Brauer, D. D. | Search bioRxiv

bioRxiv2022

Redirecting RiPP biosynthetic enzymes to proteins and backbone-modified substrates

Ribosomally synthesized and post-translationally modified peptides (RiPPs) are peptide-derived natural products that include the FDA-approved analgesic ziconotide1,2 as well as compounds with potent antibiotic, antiviral, and anticancer properties.3 RiPP enzymes known as cyclodehydratases and dehydrogenases represent an exceptionally well-studied enzyme class.3 These enzymes work together to catalyze intramolecular, interresidue condensation3,4 and aromatization reactions that install oxazoline/oxazole and thiazoline/thiazole heterocycles within ribosomally produced polypeptide chains. Here we show that the previously reported enzymes MicD-F and ArtGox accept backbone-modified monomers, including aramids and beta-amino acids, within leader-free polypeptides, even at positions immediately preceding or following the site of cyclization/dehydrogenation. The products are sequence-defined chemical polymers with multiple, diverse, non-alpha-amino acid subunits. We show further that MicD-F and ArtGox can install heterocyclic backbones within protein loops and linkers without disrupting the native tertiary fold. Calculations reveal the extent to which these heterocycles restrict conformational space; they also eliminate a peptide bond. Both features could improve the stability or add function to linker sequences now commonplace in emerging biotherapeutics. Moreover, as thiazoles and thiazoline heterocycles are replete in natural products,5-7 small molecule drugs,8,9 and peptide-mimetic therapeutics,10 their installation in protein-based biotherapeutics could improve or augment performance, activity, stability, and/or selectivity. This work represents a general strategy to expand the chemical diversity of the proteome beyond and in synergy with what can now be accomplished by expanding the genetic code.

biochemistry↗

bioRxiv2021

Comprehensive Fitness Landscape of a Multi-Geometry Protein Capsid Informs Machine Learning Models of Assembly

Virus-like particles (VLPs) are non-infections viral-derived nanomaterials poised for biotechnological applications due to their well-defined, modular self-assembling architecture. Although progress has been made in understanding the complex effects that mutations may have on VLPs, nuanced understanding of the influence particle mutability has on quaternary structure has yet to be achieved. Here, we generate and compare the apparent fitness landscapes of two capsid geometries (T=3 and T=1 icosahedral) of the bacteriophage MS2 VLP. We find significant shifts in mutability at the symmetry interfaces of the T=1 capsid when compared to the wildtype T=3 assembly. Furthermore, we use the generated landscapes to benchmark the performance of in silico mutational scanning tools in capturing the effect of missense mutation on complex particle assembly. Finding that predicted stability effects correlated relatively poorly with assembly phenotype, we used a combination of de novo features in tandem with in silico results to train machine learning algorithms for the classification of variant effects on assembly. Our findings not only reveal ways that assembly geometry affects the mutable landscape of a self-assembled particle, but also establish a template for the generation of predictive mutational models of self-assembled capsids using minimal empirical training data.

bioengineering↗