frontier-banner
Frontiers
Home>Frontiers>

Nature Communications | Generation of Mixed-Valency, Modular Multispecific Antibodies via Disulfide-Linked Fc–FcγR Complexes

Nature Communications | Generation of Mixed-Valency, Modular Multispecific Antibodies via Disulfide-Linked Fc–FcγR Complexes
--

This study presents an innovative strategy for designing highly stable and modular multispecific antibodies, greatly simplifying traditional complex antibody engineering processes. It offers a scalable platform for developing novel T-cell engagers targeting tumor immune evasion mechanisms.

 

Literature Overview

The paper titled 'Generation of mixed-valency, modular multispecific antibodies using disulfide-linked Fc–FcγR complexes,' published in Nature Communications, systematically explores how the natural interaction between IgG Fc and FcγR can be leveraged to achieve stable covalent linkage via a single symmetric disulfide bond, enabling the construction of functionally diverse and easily producible multispecific antibodies. This strategy not only circumvents the chain-mismatch issues common in traditional bispecific antibodies but also retains the long serum half-life advantage of IgG. The study further demonstrates the platform's potential in T-cell engagement, co-stimulatory signal integration, and conditional cytokine release.

Background Knowledge

Currently, although several bispecific antibodies have been approved for hematologic malignancies—such as blinatumomab targeting CD19 and CD3—most adopt a monovalent binding mode, limiting their ability to discriminate based on target antigen density and increasing the risk of on-target, off-tissue toxicity. Additionally, most platforms rely on complex CH3 domain engineering (e.g., knobs-into-holes) or additional Fc-silencing mutations (e.g., LALA, STR) to prevent binding to endogenous FcγRs, significantly increasing development complexity. This study ingeniously exploits the naturally high-affinity interface between FcγRIIIa and IgG1 Fc, introducing site-specific A330C and I106C mutations to form a disulfide bond, enabling a plug-and-play modular assembly. This design not only avoids chain mispairing but also inherently blocks the FcγR binding site without requiring additional mutations. Moreover, the system allows independent fusion of different functional domains—such as scFv, IL-2, or CD28—to the N- and C-termini of FcγR, enabling integration of multiple signals. The research addresses key bottlenecks in current platforms related to manufacturing complexity, valency control, and functional expandability, offering a new path for next-generation immunotherapies targeting antigen heterogeneity in solid tumors.

 

 

Research Methods and Experiments

The authors employed an ExpiCHO cell co-expression system, co-transfecting IgG1 Fc carrying the A330C mutation with soluble FcγRIIIa carrying the I106C mutation, followed by purification using Protein G affinity chromatography and size-exclusion chromatography to obtain a stable disulfide-linked complex. The complex showed the expected molecular weight shift in non-reducing SDS-PAGE and remained intact over a 7-day serum stability test, indicating excellent in vivo stability. SPR and flow cytometry confirmed that the complex retained bispecific binding to Her2 and CD3, normal binding to FcRn, and effectively blocked binding to FcγRIIIa.

To validate functional activity, the authors constructed an αHer2–αCD3 complex and evaluated T-cell-mediated cytotoxicity in Her2-positive tumor cell lines such as BT474 and SKOV3. Using a Jurkat-NFAT-luciferase reporter system and real-time impedance analysis with primary T cells, they demonstrated that the complex activated T cells and induced potent tumor killing at picomolar concentrations, with activity dependent on bispecific binding. Furthermore, multifunctional complexes incorporating anti-CD28 scFv or IL-2 were constructed, and their synergistic effects with T-cell engagers—exhibiting 'AND' logic gating—were demonstrated.

Key Conclusions and Perspectives

  • The Fc–FcγR complex formed via the A330C–I106C disulfide bond achieves high yield, high purity, and exceptional stability, solving long-standing challenges of chain mispairing and low expression in traditional bispecific antibodies, thus providing a universal platform for rapid construction of multispecific antibodies.
  • The complex exhibits bivalent binding to tumor antigens and monovalent binding to T cells, significantly enhancing selective killing of Her2-high cells while reducing toxicity to low-expressing cells, thereby widening the therapeutic window and offering important guidance for developing engagers targeting solid tumors.
  • Fusing different functional modules (e.g., IL-2, CD28) to both ends of FcγR enables signal synergy, enhancing T-cell activation and persistence, providing a novel strategy to overcome immunosuppression in the tumor microenvironment.
  • The constructed protease-activatable IL-2 prodrug (αHer2–Ta–IL2) releases active IL-2 in the presence of MMP7, enabling tumor microenvironment-specific immune activation and offering a programmable template for conditional immunotherapies.
  • In an AML xenograft model, the αIL1RAP–αCD3 complex significantly suppressed tumor growth and prolonged survival, validating its in vivo antitumor activity and supporting its potential for clinical translation.

Research Significance and Prospects

The 'templated disulfide' strategy introduced in this study offers a highly modular and scalable solution for multispecific antibody engineering, dramatically lowering development barriers. Its preserved FcRn binding ensures a long half-life, reducing dosing frequency and improving patient compliance. More importantly, the platform supports flexible integration of multiple functional modules, laying the foundation for 'smart' immunotherapies (e.g., AND/OR logic gating, conditional activation).

From a drug development perspective, this technology accelerates the screening of multiple target combinations, particularly for solid tumors with high antigen heterogeneity. Simultaneously, its inherent FcγR shielding avoids complex Fc-silencing engineering, streamlining CMC processes. Future applications could include CAR-T cell preconditioning and dual-functional checkpoint blockade.

 

 

Conclusion

This study repurposes the natural interaction between IgG Fc and FcγR to develop a disulfide-linked, modular multispecific antibody platform enabling efficient and stable antibody functionalization. The platform not only overcomes challenges of chain mispairing and pharmacokinetic optimization in traditional bispecific antibody production but also enhances tumor cell selectivity through mixed-valency design, reducing off-target toxicity. Its scalability supports flexible integration of diverse immune-modulating payloads, providing a powerful tool for next-generation 'smart' immunotherapies. Particularly in the treatment of hematologic malignancies and solid tumors, this technology could accelerate the iterative development of T-cell engagers targeting antigens such as CD19, CD33, and IL1RAP. Combined with conditional activation strategies—such as protease-responsive IL-2 release—it further enhances therapeutic safety and precision. From bench to bedside, this platform offers a universal solution for accelerating antibody drug discovery and optimizing efficacy and safety profiles, potentially becoming a cornerstone in the future development of immunotherapies.

 

Reference:
Miso Park, Kevin Ly, Bea Parcutela, Guido Marcucci, and John C Williams. Generation of mixed-valency, modular multispecific antibodies using disulfide-linked Fc–FcγR complexes. Nature Communications.
Post-translational modifications (PTMs) are key regulators of protein function, stability, and interactions, and are critical in cellular signaling, localization, and disease mechanisms. However, experimental identification of PTMs (e.g., mass spectrometry, western blotting, radioactive labeling) is costly and time-consuming, making computational approaches attractive alternatives. Traditional computational models rely only on local sequence features around PTM sites. Many existing pretrained protein language models (PLMs) are sequence-only, lack structural information, and are often single-task, preventing feature sharing across PTM types and limiting knowledge transfer and prediction performance.