Abstract
Immunoglobulin G4 antibodies exhibit unusual properties with important biological consequences. We report the structure of the human full-length IgG4 S228P anti-PD1 antibody pembrolizumab, solved to 2.3-Å resolution. Pembrolizumab is a compact molecule, consistent with the presence of a short hinge region. The Fc domain is glycosylated at the CH2 domain on both chains, but one CH2 domain is rotated 120° with respect to the conformation observed in all reported structures to date, and its glycan chain faces the solvent. We speculate that this new conformation is driven by the shorter hinge. The structure suggests a role for the S228P mutation in preventing the IgG4 arm exchange. In addition, this unusual Fc conformation suggests possible structural diversity between IgG subclasses and shows that use of isolated antibody fragments could mask potentially important interactions, owing to molecular flexibility.
This is a preview of subscription content, access via your institution
Access options
Subscribe to this journal
Receive 12 print issues and online access
$189.00 per year
only $15.75 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
References
Mellman, I., Coukos, G. & Dranoff, G. Cancer immunotherapy comes of age. Nature 480, 480–489 (2011).
Mullard, A. New checkpoint inhibitors ride the immunotherapy tsunami. Nat. Rev. Drug Discov. 12, 489–492 (2013).
Aalberse, R.C. & Schuurman, J. IgG4 breaking the rules. Immunology 105, 9–19 (2002).
van der Neut Kolfschoten, M. et al. Anti-inflammatory activity of human IgG4 antibodies by dynamic Fab arm exchange. Science 317, 1554–1557 (2007).
Jefferis, R. & Lund, J. Interaction sites on human IgG-Fc for FcγR: current models. Immunol. Lett. 82, 57–65 (2002).
Jefferis, R. Isotype and glycoform selection for antibody therapeutics. Arch. Biochem. Biophys. 526, 159–166 (2012).
Kratzin, H.D. et al. The primary structure of crystallizable monoclonal immunoglobulin IgG1 Kol. II. Amino acid sequence of the L-chain, gamma-type, subgroup I. Biol. Chem. Hoppe Seyler 370, 263–272 (1989).
Guddat, L.W., Herron, J.N. & Edmundson, A.B. Three-dimensional structure of a human immunoglobulin with a hinge deletion. Proc. Natl. Acad. Sci. USA 90, 4271–4275 (1993).
Harris, L.J., Skaletsky, E. & McPherson, A. Crystallographic structure of an intact IgG1 monoclonal antibody. J. Mol. Biol. 275, 861–872 (1998).
Saphire, E.O. et al. Crystal structure of a neutralizing human IgG against HIV-1: a template for vaccine design. Science 293, 1155–1159 (2001).
Harris, L.J., Larson, S.B., Hasel, K.W. & McPherson, A. Refined structure of an intact IgG2a monoclonal antibody. Biochemistry 36, 1581–1597 (1997).
Davies, A.M. et al. Structural determinants of unique properties of human IgG4-Fc. J. Mol. Biol. 426, 630–644 (2014).
Radaev, S. & Sun, P. Recognitions of immunoglobulins by Fcγ receptors. Mol. Immunol. 38, 1073–1083 (2002).
Bruhns, P. et al. Specificity and affinity of human Fcγ receptors and their polymorphic variants for human IgG subclasses. Blood 113, 3716–3725 (2009).
Schneider, S. & Zacharias, M. Atomic resolution model of the antibody Fc interaction with the complement C1q component. Mol. Immunol. 51, 66–72 (2012).
Pervushin, K.V., Wider, G. & Wüthrich, K. Single transition-to-single transition polarization transfer (ST2-PT) in [15N,1H]-TROSY. J. Biomol. NMR 12, 345–348 (1998).
Arbogast, L.W., Brinson, R.G. & Marino, J.P. Mapping monoclonal antibody structure by 2D 13C NMR at natural abundance. Anal. Chem. 87, 3556–3561 (2015).
Tian, X. et al. In-depth analysis of subclass-specific conformational preferences of IgG antibodies. IUCrJ. 2, 9–18 (2015).
Angal, S. et al. A single amino acid substitution abolishes the heterogeneity of chimeric mouse/human (IgG4) antibody. Mol. Immunol. 30, 105–108 (1993).
Richardson, J.S. The anatomy and taxonomy of protein structures. Adv. Protein Chem. 34, 167–339 (1981).
Poppe, L. et al. Profiling formulated monoclonal antibodies by 1H NMR spectroscopy. Anal. Chem. 85, 9623–9629 (2013).
Rayner, L.E. et al. The Fab conformations in the solution structure of human immunoglobulin G4 (IgG4) restrict access to its Fc region: implications for functional activity. J. Biol. Chem. 289, 20740–20756 (2014).
Shields, R.L. et al. High resolution mapping of the binding site on human IgG1 for Fc gamma RI, Fc gamma RII, Fc gamma RIII, and FcRn and design of IgG1 variants with improved binding to the Fc gamma R. J. Biol. Chem. 276, 6591–6604 (2001).
