Solution mapping of MHC-I:TCR interactions using a minimalistic protein system
Edited by G. Marius Clore, National Institutes of Health, Bethesda, MD; received March 17, 2025; accepted May 1, 2025
Abstract
Recognition of epitopic peptide antigens presented on class I major histocompatibility complex (MHC-I) proteins by T cell receptors (TCRs) forms the cornerstone of immune surveillance, leading to a plethora of adaptive immune responses. Characterization of TCR:peptide/MHC-I interactions is critical for understanding immune recognition, and developing immunotherapies, but the large variation in docking orientations of TCRs on their peptide/MHC-I targets challenges structural modeling. NMR spectroscopy could potentially resolve this ambiguity, but the large size of the TCR:peptide/MHC-I complex limits data quality. Here, we demonstrate that a designed MHC-I protein, SMART A*02:01, enables facile solution mapping of MHC-I:TCR interactions at scale. Our approach can be combined with computational modeling and structure-guided engineering to aid the development of TCR-based therapeutics.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.
MHC-I proteins are essential components of the adaptive immune system (1). Through presentation of short epitopic peptides at the cell surface, these molecules provide a basis for immune surveillance by TCRs via recognition of aberrant or foreign antigens, triggering CD8+ cytotoxic T cell responses (2). Classical MHC (also called human leukocyte antigens, HLAs) class I complexes are composed of a short (8 to 15 amino acids) peptide, an invariable light chain (β2 microglobulin, β2m), and a highly polymorphic heavy chain which forms the peptide-binding groove. This functional polymorphism of MHC-I proteins results in diverse repertoires of displayed peptides and therefore a large variability in interactions with clonotypic TCRs present in the human population (3–5). Approximately 250 peptide-bound MHC-I (pMHC-I):TCR structures have currently been solved (6, 7), revealing an extensive variability in docking orientations among known TCR:pMHC-I complexes (2, 6, 8, 9). Recent findings have revealed unconventional docking orientations, underscoring the potential for unique binding mechanisms which may remain to be discovered (10). Crystallography and cryo-EM-based structure determination methods of MHC-I complexes (11–13) are hindered by low-throughput, and require a significant commitment of time and resources (14). Solution NMR spectroscopy is limited by the large molecular weight of the complex (~95 kDa), which causes peak broadening and spectral overlap (15–17). We have recently developed a computationally designed SMART MHC-I protein, comprising a single-chain peptide-binding groove with the light chain (β2m) and α3 domains replaced by a helical stabilizer protein domain (18). This design leads to prolonged stability (by preventing β2m dissociation, which leads to irreversible aggregation) and reduced molecular weight (29 vs. 45 kDa). The SMART MHC-I design aims to preserve the heavy chain/β2m interface, and maintain the native properties of the peptide-binding groove (19, 20). Here, we demonstrate the use of SMART A*02:01 (containing the heavy chain sequence of the common HLA-A*02:01 allele) alongside NMR-based solution mapping experiments to rapidly determine its TCR docking orientation.
Results
To determine whether SMART A*02:01 shares structural and conformational characteristics similar to the native HLA-A*02:01, we expressed and purified isotopically labeled SMART A*02:01 through in vitro refolding (21). We prepared complexes with two known high-affinity peptides, TAX9 (LLFGYPVYV) and NYESOV (SLLMWITQV), and carried out backbone assignments (16). To validate that the average solution conformation of SMART A*02:01 recapitulates the native peptide-binding groove, we extracted Cβ secondary chemical shifts which report on protein secondary structure (22–24). The Cβ secondary chemical shifts of SMART A*02:01 are in good agreement with the canonical fold of the native HLA-A*02:01/β2m/TAX9 peptide-binding groove, as expected (25–27). To detect local effects on the SMART A*02:01 protein resulting from differences between the TAX9 and NYESOV peptides, we calculated chemical shift perturbations (CSPs) by overlaying 2D 1H–15N transverse relaxation-optimized spectroscopy (TROSY) spectra (Fig. 1). Significant CSPs are found at residues that either directly interact with the peptide or are allosterically impacted by peptide/MHC interactions (Fig. 1A). Prominent examples include Y229 and Q225 at the α2 helix, T143 and D147 on the α1 helix, and W217 on the α2-1 helix, all of which are impacted by chemical and structural differences between the two peptides (Fig. 1B). We also detect CSPs for a cluster of residues, including W130, E123, E116, and R135, near the peptide C-terminus, highlighting the sensitivity of solution NMR for revealing induced conformational changes (Fig. 1B). These results demonstrate that SMART A*02:01 binds high affinity peptides in a manner consistent with the existing structural data for native MHC-I molecules (15–17, 26, 28–31).
