Protein conformational dynamics often govern protein–protein interaction mechanisms and function, as flexibility enables adaptation to different environments. Yet, the extent to which current physics‑based and emerging AI‑driven methods can predict flexibility‑dependent physicochemical properties remains to be determined. Here, we studied the HyHEL10 (Hy10) antibody lineage binding to lysozymes with different conformational flexibility. Contrary to traditional models, the affinity-matured Hy104X antibody achieves high affinity by actively destabilizing its CDRH1 loop, enabling the capture of mobile antigen variants via entropy-buffered conformational selection. This structural plasticity presents a significant challenge to standard computational protocols. We found that the polarization effect is enhanced with increasing protein flexibility. Our benchmarking of MM/PBSA and AI-driven structure predictors reveals that both approaches initially fail to accurately rank affinities in more flexible systems, primarily due to sampling limitations and inaccurate structure prediction. However, the inclusion of structural templates in Chai-1 effectively mitigated these limitations, leading to enhanced correlation with experimental affinities. In the future, by coupling AI’s quick screening with robust physics-based evaluations, we may have a potential pathway for in silico engineering of therapeutic antibodies against the most elusive, shapeshifting targets in the proteome.
Fluorescence molecular imaging using fluorescently labeled antibodies enables identification of target cells and provides pharmacological insight into local drug distribution. The current GMP manufacturing method of tracers using PD-10 columns is subject to methodological constraints when handling low antibody concentrations, large quantities, or incompatible antibodies. This study presents Tangential Flow Filtration (TFF) as a novel and scalable approach to manufacture clinical-grade fluorescent tracers and demonstrates its applicability using cetuximab-800CW and ustekinumab-800CW. TFF system settings were optimized for purification of cetuximab-800CW and ustekinumab-800CW with labeling parameters and quality control similar to tracer production with PD-10 columns. Cetuximab-800CW production was successfully transitioned from PD-10 to TFF. QC-data after manufacturing with TFF (n = 7) or PD-10 (n = 21) were compared at release and up to 24 months of storage at 2–8°C. This revealed TFF purification to be more efficient and robust, with label efficiencies of 93.5 ± 2.3% with TFF compared to 68.3 ± 6.9% with PD-10, and 0.0 ± 0.1% unconjugated dye with TFF compared to 0.5 ± 0.3% with PD-10. Stability data were comparable for both methods. TFF manufacturing yielded clinical-grade cetuximab-800CW batches stable for at least 24 months. Concentrations ranged from 0.99–1.08 mg/mL and target binding affinity from 65–98%. Detected free dye and aggregates remained well below 3.0%. Moreover, ustekinumab-800CW was successfully manufactured using TFF after PD-10 manufacturing failed due to particle formation. Ustekinumab-800CW was stable up to 24 months at 2–8°C. Together, this demonstrates the feasibility of using TFF for clinical-grade fluorescent tracer manufacturing under GMP conditions.
Accurate mapping of paratopes and epitopes at antibody-antigen (Ab-Ag) interfaces is essential for the rational design of therapeutic antibodies in early drug development. Traditional methods for determining high-resolution Ab-Ag complex structures, such as X-ray crystallography and cryo-electron microscopy, are labor-intensive and require substantial time and resources. Although recent computational approaches have increased throughput, prediction accuracy and reliability remain limited, particularly without experimental structural constraints. Here, we present a hydrogen-deuterium exchange mass spectrometry (HDX-MS)-guided structural modeling workflow for efficient and accurate epitope and paratope mapping across three prospective case studies in therapeutic protein discovery and engineering. Across these systems, peptide-level HDX-MS data were sufficient to guide docking and structural modeling, whereas targeted electron transfer dissociation (ETD) HDX-MS was applied in one system as a higher-resolution, orthogonal approach to further localize critical binding residues. The workflow was applied to diverse Ab-Ag architectures spanning Fab and VHH binders and homodimeric and heterodimeric antigen ectodomains, and HDX-MS data were integrated with HADDOCK, AlphaFold2, and Protenix to generate Ab-Ag complex models. Comparison with subsequently obtained in-house X-ray crystal structures, used here as orthogonal high-resolution reference structures, showed strong agreement in overall binding orientation and interface placement. These results demonstrate the robustness and practical utility of this integrative approach for structure-informed antibody engineering in therapeutic protein discovery.
In this work, we developed bispecific antibody (bsAb)-derived surrogate agonists which mimic the function of IL-21 by targeting the IL-21 receptor composed of IL-21 R (CD360) and IL-2 Rγ (CD132). For this, antigen-specific VHHs (variable domains of the heavy chain of heavy-chain-only antibodies) were obtained by immunization of camelids and isolated using yeast surface display. Combinatorial reformatting of IL-21 R-specific single‑domain antibodies (sdAbs) and IL-2 Rγ-targeting paratopes into a monovalent bispecific antibody architecture enabled the identification of IL-21 mimetics displaying attenuated capacities in triggering STAT3 phosphorylation compared to the wild-type cytokine as demonstrated in NK-92 cells as well as peripheral blood mononuclear cells (PBMCs). Moreover, by applying different protein engineering strategies, we demonstrate that agonism capacities of the generated IL-21 mimetics, such as pSTAT3 induction or Granzyme B expression of cytotoxic T cells, can be significantly optimized. For this, framework mutations were introduced to engineer VHH:VHH interactions within the bispecific sdAb-Fc fusion geometry for a more rigid receptor targeting. Furthermore, we show that antibody format engineering, in which the VHHs were arranged in an IgG-like scaffold that replaces the conventional IgG VH and VL domains with the corresponding VHHs, combined with rigidifying mutations, enables IL-21 R agonism comparable to the wild-type cytokine. Taken together, these findings show that IL-21 receptor agonism can be substantially optimized by adapting the spatial orientation of paratopes targeting both receptor subunits via forced dimerization, without altering paratope valencies.
Redirection of T lymphocytes toward cancer cells has been one of the most promising treatment concepts in oncology developed over the past years and leading to multiple regulatory approvals for both adoptive cell therapy (ACT) and T-cell-engaging bispecific (TEB) molecules. However, progress has been achieved predominantly in hematological indications by aiming at so-called lineage antigens (i.e. CD19, CD20, BCMA). Targeting of peptide-HLA antigens (pHLA) has been proposed as a strategy to overcome the paucity of suitable surface antigens in solid cancers, yet the technical hurdle to specifically address this class of low copy and highly promiscuous antigens has so far hampered its use in a broader patient population. We here describe the successful development of TCER (T-Cell-Engaging Receptor), a novel and highly versatile class of T-cell receptor (TCR)-based TEB with optimized in vivo efficacy, tolerability, plasma half-life, and stability characteristics for a broad use targeting pHLA in oncology and beyond.
