4x Native-PAGE™ Sample Buffer per 10 pg (10 pL) of sample. Samples were loaded onto 4-16% Native-PAGE™ Novex Bis-Tris gels with NativeMark™ Unstained Protein Standard used as the ladder control. Gel electrophoresis was carried out at 150 Volts for 2 h at RT with a current gradient ranging from 15-16 to 2-4 mAmps. Gels were subsequently rinsed three times in MilliQ water and stained for 2-3 h in SimplyBlue™ Safe Stain at RT. The MiliQ water was changed a number of times over a 24-48 h period to enable gel de-staining and a BioDoc IT Imaging System was used to obtain images.
Size exclusion chromatography-coupled small angle x-ray scattering
125mL HLA-E- 2M-peptide refolds were assembled in the L-Arginine-Tris macro refolding buffer according to the protein refolding method detailed above and incubated for 72hrs at 4°C prior to concentration with the VivaFlow 50R system with a lOkDa molecular weight cut-off (Sartorius). 45pL of pre-refolded HLA-E- 2M-peptide sample concentrated to lOmgs/mL, was subsequently injected onto a high-performance liquid chromatography (HPLC) KW402.5 column at Diamond Light Source Beamline B21, several hours post-concentration. HPLC elution buffers corresponded exactly to the L- Arginine Tris pH 8 macro-refolding injection buffer minus the protein components - although 60pM or 120pM excess peptide was added to the elution buffer for certain SEC-SAXS experiments. SEC-SAXS data were collected at Diamond Light Source Beamline B21 and images were taken every 3 seconds of X-ray-exposed HPLC-purified material over the course of a 32 minute elution period. Scattering data were circularly integrated prior to buffer subtraction followed by Guinier fitting and pairwise distribution function calculations, which were performed in the SAXS-dedicated software, ScAtter, developed by Robert Rambo (Franke et al., 2017). Ab initio bead models representing the average protein conformation in solution were generated by DAMMIF (Franke and Svergun, 2009) (Svergun, 1999) and the superposition of such bead models to crystallographic coordinates was carried out using the Supcomb package of ATSAS (Kozin and Svergun, 2001).
Peptide exchange ELISA-based HLA-E peptide binding assay
Peptide exchange ELISA-based HLA-E peptide binding assays were conducted according to a previously published method that was developed and optimised by the inventors. Peptide exchange micro-reactions were assembled in the traditional macro refolding buffer for MHC class I molecules comprising 100 mM Tris pH8.0, 400 mM 1- arginine monohydrochloride, 2 mM EDTA, 5 mM reduced glutathione, and 0.5 mM oxidized Glutathione, prepared in MiliQ water. 3 pg of previously purified HLA-E refolded with a UV-sensitive VL9 peptide variant (VMAP(J*)TLVL) and 100 pM of “exchange” peptide were added to polypropylene V-shaped 96-well plates and final reaction volumes adjusted to 125 pL prior to 5 h incubations on ice. Although HLA-E complexes previously refolded with a UV-sensitive peptide were used, peptide exchange reactions were not irradiated in a dedicated UV cabinet since this step was previously determined unnecessary for optimal peptide exchange. For titrations experiments, the final peptide concentration of the peptide exchange reaction was varied from 7.5 pM to 1.2 mM.
