AI-guided pipeline for protein-protein interaction drug discovery identifies a SARS-CoV-2 inhibitor

Protein-protein interactions (PPIs) offer great opportunities to expand the druggable proteome and therapeutically tackle various diseases, but remain challenging targets for drug discovery. Here, we provide a comprehensive pipeline that combines experimental and computational tools to identify and validate PPI targets and perform early-stage drug discovery. We have developed a machine learning approach that prioritizes interactions by analyzing quantitative data from binary PPI assays or AlphaFold-Multimer predictions. Using the quantitative assay LuTHy together with our machine learning algorithm, we identified high-confidence interactions among SARS-CoV-2 proteins for which we predicted three-dimensional structures using AlphaFold-Multimer. We employed VirtualFlow to target the contact interface of the NSP10-NSP16 SARS-CoV-2 methyltransferase complex by ultra-large virtual drug screening. Thereby, we identified a compound that binds to NSP10 and inhibits its interaction with NSP16, while also disrupting the methyltransferase activity of the complex, and SARS-CoV-2 replication. Overall, this pipeline will help to prioritize PPI targets to accelerate the discovery of early-stage drug candidates targeting protein complexes and pathways.

Neoleukin-2 enhances anti-tumour immunity downstream of peptide vaccination targeted by an anti-MHC class II VHH

Cancer-specific mutations can lead to peptides of unique sequence presented on MHC class I to CD8 T cells. These neoantigens can be potent tumour-rejection antigens, appear to be the driving force behind responsiveness to anti-CTLA-4 and anti-PD1/L1-based therapies and have been used to develop personalized vaccines. The platform for delivering neoantigen-based vaccines has varied, and further optimization of both platform and adjuvant will be necessary to achieve scalable vaccine products that are therapeutically effective at a reasonable cost. Here, we developed a platform for testing potential CD8 T cell tumour vaccine candidates. We used a high-affinity alpaca-derived VHH against MHC class II to deliver peptides to professional antigen-presenting cells. We show in vitro and in vivo that peptides derived from the model antigen ovalbumin are better able to activate naive ovalbumin-specific CD8 T cells when conjugated to an MHC class II-specific VHH when compared with an irrelevant control VHH. We then used the VHH-peptide platform to evaluate a panel of candidate neoantigens in vivo in a mouse model of pancreatic cancer. None of the candidate neoantigens tested led to protection from tumour challenge; however, we were able to show vaccine-induced CD8 T cell responses to a melanoma self-antigen that was augmented by combination therapy with the synthetic cytokine mimetic Neo2/15.

Generation of Ca2+-independent sortase A mutants with enhanced activity for protein and cell surface labeling

Sortase A, a calcium-dependent transpeptidase derived from Staphylococcus aureus, is used in a broad range of applications, such as the conjugation of fluorescent dyes and other moieties to proteins or to the surface of eukaryotic cells. In vivo and cell-based applications of sortase have been somewhat limited by the large range of calcium concentrations, as well as by the often transient nature of protein-protein interactions in living systems. In order to use sortase A for cell labeling applications, we generated a new sortase A variant by combining multiple mutations to yield an enzyme that was both calcium-independent and highly active. This variant has enhanced activity for both N- and C-terminal labeling, as well as for cell surface modification under physiological conditions.