Fig. 1.
We then prepared a 1:1 complex of SMART A*02:01/TAX9 with the A6c134 TCR (33) and collected 2D TROSY and 3D NOESY spectra to detect and assign the TCR-bound state (34). Upon A6c134 binding, we observed numerous peak shifts and several instances of exchange line broadening, due to conformational exchange between the apo and TCR-bound states occurring at the microsecond timescale (Fig. 2A). We observe slow-exchange complex formation in our NMR spectra, as expected [native complex KD = 4 nM (33)]. To map the TCR binding sites onto SMART A*02:01/TAX9, we calculated CSPs between the apo and TCR-bound TROSY spectra (Fig. 2 A and B). We observe strong CSPs and exchange broadening at the primary A6c134 binding sites on SMART A*02:01, with all significant effects localized to the α1/α2 domains, and β-sheet floor. The placement of these perturbations suggests that the TCR binding site forms a diagonal docking orientation, in agreement with the crystal structure of the native complex (Fig. 2B) (33). We observe the strongest CSPs or exchange broadened sites at E128, T143, A220, E224, Q225, L230, T233, and W237, which contribute to TCR binding in the crystal structure (Fig. 2 C and D). In addition to detecting direct binding contacts, we observe CSPs along the α2-1 helix (residues W217, E218, I194, and Y193), due to a conformational change in the orientation of the α2-1 helix in the A6c134-bound crystal structure (Fig. 2E). In summary, our data demonstrate that solution NMR can accurately map the docking orientation of A6c134 onto SMART A*02:01/TAX9, in agreement with existing structural data.
Fig. 2.
Discussion
TCR binding orientations on pMHC-I antigens can vary significantly between different cognate complexes, and conventional structure determination methods require significant labor and resource commitments (17). Here, we present the application of our SMART A*02:01 design (18) to NMR-based solution mapping studies, allowing for rapid detection of the TCR docking orientation in a physiologically relevant, aqueous environment. Our results provide a proof-of-concept of this solution mapping approach, validated by comparison with the high-resolution X-ray structure of the A*02:01/TAX9:A6c134 complex. The preservation of the overall structure of SMART A*02:01/TAX9 relative to the native HLA-A*02:01/TAX9 complex indicate that other combinations of A*02:01 peptide/binder pairings [including TCRs, antibodies, and other engineered binders (12, 35)] can be resolved using this approach. Our established SMART A*02:01 resonance assignments can be transferred to complexes with other peptide antigens using straightforward 3D NMR approaches [triple resonance approaches (HNCO) or amide–amide NOESYs (36, 37) enabling studies of other TCRs with their cognate peptide/SMART A*02:01 antigens. Next steps will expand SMART designs to other common HLA allotypes (18). A pertinent application of this work is to inform structural modeling methods for improving the predictions of TCR/pMHC-I cpmplexes (38). Given the high value associated with accurate prediction of these complexes to inform TCR engineering approaches, CSPs can be readily integrated into the structure modeling process (39). Our work presents an opportunity to combine rapid solution mapping of TCR docking with advancements in computational structure prediction (40, 41), to enable high-resolution structural characterization of pMHC-I:TCR interactions.
Data, Materials, and Software Availability
All assignments were deposited in BMRB under accession codes 52878 (SMART A*02:01/NYESOV) (23), 52879 (SMART A*02:01/TAX9) (24), and 52880 (SMART A*02:01/TAX9/A6c134) (34). Extended methods are included in SI Appendix. All other data are included in the manuscript and/or SI Appendix.
Acknowledgments
We acknowledge the biomolecular NMR facility at the Eldridge Reeves Johnson Foundation at the University of Pennsylvania, Grants R01AI143997 (N.G.S.), R35GM125034 (N.G.S.), U01DK112217 (N.G.S.), T32GM132039 (NIH training grant supporting C.H.W), NIH R01AI103867 (K.C.G.) and HHMI (K.C.G and D.B.), The Asplundh Foundation, and The Children’s Hospital of Philadelphia Cell and Gene Therapy Collaborative. This work is part of the MATCHMAKERS and NexTGen teams, supported by the Cancer Grand Challenges partnership financed by CRUK (CGCATF-2023/100004, CGCATF-2023/100006, CGCATF-2023/100008, and CGCATF-2021/100002), the National Cancer Institute (OT2CA297575, OT2CA297242, OT2CA297288, and CA278687-01), and The Mark Foundation for Cancer Research. All authors are members of the MATCHMAKERS team. N.G.S. is also a member of the NexTGen team.