Peptide exchange reactions were subsequently interrogated by sandwich ELISA. 96- well ELISA plates were coated in lOmg/mL 3D12, an anti-human HLA-E capture antibody prior to a 12 h incubation period at 4°C. Following plate washing in PBS (200 pL per well) to remove excess coating antibody, ELISA plate wells were blocked with 300 pL of 2% IgG-free BSA for 2 h at RT. Blocked wells were washed five times in 0.05% Tween-based ELISA wash buffer (BioLegend) followed by a single wash in PBS prior to the addition of 50 pL of peptide exchange reaction diluted 1: 100 in 2% BSA to each well. ELISA plates containing peptide exchange reaction samples were incubated for 1 h at RT and subsequently washed in 0.05% Tween-based ELISA wash buffer and PBS. A polyclonal anti-human b2M HRP- conjugated IgG detection antibody (ThermoFisher Scientific) was diluted 1:2500 in 2% BSA and 50pL added to each ELISA well. ELISA plates were incubated in the dark for 30 min prior to wash steps in a 0.05% Tween-based ELISA wash buffer and PBS. An anti-rabbit IgG enhancement antibody, raised in goats and conjugated to HRP (EnVision+ System-HRP from Agilent) was diluted 1:15 in 2% BSA (containing 1% mouse serum) and subsequently added to each well. Enhancement antibodies were incubated for 15 min in ELISA wells prior to wash steps in a 0.05% Tween-based ELISA wash buffer and PBS. ELISA plates were developed in 100 pL of 3,3',5,5'-tetramethyl benzidine (TMB) substrate (BioLegend), dark incubated at RT for 10 min and reactions terminated by the addition of 100 pL H2S04 STOP solution (BioLegend). Absorbance readings were measured at 450 nm on a FLUOstar OMEGA plate reader and such inter-assay readings normalised via background subtraction and expression of each signal as a percentage of the positive control signal. Pearson product-moment correlation coefficients were generated for normalised ELISA signals versus various parameters obtained from SEC-SAXS analyses.
Differential scanning fluorimetry
Similar to previously published differential scanning fluorimetry (DSF) methodology (Anjanappa et al., 2020), the thermal stability of empty HLA-E (0.45 mg/ml) with added peptide was measured by a 30 minute co-incubation at room temperature with 12 Molar excess of peptide (120uM) dissolved in L-Arginine redox solution (400mM L-Arginine monohydrochlride, 5mM Reduced Glutathione, 0.5mM Oxidised Glutathione, 2mM EDTA, lOOmM Tris pH 8). Following incubation, two Prometheus NT.48 Series nanoDSF Grade Standard Capillaries (Nanotemper, Munich, Germany) were filled per sample and loaded into a Prometheus NT.48 fluorimeter (Nanotemper) controlled by PR.ThermControl (version 2.1.5). Excitation power was pre-adjusted to obtain between 8000 and 20,000 Raw Fluorescence Units for fluorescence emission at 330 nm and 350 nm, and samples were then heated a rate of 1 °C/min from 20 °C to 95 °C.
Statistical analysis
Pearson product-moment correlation coefficients were calculated for DSF-determined melting temperatures and normalised ELISA-based peptide binding signals to establish the congruity between these techniques. Additionally, linear correlations were identified between melting temperatures or normalised ELISA-based signals and various SEC-SAXS parameters including the maximal dimensions of the average conformation in solution (dmax) or the volume of the DAMMIF ab initio molecular envelope model (A3). X-ray crystallography lOOnL of FPLC-purified peptide-refolded HFA-E at lOmg/mF was mixed with lOOnF reservoir buffer in crystallisation wells by the mosquito protein crystallisation robot (TTP FabTech) and equilibrated by sitting drop vapour-diffusion at 20 °C (Walter et al., 2005). Crystals of IF9 (IMYNYPAMF)-bound HFA-E grew in 2.2 M ammonium sulphate, 0.1 M MES at pH 5.8 whereas Mtbl4 (RMAATAQVF)-bound HFA-E crystals grew in 3 M ammonium sulphate 0.1 M MES at pH 6. Crystals were cryopreserved in 25% glycerol by Dr. Karl Harlos of Oxford University and diffraction data were collected at the Diamond Light Source, beamlines