Sondermann, P., Huber, R., Oosthuizen, V. & Jacob, U. The 3.2-Å crystal structure of the human IgG1 Fc fragment–FcγRIII complex. Nature 406, 267–273 (2000).
Hamilton, R.G. The Human IgG Subclasses (Calbiochem, 2001).
Tao, M.H., Smith, R.I. & Morrison, S.L. Structural features of human immunoglobulin G that determine isotype-specific differences in complement activation. J. Exp. Med. 178, 661–667 (1993).
Oganesyan, V. et al. Structural insights into neonatal Fc receptor-based recycling mechanisms. J. Biol. Chem. 289, 7812–7824 (2014).
Martin, W.L., West, A.P., Gan, L. & Bjorkman, P.J. Crystal structure at 2.8 Å of an FcRn/hetrodimeric Fc complex: mechanism of pH-dependent binding. Mol. Cell 7, 867–877 (2001).
West, A.P. & Bjorkman, P.J. Crystal structure and immunoglobulin G binding properties of the human major histocompatibility complex-related Fc receptor. Biochemistry 39, 9698–9708 (2000).
Arnold, J.N., Wormald, M.R., Sim, R.B., Rudd, P.M. & Dwek, R.A. The impact of glycosylation on the biological function and structure of human immunoglobulins. Annu. Rev. Immunol. 25, 21–50 (2007).
Quast, I. & Lünemann, J.D. Fc glycan-modulated immunoglobulin G effector functions. J. Clin. Immunol. 34, S51–S55 (2014).
Lux, A. & Nimmerjahn, F. Impact of differential glycosylation on IgG activity. Adv. Exp. Med. Biol. 780, 113–124 (2011).
Barb, A.W. & Prestegard, J.H. NMR analysis demonstrates immunoglobulin G N-glycans are accessible and dynamic. Nat. Chem. Biol. 7, 147–153 (2011).
Yang, X. et al. Comprehensive analysis of the therapeutic IgG4 antibody pembrolizumab: hinge modification blocks half molecule exchange in vitro and in vivo. J. Pharm. Sci. 10.1002/jps.24620 (26 August 2015).
Vonrhein, C. et al. Data processing and analysis with the autoPROC toolbox. Acta Crystallogr. D Biol. Crystallogr. 67, 293–302 (2011).
McCoy, A.J. et al. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674 (2007).
Collaborative Computational Project N4. The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D Biol. Crystallogr. 50, 760–763 (1994).
Murshudov, G.N., Vagin, A.A. & Dodson, E.J. Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr. D Biol. Crystallogr. 53, 240–255 (1997).
Emsley, P., Lohkamp, B., Scott, W. & Cowtan, K. Features and development of Coot. Acta Crystallogr. D Biol. Crystallogr. 66, 486–501 (2010).
Winn, M.D., Isupov, M.N. & Murshudov, G.N. Use of TLS parameters to model anisotropic displacements in macromolecular refinement. Acta Crystallogr. D Biol. Crystallogr. 57, 122–133 (2001).
Acknowledgements
The authors would like to thank J. Chapman and R. Ruzanski for performing receptor binding and capillary electrophoresis assays and M. Hohn for help in generation of the N15 labeled antibody. Research described in this paper was performed at the Canadian Light Source, which is supported by the Canadian Foundation for Innovation, Natural Sciences and Engineering Research Council of Canada, the University of Saskatchewan, the Government of Saskatchewan, Western Economic Diversification Canada, the National Research Council Canada and the Canadian Institutes of Health Research. The X-ray diffraction data were collected by Shamrock, and we thank G. Ranieri, J. Carter and R. Walter for collecting the data.
Author information
Authors and Affiliations
Contributions
G.S. carried out crystallographic experiments and structure determination. X.Y. prepared pembrolizumab and Fab fragments for crystallization and NMR experiments, designed and carried out the deglycosylation experiments, designed the receptor binding experiments and analyzed the resulting data. W.W.P. and P.R. established the crystallization procedures. M.McC. carried out NMR experiments and analyzed the resulting data. J.M.J. performed molecular dynamics simulations and analyzed the resulting data. R.S.K. supervised antibody sourcing and formulation for crystallization studies. G.S. and C.S. wrote the paper, and all authors contributed to editing of the manuscript.
Corresponding author
Ethics declarations
Competing interests
All authors are Merck & Co. Inc. employees, and pembrolizumab is a Merck marketed product called KEYTRUDA®.
Integrated supplementary information
Supplementary Figure 1 Surface representation of the upper linker region in pembrolizumab.