Author contributions
C.H.W., A.C., X.T.C., W.L.W., K.C.G., D.B., and N.G.S. designed research; C.H.W., A.C., X.T.C., and W.L.W. performed research; C.H.W., A.C., X.T.C., and W.L.W. contributed new reagents/analytic tools; C.H.W., A.C., X.T.C., and W.L.W. analyzed data; and C.H.W., K.C.G., D.B., and N.G.S. wrote the paper.
Competing interests
The authors declare no competing interest.
Supporting Information
Appendix 01 (PDF)
- Download
- 193.55 KB
References
1
P. Parham, T. Ohta, Population biology of antigen presentation by MHC class I molecules. Science 272, 67–74 (1996), https://doi.org/10.1126/science.272.5258.67.
2
J. Rossjohn et al., T cell antigen receptor recognition of antigen-presenting molecules. Annu. Rev. Immunol. 33, 169–200 (2015), https://doi.org/10.1146/annurev-immunol-032414-112334.
3
P. J. Bjorkman et al., Structure of the human class I histocompatibility antigen, HLA-A2. Nature 329, 506–512 (1987), https://doi.org/10.1038/329506a0.
4
D. J. Barker et al., The IPD-IMGT/HLA database. Nucleic Acids Res. 51, D1053–D1060 (2023), https://doi.org/10.1093/nar/gkac1011.
5
J. Robinson et al., Distinguishing functional polymorphism from random variation in the sequences of >10,000 HLA-A, -B and -C alleles. PLoS Genet 13, e1006862 (2017), https://doi.org/10.1371/journal.pgen.1006862.
6
R. Gowthaman, B. G. Pierce, TCR3d: The T cell receptor structural repertoire database. Bioinformatics 35, 5323–5325 (2019), https://doi.org/10.1093/bioinformatics/btz517.
7
S. Gupta, S. Nerli, S. Kutti Kandy, G. L. Mersky, N. G. Sgourakis, HLA3DB: Comprehensive annotation of peptide/HLA complexes enables blind structure prediction of T cell epitopes. Nat. Commun. 14, 6349. (2023), https://doi.org/10.1038/s41467-023-42163-z.
8
G. F. Papadaki et al., Decoupling peptide binding from T cell receptor recognition with engineered chimeric MHC-I molecules. Front. Immunol. 14, 1116906 (2023), https://doi.org/10.3389/fimmu.2023.1116906.
9
M. G. Rudolph, R. L. Stanfield, I. A. Wilson, How TCRs bind MHCs, peptides, and coreceptors. Annu. Rev. Immunol. 24, 419–466 (2006).
10
S. Gras et al., Reversed T cell receptor docking on a major histocompatibility class I complex limits involvement in the immune response. Immunity 45, 749–760 (2016), https://doi.org/10.1016/j.immuni.2016.09.007.
11
L. Sušac et al., Structure of a fully assembled tumor-specific T cell receptor ligated by pMHC. Cell 185, 3201–3213.e19 (2022), https://doi.org/10.1016/j.cell.2022.07.010.
12
Y. Sun et al., Structural principles of peptide-centric chimeric antigen receptor recognition guide therapeutic expansion. Sci. Immunol. 8, eadj5792 (2023), https://doi.org/10.1126/sciimmunol.adj5792.
13
Y. Sun et al., CryoEM structure of an MHC-I/TAPBPR peptide-bound intermediate reveals the mechanism of antigen proofreading. Proc. Natl. Acad. Sci. U.S.A. 122, e2416992122 (2025).
14
T. P. Riley, N. K. Singh, B. G. Pierce, Z. Weng, B. M. Baker, “Computational modeling of T cell receptor complexes” in Computational Design of Ligand Binding Proteins. Methods in Molecular Biology, B. L. Stoddard, Ed. (Springer, New York, NY, 2016), vol. 1414, pp. 319–340, https://doi.org/10.1007/978-1-4939-3569-7_19.
15
Y. Patskovsky et al., Molecular mechanism of phosphopeptide neoantigen immunogenicity. Nat. Commun. 14, 3763 (2023), https://doi.org/10.1038/s41467-023-39425-1.
16
N. G. Sgourakis et al., A novel MHC-I surface targeted for binding by the MCMV M06 immunoevasin revealed by solution NMR. J. Biol. Chem. 290, 28857–28868 (2015), https://doi.org/10.1074/jbc.M115.689661.
17
L. Varani et al., Solution mapping of T cell receptor docking footprints on peptide-MHC. Proc. Natl. Acad. Sci. U.S.A. 104, 13080–13085 (2007). https://doi.org/10.1073/pnas.0703702104.