i04 (HLA-E-Mtbl4 structure) and i03 (HLA-E-IL9 structure). Diffraction data were auto-indexed by Xia2 DIALS. Since the outer shell CCl/2 exceeded the minimum threshold (>0.3) for both datasets, no reflections were excluded from downstream analysis. A more conservative data truncation approach according to the Rmerge and I/sigma cut-offs has been shown to result in the elimination of useful data which would otherwise have contributed to model quality (Karplus and Diederichs, 2012). Thus, the well-established and reproducible indicator of crystallographic data quality, CCl/2, superseded such traditional thresholds and was adopted as a determinant for data truncation for the crystal structures presented in this study. Molecular replacement was conducted by MolRep of the CCP4i suite using published crystallographic coordinates of HLA-E (PDB ID: 6GH1), with the Mtb44 (RLPAKAPLL) peptide coordinates removed, as an initial phasing model (Walters et al., 2018b) (Winn et al., 2011) (Vagin and Teplyakov, 2010) (Murshudov et al., 2011). Rigid body, restrained and TLS refinement were computed by the CCP4i REFMAC5 (Murshudov et al., 2011) or Phenix. refine (Afonine et al., 2012) between iterative cycles of manual model building in Coot (Emsley et al., 2010). Geometry was validated by MolProbity (Chen et al., 2010), prior to visualisation of the model in the PyMOL Molecular Graphics System, version 2.0 (Schrodinger, LLC) and further investigation using PDBePISA (Krissinel and Henrick, 2007) and PDBeFOLD (Krissinel and Henrick, 2004). Thermal melt assay
By ROX Dye Incorporation: The thermostability of canonically refolded HEA-E-b2ih peptide complexes and C terminus extended peptides with a cysteine refolded with HLA-E containing a tyrosine to cysteine mutation was determined by heat-induced fluorescent dye incorporation, using the commercially available Protein Thermal Shift Dye Kit™ (Applied Biosystems). 5 pg of test HEA-E-b2ih complexes was aliquoted into 0.1 mL MicroAmp Fast Optical 96-well plates containing pre-mixed Protein Thermal Shift Dye and Protein Thermal Shift Buffer. Sample buffer (either PBS or Tris pH8, 100 mM NaCL) was added to achieve a final volume of 20 pL. Control samples reconstituted with buffer were prepared to monitor background fluorescent signal. Both samples and controls were set up in quadruplicate. Thermal-driven dye incorporation was measured on an Applied Biosystem Real-Time 7500 Fast PCR System. Data was collected over a temperature ramp ranging from 25 to 95 °C, with 1 °C intervals. Melt curve data were analysed using Protein thermal Shift Software vl.3, and median Derivative Tm values (°C) are reported.
By Differential Scanning Fluorimetry: The thermal stability of canonical HFA-E and HFA-E H99Y, HFA-E F116Y and HFA-E S147W material was measured by Differential scanning fluorimetry analysis. The individual HLA-E proteins were incubated at 0.45 mg/ml with Molar excess of peptide in 20mM Tris pH7, lOOmM NaCl buffer at a final volume of 20uF. Following a 30 minute incubation at room temperature, individual samples were subsequently split, transferred into two Prometheus NT.48 Series nanoDSF Grade Standard Capillaries (Nanotemper, Munich, Germany) and then placed in the capillary tray of a Prometheus NT.48 fluorimeter (Nanotemper). Excitation power was pre-adjusted to obtain between 8000 and 20,000 Raw Fluorescence Units for fluorescence emission at 330 nm and 350 nm. A thermal ramp ranging from 20 °C to 95 °C, at a rate of 1 °C/min, was applied. Automated thermal melt data calling was generated by the analysis software within PR.ThermControl, (version 2.1.5) software.

