Birth of new protein folds and functions in the virome

Abstract

Rapid virus evolution generates proteins essential to infectivity and replication but with unknown function due to extreme sequence divergence. Using a database of 67,715 newly predicted protein structures from 4,463 eukaryotic viral species, we found that 62% of viral proteins are evolutionarily young and lack homologs in the Alphafold database. Among the 38% of more ancient viral proteins, many have non-viral structural homologs that revealed surprising similarities between human pathogens and their eukaryotic hosts. Structural comparisons suggested putative functions for >25% of unannotated viral proteins, including those with roles in the evasion of innate immunity. In particular, RNA ligase T- (ligT) like phosphodiesterases were found to resemble phage-encoded proteins that hydrolyze the host immune-activating cyclic dinucleotides 3'3' and 2'3' cyclic G-A monophosphate (cGAMP). Experimental analysis showed that ligT homologs encoded by avian poxviruses likewise hydrolyze 2'3' cGAMP, showing that ligT-mediated targeting of cGAMP is an evolutionarily conserved mechanism of immune evasion present in both bacteriophage and eukaryotic viruses. Together, the viral protein structural database and analytics presented here afford new opportunities to identify mechanisms of virus-host interactions that are common across the virome.

Sequence, Structure and Functional space ofDrosophila de novoproteins

“…This observation might suggest that the emergence of function from novel proteins, even through random sequences, could be a more probable phenomena than previously thought. Finally we emphasize that, while efforts to deduce protein function based on structural similarity is ongoing [Nomburg et al, 2024, Gligorijević et al, 2021], numerous instances exist where proteins with similar structures perform different functions, and vice versa [Finkelstein et al, 1993, Govindarajan and Goldstein, 1996, Galperin et al, 1998, Martin et al, 1998].…”

“…A protein’s structure can provide some clues about its function [Orengo et al, 1999]. For example, one can reasonably guess the function of an uncharacterized protein by comparing its structure to that of a known functional protein [Nomburg et al, 2024]. Although, protein function is often attributed to its structure, and unfolded proteins were assumed to be toxic, many studies show that disordered proteins can be functional [Deiana et al, 2019, Jemth et al, 2018, Ali and Ivarsson, 2018].…”

Structure-guided discovery of ancestral CRISPR-Cas13 ribonucleases

“…We further leveraged structural and sequence comparisons to resolve complex phylogenetic relationships, enabling the discovery of recurrent themes underlying CRISPR-Cas enzyme evolution. Although this study focused on the AlphaFold database, our strategy generalizes to other structure prediction databases, including those of metagenomic or viral origin ( 15 , 41 ). As structure prediction methods and associated databases continue to advance, structure-guided protein mining will become increasingly powerful, enabling greater access to biological insights that have long evaded detection.…”

Hidden evolutionary constraints dictate the retention of coronavirus accessory genes

“…We used recombinant MHV encoding a cellular PDE, the AKAP7 central domain (AKAP7), as a functional replacement for an inactive NS2 (ref 8) with a substitution at position 126 (His126Arg), that results in restricted viral replication 18 . Viral PDEs were derived via capture of cellular AKAP7-like genes 19,20 so using this virus, MHV AKAP7 , simulates a horizontal transfer event giving rise to a novel viral gene. Critically, the AKAP7 PDE is inserted in place of ORF4 (ref 8), allowing us to probe the relationship between selection and gene retention without substantial expansion of the genome.…”