18
W. L. White et al., Design of solubly expressed miniaturized SMART MHCs. bioRxiv [Preprint] (2025). https://doi.org/10.1101/2025.03.14.643101 (Accessed 14 March 2025).
19
A. C. McShan et al., Conformational plasticity of RAS Q61 family of neoepitopes results in distinct features for targeted recognition. Nat. Commun. 14, 8204 (2023), https://doi.org/10.1038/s41467-023-43654-9.
20
J. Ma et al., Dynamic allostery in the peptide/MHC complex enables TCR neoantigen selectivity. Nat. Commun. 16, 849 (2025), https://doi.org/10.1038/s41467-025-56004-8.
21
D. N. Garboczi, D. T. Hung, D. C. Wiley, HLA-A2-peptide complexes: Refolding and crystallization of molecules expressed in Escherichia coli and complexed with single antigenic peptides. Proc. Natl. Acad. Sci. U.S.A. 89, 3429–3433 (1992), https://doi.org/10.1073/pnas.89.8.3429.
22
D. S. Wishart, B. D. Sykes, F. M. Richards, The chemical shift index: A fast and simple method for the assignment of protein secondary structure through NMR spectroscopy. Biochemistry 31, 1647–1651 (1992), https://doi.org/10.1021/bi00121a010.
23
C. H. Woodward, N. G. Sgourakis, SMART A*02:01 in complex with NYESOV peptide. Biological Magnetic Resonance Bank (BMRB). https://bmrb.io/data_library/summary/index.php?bmrbId=52878. Deposited 21 January 2025.
24
C. H. Woodward, N. G. Sgourakis, Backbone assignments for SMART A*02:01 in complex with TAX9 peptide. Biological Magnetic Resonance Bank (BMRB). https://bmrb.io/data_library/summary/index.php?bmrbId=52879. Deposited 21 January 2025.
25
Y. Sun et al., Universal open MHC-I molecules for rapid peptide loading and enhanced complex stability across HLA allotypes. Proc. Natl. Acad. Sci. U.S.A. 120, e2304055120 (2023), https://doi.org/10.1073/pnas.2304055120.
26
A. C. McShan et al., Molecular determinants of chaperone interactions on MHC-I for folding and antigen repertoire selection. Proc. Natl. Acad. Sci. U.S.A. 116, 25602–25613 (2019). https://doi.org/10.1073/pnas.1915562116.
27
Y. Shen, A. Bax, Protein structural information derived from NMR chemical shift with the neural network program TALOS-N. Methods Mol. Biol. 1260, 17–32 (2015), https://doi.org/10.1007/978-1-4939-2239-0_2.
28
D. R. Madden, D. N. Garboczi, D. C. Wiley, The antigenic identity of peptide-MHC complexes: A comparison of the conformations of five viral peptides presented by HLA-A2. Cell 75, 693–708 (1993), https://doi.org/10.1016/0092-8674(93)90490-H.
29
A. I. Webb et al., Functional and structural characteristics of NY-ESO-1-related HLA A2-restricted epitopes and the design of a novel immunogenic analogue. J. Biol. Chem. 279, 23438–23446 (2004), https://doi.org/10.1074/jbc.M314066200.
30
W. F. Hawse et al., Peptide modulation of class I major histocompatibility complex protein molecular flexibility and the implications for immune recognition. J. Biol. Chem. 288, 24372–24381 (2013), https://doi.org/10.1074/jbc.M113.490664.
31
R. J. Mallis, H. Arthanari, M. J. Lang, E. L. Reinherz, G. Wagner, NMR-directed design of pre-TCRβ and pMHC molecules implies a distinct geometry for pre-TCR relative to αβTCR recognition of pMHC. J. Biol. Chem. 293, 754–766 (2018), https://doi.org/10.1074/jbc.M117.813493.
32
Y. Song et al., High-resolution comparative modeling with RosettaCM. Structure 21, 1735–1742 (2013), https://doi.org/10.1016/j.str.2013.08.005.
33
D. K. Cole et al., Increased peptide contacts govern high affinity binding of a modified TCR whilst maintaining a native pMHC docking mode. Front. Immunol. 4, 168 (2013), https://doi.org/10.3389/fimmu.2013.00168.
34
C. H. Woodward, N. G. Sgourakis, Backbone assignments for SMART A*02:01 in complex with TAX9 peptide and the A6c134 T cell receptor. Biological Magnetic Resonance Bank (BMRB). https://bmrb.io/data_library/summary/index.php?bmrbId=52880. Deposited 21 January 2025.