Claims

1. A method of identifying a peptide which is capable of forming an HLA-E:peptide complex and being recognised by a T-cell and/or B-cell, comprising the steps of:
(a) Performing an assay to determine the level of binding of a peptide of interest in a HLA-E:peptide complex, relative to the level of binding of a reference peptide to HLA-E in a reference HLA-E complex;
(b) Stratifying the peptide of interest into one of the following groups based on its relative binding determined in step (a):
(i) More than about 70% of the level of binding to HLA-E relative to the level of binding of the reference peptide;
(ii) less than about 70% of the level of binding to HLA-E relative to the level of binding of the reference peptide; and
(c) Performing an assay to determine whether a HLA-E:peptide complex comprising the peptide of interest is recognised by a T-cell and/or B-cell.
2. The method of claim 1, wherein a peptide of interest is stratified into group (ii) in step (b) is further stratified into one of the following groups based on its relative binding determined in step (a):
(al) Between about 35% to about 70% of the level of binding to HLA-E relative to the level of binding of the reference peptide;
(bl) Between about 15% to about 35% of the level of binding to HLA-E relative to the level of binding of the reference peptide; or (cl) Under 15% of the level of binding to HLA-E relative to the level of binding of the reference peptide.
3. The method of claim 1 or claim 2, wherein the reference peptide is VL9 and the reference HLA-E:peptide complex comprises VL9, HLA-E and B2 microglobulin.
4. The method of any of claims 1-3, wherein the peptide of interest is identified using one or more amino acid sequence prediction process.
5. The method of any of claims 1-4, further comprising: before step (a), performing an initial screening assay to determine whether a HLA-E:peptide complex comprising the peptide of interest is capable of being expressed on a cell surface, wherein if the HLA-E:peptide complex comprising the peptide of interest is determined to be capable of being expressed on a cell surface, the method proceeds to step (a).
6. The method of claim 5, wherein the screening assay comprises transfecting into a cell a nucleic acid encoding a polypeptide comprising amino acids of the peptide of interest, B2microglobulin and HLA-E heavy chain.
7. The method of any of claims 1-6, wherein the assay of step (a) comprises an Enzyme-linked immunosorbent assay (ELISA).
8. The method of any of claims 1-7, further comprising: after the assay of step (b), performing one or more further assays to determine the biophysical and/or biochemical characteristics of the HLA-E:peptide complex comprising the peptide of interest.
9. The method of claim 8, wherein the one or more further assays comprises Native PAGE, Thermal melt (Tm) analysis, and/or Small Angle X-ray Scattering (SAXS).
10. The method of any of claims 1-9, comprising: in the assay of step (c), using multimers of the HLA-E:peptide complex comprising the peptide of interest to determine whether a HLA-E:peptide complex comprising the peptide of interest is recognised by a T-cell and/or B-cell.
11. The method of any of claims 1-10, comprising: in the assay of step (c), further determining whether an HLA-E:peptide complex comprising the peptide of interest is capable of stimulating an immune response in one or more T-cell and/or B-cell.
12. The method of any of claims 1-11, further comprising performing one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest in one or more of: the assay of step (a); the initial screening assay of claim 5; and/or the assay of step (c).
13. The method of claim 12, wherein the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest comprises one or more of:
(A) First forming a HLA-E:peptide complex using a UV-sensitive reference peptide and subsequently exchanging this with the peptide of interest such that an HLA- E:peptide complex comprising the peptide of interest is formed, optionally wherein this step is performed in the presence of a molar excess of the peptide of interest, such as at IOOmM or more;
(B) Crosslinking one or more amino acid of the peptide of interest to one or more amino acid of HLA-E, optionally wherein the UV-sensitive reference peptide is VMAP(J*)TLVL, where J* is 3-amino-3-(2-nitrophenyl)-propionic acid;
(C) Introducing one or more crosslink between amino acids in HLA-E;
(D) Introducing one or more mutation in HLA-E to increase the stability of the binding of a peptide of interest in the HLA-E:peptide complex; or
(E) Forming a complex of the peptide of interest with Mamu-E.
14. The method of claim 13, wherein in (B) the crosslinking one or more amino acid of the peptide of interest to one or more amino acid of HLA-E comprises: mutating one or more residues of HLA-E to a cysteine or lysine, and optionally substituting one or more residues of the peptide of interest to a cysteine, lysine or synthetic amino acid such that a crosslink is capable of being formed between the peptide of interest and HLA-E; or mutating one or more amino acids of HLA-E to a positively or negatively charged amino acid, and forming a salt bridge with an oppositely charged amino acid in the peptide, wherein the oppositely charged amino acid in the peptide is optionally substituted in.
15. The method of claim 14, wherein substituting one or more residues of the peptide of interest comprises substituting the residue in the first or second position of the peptide of interest to a cysteine, a homocysteine, or a synthetic amino acid comprising a free sulphydryl group, such that a crosslink in the form of a disulphide bond can be formed between a mutant amino acid in the HLA-E heavy chain and the amino acid at the first or second position in the peptide.