The surface view of the upper linker region in pembrolizumab, including the hinge peptide (residues Val218–Pro225) and the nearby peptides Pro130–Thr138 and Leu196–Gly197 clearly shows the interdigitation occurring in this area. In the IgG1 structures 1HZH (Saphire, E.O., et al., Science. 293, 1155-1159, 2001) and 1IGY (Harris, L.J., et al., J. Mol. Biol. 275, 861-872, 1998) there are no contacts between the two upper linker regions. The compact hinge seen in the crystal structure differs from the extended hinge conformation proposed for the solution structure of human IgG4, both wild type and S228P mutant (Rayner, L.E., et al., J Biol Chem. 289, 20740–20756, 2014). The models generated in the paper to fit the BioSAXs data were calculated from the IgG1 structure, constraining the full hinge region (Val218–Pro238) to be of a minimum length of between 50 and 73.5 Å (or between 22 to 31.5 Å for the Val218–Cys226 peptide), values derived from the 1HZH structure. From the pembrolizumab structure the length (Ca to CA) of the very well ordered Val218–Cys226 peptide is about 13 Å, while the distance between Val218 and Pro238 is 24 Å in one chain and 34 Å in the other, much shorter than the values used for the modeling of the bioSAXs data.
Supplementary Figure 2 Distribution of phi and psi angles for the wild-type and S228P pembrolizumab linker peptides Val218–Gly236.
Top panel: Phi/Psi populations (frequency versus angle) plotted for residues P227 and P(S)228 of the linker region. Phi/Psi populations derived from last 36 nanoseconds of simulations of linker peptide dimers (20 residues, Val218–Gly236), with either a Pro or a Ser at position 228. The missing portions of the linker (230PAP232 in chain B and 230PAPEFL235 in chain G) were built and minimized in the context of the entire protein using the ab initio loop builder functionality of MOE 2013.08. The linker region was extracted and prepared using the prep-wizard in Maestro (Schrodinger) and to generate the WT linker, P228 was mutated back to S. Both WT and S228P pembrolizumab linker dimers were simulated for 48 ns each, in cubic boxes of ~8000 water molecules using the Schrodinger implementation of Desmond molecular dynamics. After the standard restrained equilibration procedure (default settings in Schrodinger 2014-4), the first 12 nanoseconds were also discarded as ‘equilibration’ from both simulations. Bottom panel: Distribution of the S-S distances between residues C226 and C229 during the two simulations for the S228P (left) and S228 (right) peptide. Both S228P and WT (S228) linkers were simulated as single chains, in exactly the same way as described above, but with the Cysteine S atoms protonated. The dual distribution in the S228 peptide suggests that there is a tendency for Cys226 and Cys229 to move closer to each other thus facilitating the formation of an intra-chain disulfide bond.
Supplementary Figure 3 Structural conservation between the Fc domains in pembrolizumab.
A) Overlay of the two CH2 domains observed in pembrolizumab: residues 240–245 and 257–340 of the conventional CH2 (Chain A, yellow trace) and the flipped CH2 (chain B, cyan trace) align with an RMSD on CA of 0.76 Å (1.24 Å if calculated for all atoms). The major difference is for the loop spanning residues 246–256 which in the flipped CH2 assumes a different conformation because of crystal contacts. The sugar residues also align well: the RMSD for the core residues is 0.1 Å. B) Overlay of the glycans in chain A (black carbon), PDB entry 4C54 (yellow carbons, RMSD with chain A is 0.2 Ang) and 1HZH (magenta carbon, rmsd with Chain A is 0.4 Ang).
Supplementary Figure 4 Representative sensorgrams from the single-cycle kinetic SPR measurements.
Left Panel: representative sensorgrams for the single cycle kinetic (SCK) SPR method used to measure the affinity (KD) of pembrolizumab (A) and IgG1 (B) to FcγRI; Right panel: Representative sensorgrams for the multi-cycle steady state affinity method used to measure the affinity of pembrolizumab (C) and IgG1 (D) to FcγRIIAH131
Supplementary information
Supplementary Text and Figures
Supplementary Figures 1–4 and Supplementary Table 1 (PDF 696 kb)
Source data
Rights and permissions
About this article
Cite this article
Scapin, G., Yang, X., Prosise, W. et al. Structure of full-length human anti-PD1 therapeutic IgG4 antibody pembrolizumab. Nat Struct Mol Biol 22, 953–958 (2015). https://doi.org/10.1038/nsmb.3129
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/nsmb.3129
This article is cited by
-
Progress in programmed cell death-1/programmed cell death-ligand 1 pathway inhibitors and binding mode analysis
Molecular Diversity (2023)
-
Systems glycobiology for discovering drug targets, biomarkers, and rational designs for glyco-immunotherapy
Journal of Biomedical Science (2021)
-
Use of checkpoint inhibitors in patients with lymphoid malignancies receiving allogeneic cell transplantation: a review
Bone Marrow Transplantation (2021)
-
Intracellular accumulation of PD-1 molecules in circulating T lymphocytes in advanced malignant melanoma: an implication for immune evasion mechanism
International Journal of Clinical Oncology (2020)
-
Antibody responses to viral infections: a structural perspective across three different enveloped viruses
Nature Microbiology (2019)