35
H. Du et al., Targeting peptide antigens using a multiallelic MHC I-binding system. Nat. Biotechnol. 1–11 (2024), https://doi.org/10.1038/s41587-024-02505-8.
36
S. Grzesiek, A. Bax, Improved 3D triple-resonance NMR techniques applied to a 31 kDa protein. J. Magn. Reson. (1969) 96, 432–440 (1992), https://doi.org/10.1016/0022-2364(92)90099-S.
37
P. Rossi, Y. Xia, N. Khanra, G. Veglia, C. G. Kalodimos, 15N and 13C-SOFAST-HMQC editing enhances 3D-NOESY sensitivity in highly deuterated, selectively [1H,13C]-labeled proteins. J. Biomol. NMR 66, 259–271 (2016), https://doi.org/10.1007/s10858-016-0074-5.
38
P. Bradley, Structure-based prediction of T cell receptor:peptide-MHC interactions. eLife 12, e82813 (2023), https://doi.org/10.7554/eLife.82813.
39
S. Nerli, A. C. McShan, N. G. Sgourakis, Chemical shift-based methods in NMR structure determination. Prog. Nucl. Magn. Reson. Spectrosc. 106–107, 1–25 (2018), https://doi.org/10.1016/j.pnmrs.2018.03.002.
40
M. Culka et al., Predicting specificity of TCR-pMHC interactions using machine learning and biophysical models. bioRxiv [Preprint] (2025). https://doi.org/10.1101/2025.04.04.647165 (Accessed 7 April 2025).
41
J. Wohlwend et al., Boltz-1 democratizing biomolecular interaction modeling. bioRxiv [Preprint] (2024). https://doi.org/10.1101/2024.11.19.624167 (Accessed 6 May 2025).
Information & Authors
Information
Published in
Classifications
Copyright
Copyright © 2025 the Author(s). Published by PNAS. This open access article is distributed under Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND).
Data, Materials, and Software Availability
All assignments were deposited in BMRB under accession codes 52878 (SMART A*02:01/NYESOV) (23), 52879 (SMART A*02:01/TAX9) (24), and 52880 (SMART A*02:01/TAX9/A6c134) (34). Extended methods are included in SI Appendix. All other data are included in the manuscript and/or SI Appendix.
Submission history
Received: March 17, 2025
Accepted: May 1, 2025
Published online: June 9, 2025
Published in issue: June 17, 2025
Keywords
Acknowledgments
We acknowledge the biomolecular NMR facility at the Eldridge Reeves Johnson Foundation at the University of Pennsylvania, Grants R01AI143997 (N.G.S.), R35GM125034 (N.G.S.), U01DK112217 (N.G.S.), T32GM132039 (NIH training grant supporting C.H.W), NIH R01AI103867 (K.C.G.) and HHMI (K.C.G and D.B.), The Asplundh Foundation, and The Children’s Hospital of Philadelphia Cell and Gene Therapy Collaborative. This work is part of the MATCHMAKERS and NexTGen teams, supported by the Cancer Grand Challenges partnership financed by CRUK (CGCATF-2023/100004, CGCATF-2023/100006, CGCATF-2023/100008, and CGCATF-2021/100002), the National Cancer Institute (OT2CA297575, OT2CA297242, OT2CA297288, and CA278687-01), and The Mark Foundation for Cancer Research. All authors are members of the MATCHMAKERS team. N.G.S. is also a member of the NexTGen team.
Author contributions
C.H.W., A.C., X.T.C., W.L.W., K.C.G., D.B., and N.G.S. designed research; C.H.W., A.C., X.T.C., and W.L.W. performed research; C.H.W., A.C., X.T.C., and W.L.W. contributed new reagents/analytic tools; C.H.W., A.C., X.T.C., and W.L.W. analyzed data; and C.H.W., K.C.G., D.B., and N.G.S. wrote the paper.
Competing interests
The authors declare no competing interest.
Authors
Metrics & Citations
Metrics
Altmetrics
Citations
Cite this article
Solution mapping of MHC-I:TCR interactions using a minimalistic protein system, Proc. Natl. Acad. Sci. U.S.A.
122 (24) e2506016122,
https://doi.org/10.1073/pnas.2506016122
(2025).
Copied!
Copying failed.
Export the article citation data by selecting a format from the list below and clicking Export.
View Options
View options
PDF format
Download this article as a PDF file
DOWNLOAD PDFLogin options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginRecommend to a librarian
Recommend PNAS to a LibrarianPurchase options
Purchase this article to access the full text.