Current antibody-based immunotherapy depends on tumor antigen shedding for proper T cell priming. Here we select a novel human CD40 agonistic drug candidate and generate a bispecific antibody, herein named BiA9*2_HF, that allows for rapid antibody-peptide conjugate formation. The format is designed to facilitate peptide antigen delivery to CD40 expressing cells combined with simultaneous CD40 agonistic activity. In vivo, the selected bispecific antibody BiA9*2_HF loaded with peptide cargos induces improved antigen-specific proliferation of CD8+ (10-15 fold) and CD4+ T cells (2-7 fold) over control in draining lymph nodes. In both virus-induced and neoantigen-based mouse tumor models, BiA9*2_HF demonstrates therapeutic efficacy and elevated safety profile, with complete tumor clearance, as well as measured abscopal impact on tumor growth. The BiA9*2_HF drug candidate can thus be utilized to tailor immunotherapeutics for cancer patients.

G3BP is the central node within stress-induced protein-RNA interaction networks known as stress granules (SGs). The SG-associated proteins Caprin-1 and USP10 bind mutually exclusively to the NTF2 domain of G3BP1, promoting and inhibiting SG formation, respectively. Herein, we present the crystal structure of G3BP1-NTF2 in complex with a Caprin-1-derived short linear motif (SLiM). Caprin-1 interacts with His-31 and His-62 within a third NTF2-binding site outside those covered by USP10, as confirmed using biochemical and biophysical-binding assays. Nano-differential scanning fluorimetry revealed reduced thermal stability of G3BP1-NTF2 at acidic pH. This destabilization was counterbalanced significantly better by bound USP10 than Caprin-1. The G3BP1/USP10 complex immunoprecipated from human U2OS cells was more resistant to acidic buffer washes than G3BP1/Caprin-1. Acidification of cellular condensates by approximately 0.5 units relative to the cytosol was detected by ratiometric fluorescence analysis of pHluorin2 fused to G3BP1. Cells expressing a Caprin-1/FGDF chimera with higher G3BP1-binding affinity had reduced Caprin-1 levels and slightly reduced condensate sizes. This unexpected finding may suggest that binding of the USP10-derived SLiM to NTF2 reduces the propensity of G3BP1 to enter condensates.

The agonistic potentials of therapeutic anti-CD40 antibodies have been profiled in relation to antibody isotype and epitope specificity. Still, clinical impact relies on a well-balanced clinical efficacy versus target-mediated toxicity. As CD40-mediated immune activation must rely on a combination of stimulation of antigen-presenting cells (APCs) alongside antigen presentation, for efficient T cell priming, alternative approaches to improve the therapeutic outcome of CD40-targeting strategies should focus on providing optimal antigen presentation together with CD40 stimulation. Herein, a bispecific antibody targeting CD40 as a means to deliver cargo (i.e., synthetic peptides) into APCs through a non-covalent, high-affinity interaction between the antibody and the cargo peptide, further referred to as the Adaptable Drug Affinity Conjugate (ADAC) technology, has been developed. The ADAC platform demonstrated a target-specific CD4+ and CD8+ T cell expansion in vitro and significantly improved peptide-specific CD8+ T cell proliferation in vivo. In addition, the strategy dramatically improved the in vitro and in vivo half-life of the synthetic peptides. Future applications of ADAC involve pandemic preparedness to viral genetic drift as well as neoepitope vaccination strategies where the bispecific antibody is an off-the-shelf product, and the peptide antigen is synthesized based on next-generation sequencing data mining. 

The transition from preclinical biological drug development into clinical trials requires an efficient upscaling process. In this context, bispecific antibody drugs are particularly challenging due to their propensity to form aggregates and generally produce low titers. Here, the upscaling process for a tetravalent bispecific antibody expressed by a piggyBac transposon-mediated stable HEK293 cell pool has been evaluated. The project was performed as a case study at Testa Center, a non-GMP facility for scale-up testing of biologics in Sweden, and encompassed media adaptation strategies, fed-batch optimization and a novel antibody purification technology. The cell pool was adapted to different culture media for evaluation in terms of cell viability and titers compared to its original Expi293 Expression Medium. These parameters were assessed in both sequential stepwise adaption and direct media exchanges. By this, a more affordable medium was identified that did not require stepwise adaptation and with similar titers and viability as in the Expi293 Expression Medium. Fed-batch optimizations resulted in culture densities reaching up to 20 x 106 viable cells/mL with over 90 % viability 12 days postinoculum, and antibody titers three times higher than corresponding batch cultures. By implementing a novel high-speed protein A fiber technology (Fibro PrismA) with a capture residence time of only 7.5 s, 8 L of supernatant could be purified in 4.5 h without compromising the purity, structural integrity and function of the bispecific antibody. Results from this study related to medium adaptation and design of fed-batch protocols will be highly beneficial during the forthcoming scale-up of this therapeutic antibody.

The interactions between proteins and biological membranes are important for drug development, but remain notoriously refractory to structural investigation. We combine non-denaturing mass spectrometry (MS) with molecular dynamics (MD) simulations to unravel the connections among cofactor, lipid, and inhibitor binding in the peripheral membrane protein dihydroorotate dehydrogenase (DHODH), a key anticancer target. Interrogation of intact DHODH complexes by MS reveals that phospholipids bind via their charged head groups at a limited number of sites, while binding of the inhibitor brequinar involves simultaneous association with detergent molecules. MD simulations show that lipids support flexible segments in the membrane-binding domain and position the inhibitor and electron acceptor-binding site away from the membrane surface, similar to the electron acceptor-binding site in respiratory chain complex I. By complementing MS with MD simulations, we demonstrate how a peripheral membrane protein uses lipids to modulate its structure in a similar manner as integral membrane proteins.