16. The method of any of claims 14 or 15, wherein the method comprises: mutating the tyrosine at position 84 of HLA-E to a cysteine, adding a glycine and cysteine to the carboxy terminus of the peptide of interest, and forming a disulphide bond between the cysteine at position 84 of HLA-E and the cysteine added to the carboxy terminus of the peptide of interest; and/or mutating the methionine at position 45 of HLA-E to a cysteine, substituting the amino at position two of the peptide of interest to a cysteine, a homocysteine, or a synthetic amino acid comprising a free sulphydryl group, and forming a disulphide bond between the cysteine at position 45 of HLA-E and the cysteine, homocysteine, or a synthetic amino acid at position two of the peptide of interest.
17. The method of claim 15 or 16, wherein the synthetic amino acid comprising a free sulphydryl group is a homocysteine analogue, (2S)-2-amino-5-sulfanylpentanoic acid or (2S)-2-amino-6 sulfanylhexanoic acid.
18. The method of claim 13, wherein: in (C) the one or more crosslink between amino acids in HLA-E is introduced bymutating the tyrosine at position 84 of HLA-E to a cysteine, mutating the alanine at position 139 of HLA-E to a cysteine, and forming a disulphide bond between the two cysteine residues at position 84 andl39 of HLA-E;. in (D) the one or more mutation in HLA-E comprises or consists of mutating one or more of, such as one of, two of, or all of, Histidine at position 99 to Tyrosine, mutating Phenylalanine at position 116 to Tyrosine or mutating Serine at position 147 to Tryptophan; or in (E), the Mamu-E comprises one or more of the following mutations: P57S, E79R, and/or G150A.
19. The method of any of claims 12-18, wherein: if in the assay of step (b) the peptide of interest is stratified into group (i), one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest does not need to be performed; or if in the assay of step (b) the peptide of interest is stratified into group (ii), one or more step to improve the stability of the HLA-E:peptide complex comprising is performed.
20. The method of any of claims 12-19, wherein if in the assay of step (b) the peptide of interest is stratified into group (al), the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest comprises:
(A) First forming a HLA-E:peptide complex using a UV-sensitive reference peptide and subsequently exchanging this with the peptide of interest such that an HLA- E:peptide complex comprising the peptide of interest is formed; or
(E) Forming a complex of the peptide of interest with Mamu-E.
21. The method any of claims 12-19, wherein if in the assay of step (b) the peptide of interest is stratified into group (bl), the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest comprises:
(C) Introducing one or more crosslink between amino acids in HLA-E; and/or
(D) Introducing one or more mutation in HLA-E to increase the stability of the binding of a peptide of interest in the HLA-E:peptide complex; or
(E) Forming a complex of the peptide of interest with Mamu-E.
22. The method of any of claims 12-19, wherein if in the assay of step (b) the peptide of interest is stratified into group (cl), the one or more step to improve the stability of the HLA-E:peptide complex comprising the peptide of interest comprises:
(B) Crosslinking one or more amino acid of the peptide of interest to one or more amino acid of HLA-E; or
(E) Forming a complex of the peptide of interest with Mamu-E.
23. The method of any of claims 1-22, further comprising:
Identifying one or more T-cell or B- cell which recognises the HLA-E:peptide complex comprising the peptide of interest.
24. The method of claim 23, further comprising performing an assay to determine whether a T-cell or B-cell which is identified as recognising the HLA-E:peptide complex comprising the peptide of interest, is capable of recognising and/or being activated by an abnormal cell which comprises the peptide of interest, optionally wherein the abnormal cell is an infected cell, a tumour cell or a cell undergoing a stress response.
25. A method of identifying one or more T-cell or B- cell which recognises an HLA- E:peptide complex comprising a peptide of interest, comprising:
(a) performing the method of any of claims 1-24 to identify a peptide which is capable of forming an HLA-E:peptide complex and being recognised by a T-cell and/or B-cell;
(b) performing an assay to identify one or more T-cell or B-cell which recognises the HLA-E:peptide complex; and optionally further comprising:
(c) determining whether the T-cell and/or B-cell is capable of recognising and/or being activated by an abnormal cell which comprises the peptide of interest, optionally wherein the abnormal cell is an infected cell, a tumour cell or a cell undergoing a stress response; and/or
(d) using a T cell or an antibody produced by the B cell to determine whether the peptide of interest is presented by HLA-E on the surface of an abnormal cell or healthy cell, optionally wherein the abnormal cell is an infected cell, a tumour cell or a cell undergoing a stress response
26. A mutant HLA-E heavy chain comprising one or more mutation which permits the formation of a HLA-E:peptide complex with increased stability when compared to the complex without the mutant HLA-E heavy chain.
27. The mutant HLA-E heavy chain of claim 26, wherein the one or more mutation comprises or consists of: a) a mutation at Histidine at position 99 to Tyrosine; or b) a mutation at Phenylalanine at position 116 to Tyrosine; or c) a mutation at Serine at position 147 to Tryptophan; or d) a mutation at Histidine at position 99 to Tyrosine and a mutation at Phenylalanine at position 116 to Tyrosine; or e) a mutation at Histidine at position 99 to Tyrosine and a mutation at Serine at position 147 to Tryptophan; or f) a mutation at Histidine at position 99 to Tyrosine, a mutation at phenylalanine at position 116 to Tyrosine, and a mutation at Serine at position 147 to Tryptophan.
EP22729763.7A 2021-06-01 2022-05-31 Peptide screen Pending EP4348255A1 (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
GBGB2107821.7A GB202107821D0 (en) 2021-06-01 2021-06-01 Peptide screen
GBGB2119143.2A GB202119143D0 (en) 2021-12-31 2021-12-31 Peptide screen
PCT/GB2022/051383 WO2022254200A1 (en) 2021-06-01 2022-05-31 Peptide screen

