Engineering chimeric human and mouse major histocompatibility complex (MHC) class I tetramers for the production of T-cell receptor (TCR) mimic antibodies

Abstract

Therapeutic monoclonal antibodies targeting cell surface or secreted antigens are among the most effective classes of novel immunotherapies. However, the majority of human proteins and established cancer biomarkers are intracellular. Peptides derived from these intracellular proteins are presented on the cell surface by major histocompatibility complex class I (MHC-I) and can be targeted by a novel class of T-cell receptor mimic (TCRm) antibodies that recognise similar epitopes to T-cell receptors. Humoural immune responses to MHC-I tetramers rarely generate TCRm antibodies and many antibodies recognise the α3 domain of MHC-I and β2 microglobulin (β2m) that are not directly involved in presenting the target peptide. Here we describe the production of functional chimeric human-murine HLA-A2-H2Dd tetramers and modifications that increase their bacterial expression and refolding efficiency. These chimeric tetramers were successfully used to generate TCRm antibodies against two epitopes derived from wild type tumour suppressor p53 (RMPEAAPPV and GLAPPQHLIRV) that have been used in vaccination studies. Immunisation with chimeric tetramers yielded no antibodies recognising the human α3 domain and β2m and generated TCRm antibodies capable of specifically recognising the target peptide/MHC-I complex in fully human tetramers and on the cell surface of peptide pulsed T2 cells. Chimeric tetramers represent novel immunogens for TCRm antibody production and may also improve the yield of tetramers for groups using these reagents to monitor CD8 T-cell immune responses in HLA-A2 transgenic mouse models of immunotherapy.

Fig 1

(A) T cell receptor mimic (TCRm) antibodies recognise short peptides, which can be derived from intracellular antigens, presented by MHC-I molecules on the cell surface. The T-cell receptor (TCR) on T cells also bind to this peptide/MHC-I complex, and this process is a critical component of normal immune surveillance. (B) Design of a chimeric MHC-I molecule, in which the human α3 and β2m domains are replaced with their murine counterparts (orange), while the human antigen binding α1 and α2 domains presenting the peptide remain intact.

Fig 2. Production and validation of chimeric MHC-I tetramers.

HLA-A2 (A) and chimeric HLA-A2-H2D^d heavy chains were refolded with human (B), mouse (C), or mutated mouse β2m (D) (hβ2m, mβ2m, and mβ2m_mut, respectively), in the presence of an influenza virus peptide GILGFVFTL (Flu). The protein refolding reactions were subsequently separated by FPLC. The arrows indicated the protein peaks containing the refolded MHC-I monomers. Two independent experiments were performed for each construct Right panels: a Flu peptide-stimulated CTL cell line was stained with human and chimeric HLA-A2 / Flu tetramers and co-stained with a CD3 antibody. This was performed three times for each tetramer and twice used a CTL line from two individuals. Gate and frequencies indicate the percentage of Flu-specific T cells bound by the respective tetramer in the T-cell line.

Fig 3. p53RMP tetramer production and fusion.

(A) The p53RMP peptide was refolded with the HLA-A2-H2D^d chimeric heavy chain and mutated mouse β2m light chain, and the monomer fraction (arrow) was collected and tetramerised (n = 3) before being used as the immunogen to produce TCRm antibodies. (B) Validation of p53RMP monomers. Wild type (WT) or chimeric p53RMP and p53GLA monomers were immobilised onto an ELISA plate by Extravidin, and a p53RMP TCRm T1-116C was used to detect the monomers (n = 1). (C) Representative ELISA screening data for TCRm antibodies from chimeric tetramer fusions that were reactive with the human p53RMP tetramers used for primary screening. Hybridoma supernatants were screened against plates coated with extravidin, or extravidin tetramerised monomers comprising different combinations of HLA-A2 and HLA-A2-H2D^d heavy chains, human and mouse β2m, and p53RMP and Flu peptides. Hybridoma specificity was determined based on the ELISA screening results. ‘Other’ refers to an antibody where the domain specificity could not be assigned, as described in S1 Table. ELISA data illustrated is representative of three independent experiments.

Fig 4. Binding of TCRm antibodies to p53RMP/HLA-A2 complexes on live cells detected by flow cytometry.

(A) The left panel illustrates purified T1-29D mAb staining of T2 cells pulsed with p53RMP peptide as well as 2 other p53 peptides and peptides derived from survivin, HCMV and Flu (at 100μM). Staining was detected using an APC-conjugated anti-mouse secondary antibody. The T1-84C supernatant was tested only against Flu and p53RMP peptides (right panel). (B) T2 cells pulsed with the p53RMP peptide at 100μM were stained with the purified T1-29D mAb at the indicated concentrations. The HLA-A2 specific mAb BB7.2 was used in parallel to detect HLA-A2 expression. Mean fluorescence intensities (MFIs) of the staining were plotted on the right panel (for clarity the left panel does not contain all of the tested concentrations). (C) T2 cells pulsed with the p53RMP peptide at various concentrations were stained with T1-29D and BB7.2 at 10 μg/ml. MFIs of the staining were plotted on the right panel. Flu peptide pulsing at 100μM was used as a control. Data in each panel is representative of two independent experiments.

Fig 5. TCRm specific for p53 peptide GLAPPQHLIRV (p53GLA).

HLA-A2 and chimeric HLA-A2-H2D^d tetramers containing the human p53GLA peptide were generated and used for TCRm antibody production. Three TCRm mAbs, of which T2-108A was generated against HLA-A2 WT tetramer and T2-2A and T2-116C against chimeric tetramers, were generated. (A) Supernatants from early culture (early) and post-cloning (clone) stage hybridoma supernatants of T2-108A as well as early T2-2A and T2-116A hybridoma supernatants were tested for their binding to p53GLA-HLA-A2 tetramers by ELISA. The Flu tetramer was used as a control. Data are representative of two independent experiments. (B) The TCRm hybridoma supernatants described above were tested for their binding to peptide-pulsed T2 cells and data were analysed by FACS. Data are representative of five independent experiments. (C) T2 cells pulsed with p53GLA peptide at various concentrations were stained with purified T2-2A or T2-116A and BB7.2 mAbs at 10 μg/ml. MFIs of the staining are plotted on the right panel. Flu peptide pulsing at 100μM was used as a control. Data are representative of two independent experiments.