Customizable Click Biochemistry Strategy for the Design and Preparation of Glucagon-like Peptide-1 Conjugates and Coagonists.

The development of oligomeric glucagon-like peptide-1 (GLP-1) and GLP-1-containing coagonists holds promise for enhancing the therapeutic potential of the GLP-1-based drugs for treating type 2 diabetes mellitus (T2DM). Here, we report a facile, efficient, and customizable strategy based on genetically encoded SpyCatcher-SpyTag chemistry and an inducible, cleavable self-aggregating tag (icSAT) scheme. icSAT-tagged SpyTag-fused GLP-1 and the dimeric or trimeric SpyCatcher scaffold were designed for dimeric or trimeric GLP-1, while icSAT-tagged SpyCatcher-fused GLP-1 and the icSAT-tagged SpyTag-fused GIP were designed for dual GLP-1/GIP (glucose-dependent insulinotropic polypeptide) receptor agonist. These SpyCatcher- and SpyTag-fused protein pairs were spontaneously ligated directly from the cell lysates. The subsequent icSAT scheme, coupled with a two-step standard column purification, resulted in target proteins with authentic N-termini, with yields ranging from 35 to 65 mg/L and purities exceeding 99%. In vitro assays revealed 3.0- to 4.1-fold increased activities for dimeric and trimeric GLP-1 compared to mono-GLP-1. The dual GLP-1/GIP receptor agonist exhibited balanced activity toward the GLP-1 receptor or the GIP receptor. All the proteins exhibited 1.8- to 3.0-fold prolonged half-lives in human serum compared to mono-GLP-1 or GIP. This study provides a generally applicable click biochemistry strategy for developing oligomeric or dual peptide/protein-based drug candidates.

DeepMineLys: Deep mining of phage lysins from human microbiome.

Vast shotgun metagenomics data remain an underutilized resource for novel enzymes. Artificial intelligence (AI) has increasingly been applied to protein mining, but its conventional performance evaluation is interpolative in nature, and these trained models often struggle to extrapolate effectively when challenged with unknown data. In this study, we present a framework (DeepMineLys [deep mining of phage lysins from human microbiome]) based on the convolutional neural network (CNN) to identify phage lysins from three human microbiome datasets. When validated with an independent dataset, our method achieved an F1-score of 84.00%, surpassing existing methods by 20.84%. We expressed 16 lysin candidates from the top 100 sequences in E. coli, confirming 11 as active. The best one displayed an activity 6.2-fold that of lysozyme derived from hen egg white, establishing it as the most potent lysin from the human microbiome. Our study also underscores several important issues when applying AI to biology questions. This framework should be applicable for mining other proteins.