Tailoring nanoparticle platforms for targeted delivery of therapeutic molecules is of considerable interest. Effective platforms necessitate in vitro cargo encapsulation, target recognition, triggered disassembly, and controlled release within the cell. While several self-assembling protein nanoparticles with customized structures have been designed, they often consist of only one or two static building blocks, and their adaptation for cargo packaging and delivery remains in its early stages. Antibodies are particularly attractive targeting moieties, and several studies have explored their incorporation into nanoparticle delivery platforms. The authors previously reported the computational design of antibody-incorporating nanoparticles using a designed homooligomer to assemble the antibody into bounded, multivalent, polyhedral architectures. However, these nanoparticles possessed large pores, hindering efficient cargo packaging and retention. This research aimed to address this limitation by computationally designing pH-dependent 'plugs' to create non-porous nanoparticles for targeted drug delivery. The goal was to create a system that encapsulates molecular cargoes, selectively enters target cells via antibody-mediated targeting, and disassembles in the acidic environment of the endosome or tumor microenvironment to release its payload. The use of multiple, distinct protein components, each performing a unique function, offers a highly modular and versatile platform for targeted drug delivery.

The introduction extensively cites previous research on targeted drug delivery using nanoparticles. The authors review various self-assembling protein nanoparticles, highlighting limitations in their cargo-packaging and delivery capabilities. They also discuss existing methods for incorporating antibodies into nanoparticle delivery systems, setting the stage for their novel approach. The review emphasizes the need for a system with improved cargo retention and tunable pH-responsive disassembly, features that are lacking in currently available platforms.

The researchers employed a multi-step computational design process. First, they extended the previously designed pH-dependent trimer by fusing combinations of helical-repeat-protein building blocks. This generated a large number of C3-symmetric fusion proteins with varying geometries. These were docked into the three-fold symmetric pores of the octahedral antibody nanoparticles (O42.1) using a protein-docking approach, sampling translational and rotational degrees of freedom. The resulting 'plugged' octahedral assemblies (O432) were evaluated for designability using the residue pair transform (RPX) score and shape complementarity at the interface. Sequence optimization was performed using Rosetta sequence design calculations for the interfaces between the pH-dependent plugs and the nanoparticle. The designed trimers and tetramers were expressed, purified, and characterized by SDS-PAGE and native PAGE. For in vitro assembly, the three purified components (trimeric plug, designed tetramer, and Fc domain of human IgG1 or full-length antibody) were mixed, and assembly was confirmed using SEC, DLS, and NS-EM. Cryo-EM was used to determine the high-resolution structure of the assembled nanoparticle, revealing a close match between the experimental density map and the computational design model. To achieve cargo packaging, variants with positively and negatively charged interiors were generated by modifying the interior surface residues of the trimeric plug. Packaging of nucleic acids (pegRNA) and a positively charged GFP variant (pos36GFP) was assessed using non-denaturing electrophoresis and SEC. pH-dependent disassembly was investigated using flow cytometry by monitoring the release of fluorescently labeled plug trimers and antibody Fc-GFP fusion proteins at various pH values. Finally, the ability of the nanoparticles to enter cells via receptor-mediated endocytosis was evaluated using epifluorescence microscopy.

The authors successfully designed and experimentally validated non-porous, pH-responsive antibody nanoparticles (O432). These nanoparticles are composed of three distinct components: a pH-sensitive trimer acting as a plug, a tetramer driving assembly, and an antibody for targeting. Cryo-EM analysis confirmed the accurate assembly of the three components, with a close match to the computational design model. The largest pore size in the O432 nanoparticles was reduced from 13 nm in the original design to 3 nm, significantly enhancing cargo retention. The designed nanoparticles exhibited cooperative assembly from independently purified components, simplifying the preparation process. Both nucleic acid (pegRNA) and protein (pos36GFP) cargoes were effectively packaged and protected from enzymatic degradation within the nanoparticle. The pH-dependent disassembly was tunable, with pH values ranging from 5.9 to 6.7, depending on the design of the trimeric plug. Modifications to the histidine-containing hydrogen bond networks and hydrophobic interactions within the trimer allowed for fine-tuning of the disassembly pH. Antibody-mediated targeting and cellular uptake through receptor-mediated endocytosis were demonstrated using EGFR-targeting antibodies and A431 cells. Significant internalization of the antibody-conjugated nanoparticles was observed, while the non-targeted control showed minimal uptake.

The results demonstrate a significant advance in the design and engineering of pH-responsive protein nanoparticles. The modular design, with distinct functional roles for each component, allows for facile modification and optimization. The ability to tune the pH of disassembly and to incorporate various antibodies offers broad applicability for targeted drug delivery. The close agreement between the computational design and the experimentally determined cryo-EM structure highlights the accuracy of the design methodology. The non-porous nature of the nanoparticles significantly improves cargo retention compared to previous designs. The pH-dependent release of the cargo mimics viral entry mechanisms, offering a promising platform for delivery into the tumor microenvironment, while the tunability of the system allows for tailoring to different biological contexts. The findings represent a significant step towards developing versatile and effective nanocarriers for therapeutic molecules.

This study presents a novel approach for designing non-porous, pH-responsive antibody nanoparticles. The authors successfully designed and validated a three-component system with tunable pH-dependent disassembly and efficient cargo packaging. This versatile platform holds significant promise for targeted drug delivery applications, particularly for cancer therapeutics. Future work could focus on incorporating endosomal escape mechanisms to further enhance the efficacy of the system.

While the study demonstrates significant progress, some limitations exist. The current design may not be suitable for all types of cargoes. Further optimization might be required to accommodate larger molecules or to achieve complete cargo release at specific physiological pH values. Additionally, the study primarily focuses on in vitro assays; further in vivo studies are needed to fully assess the therapeutic potential of these nanoparticles. Finally, while the authors demonstrate pH-dependent disassembly, the kinetics of release warrant further investigation.