Publications (1)

Publication Number Publication Date
EP4348255A1 true EP4348255A1 (en) 2024-04-10

Family

ID=82019937

Family Applications (1)

Application Number Title Priority Date Filing Date
EP22729763.7A Pending EP4348255A1 (en) 2021-06-01 2022-05-31 Peptide screen

Country Status (2)

Country Link
EP (1) EP4348255A1 (en)
WO (1) WO2022254200A1 (en)

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB201909509D0 (en) * 2019-07-02 2019-08-14 Immunocore Ltd Peptide-MHC complexes

Also Published As

Publication number Publication date
WO2022254200A8 (en) 2024-01-04
WO2022254200A1 (en) 2022-12-08

Similar Documents

Publication Publication Date Title
Irving et al. Interplay between T cell receptor binding kinetics and the level of cognate peptide presented by major histocompatibility complexes governs CD8+ T cell responsiveness
Chappell et al. Expression levels of MHC class I molecules are inversely correlated with promiscuity of peptide binding
Reiser et al. Crystal structure of a T cell receptor bound to an allogeneic MHC molecule
TW202017940A (en) Method for high throughput peptide-mhc affinity screening for tcr ligands
Natarajan et al. Sodium dodecyl sulfate stability of HLA-DR1 complexes correlates with burial of hydrophobic residues in pocket 1
CN108700582B (en) Quality of immune synapses predicts potency of Chimeric Antigen Receptor (CAR) T cells
Justesen et al. Functional recombinant MHC class II molecules and high-throughput peptide-binding assays
Chen et al. Structure-based design of altered MHC class II–restricted peptide ligands with heterogeneous immunogenicity
Reiser et al. Analysis of relationships between peptide/MHC structural features and naive T cell frequency in humans
Walters et al. Primary and secondary functions of HLA-E are determined by stability and conformation of the peptide-bound complexes
Audehm et al. Key features relevant to select antigens and TCR from the MHC-mismatched repertoire to treat cancer
CA2440399C (en) Epitope testing using hla
AU2024200495A1 (en) Magea1 immunogenic peptides, binding proteins recognizing magea1 immunogenic peptides, and uses thereof
US11339383B2 (en) G proteins
EP4348255A1 (en) Peptide screen
Yadav et al. The H4b minor histocompatibility antigen is caused by a combination of genetically determined and posttranslational modifications
JP2021536578A (en) Methods for High Throughput Peptide-MHC Affinity Screening for TCR Ligands
CN118019980A (en) Peptide screening
Strausbauch et al. Identification of mimotopes for the H4 minor histocompatibility antigen.
Mizote et al. Production of NY-ESO-1 peptide/DRB1* 08: 03 tetramers and ex vivo detection of CD4 T-cell responses in vaccinated cancer patients
CN107987155A (en) Identify the φt cell receptor of SAGE1 antigen small peptides
Walters Towards a universally effective HLA-E-targeted HIV vaccine-a structural and biophysical exploration
Sakoguchi et al. RIFINs displayed on malaria-infected erythrocytes bind both KIR2DL1 and KIR2DS1
Saikia The Molecular Mechanism of Major Histocompatibility Complex Class I Peptide Binding and Exchange
IL310374A (en) Antigen binding proteins specifically binding ct45

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: UNKNOWN

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20240102

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR