bioRxiv Science⌕ Search

bioRxiv · 10.1101/157495

Evolution of an enzyme from a solute-binding protein

Abstract

Much of the functional diversity observed in modern enzyme superfamilies originates from molecular tinkering with existing enzymes1. New enzymes frequently evolve from enzymes with latent, promiscuous activities2, and often inherit key features of the ancestral enzyme, retaining conserved catalytic groups and stabilizing analogous intermediates or transition states3. While experimental evolutionary biochemistry has yielded considerable insight into the evolution of new enzymes from existing enzymes4, the emergence of catalytic activity de novo remains poorly understood. Although certain enzymes are thought to have evolved from non-catalytic proteins5-7, the mechanisms underlying these complete evolutionary transitions have not been described. Here we show how the enzyme cyclohexadienyl dehydratase (CDT) evolved from a cationic amino acid-binding protein belonging to the solute-binding protein (SBP) superfamily. Analysis of the evolutionary trajectory between reconstructed ancestors and extant proteins showed that the emergence and optimization of catalytic activity involved several distinct processes. The emergence of CDT activity was potentiated by the incorporation of a desolvated general acid into the ancestral binding site, which provided an intrinsically reactive catalytic motif, and reshaping of the ancestral binding site, which facilitated enzyme-substrate complementarity. Catalytic activity was subsequently gained via the introduction of hydrogen-bonding networks that positioned the catalytic residue precisely and contributed to transition state stabilization. Finally, catalytic activity was enhanced by remote substitutions that refined the active site structure and reduced sampling of non-catalytic states. Our work shows that the evolutionary processes that underlie the emergence of enzymes by natural selection in the wild are mirrored by recent examples of computational design and directed evolution of enzymes in the laboratory.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Clifton, B. E., Kaczmarski, J. A., Carr, P. D., Gerth, M. L., Tokuriki, N., Jackson, C. J.. 2017-06-30. Evolution of an enzyme from a solute-binding protein. https://doi.org/10.1101/157495

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

Structure of influenza D virus NS1 reveals a novel Zn-binding site and ssRNA-binding activity

Influenza D virus (IDV) is an emerging pathogen associated with bovine respiratory disease and has potential for zoonotic transmission. Although nonstructural protein 1 (NS1) is a major virulence factor in influenza A and B viruses, the structure and function of IDV NS1 remain poorly defined. Here, we determined crystal structures of the IDV NS1 N-terminal RNA-binding domain (NTD) and C-terminal effector domain (CTD). The NTD retains the conserved homodimeric six-helix fold, whereas the CTD adopts a distinct architecture containing a novel Zn2+-binding site. IDV NS1 CTD preferentially binds single-stranded RNA, and mutagenesis identified lysine and arginine residues critical for this activity. Single-residue substitutions disrupting either the Zn2+-binding site or RNA-binding interface prevented recovery of infectious recombinant IDV. These findings reveal distinct structural and functional features of IDV NS1 and establish the importance of CTD integrity for viral replication. Conservation of key CTD residues in influenza C virus NS1 suggests shared structural features that distinguish influenza C and D virus NS1 proteins from their influenza A and B counterparts.

biochemistry↗

Recognition of the centromere-binding protein Ndc10 by Scm3HJURP directs Cse4CENP-A deposition at native centromeres

Kinetochores connect centromeres to spindle microtubules, ensuring accurate chromosome segregation. Kinetochore assembly at centromeres depends on deposition of CENP-A, a histone H3 variant. Despite considerable progress, how CENP-A is specifically assembled on native centromeres remains an open question. Centromere-specific assembly of CENP-A occurs spontaneously in a cell-free extract system from the yeast Saccharomyces cerevisiae, but with low efficiency, suggesting that one or more factors are limiting. Here, we show Scm3HJURP, a specific chaperone for the centromeric histone Cse4CENP-A, is one such limiting component. The addition of a complex of Scm3, Cse4 and histone H4 strongly stimulates Cse4 assembly in a centromeric DNA sequence- and dose-dependent manner; in addition, improvements in both Cse4 and kinetochore assembly were more prominent in mitotic versus interphase (G1) extracts. Using the complemented cell-free reconstitution system, we show that distinct Scm3 interfaces with Cse4 and with the Ndc10 subunit of the yeast centromere DNA sequence-recognizing CBF3 complex are essential for centromere-specific Cse4 assembly and cell viability. Thus, recognition of Ndc10 directs Scm3-dependent and locus-specific centromeric histone deposition.

biochemistry↗

Heterogeneous primary nucleation determines Aβ42 fibrilstructure.

The current developments in the field of amyloid fibril structure determination has revealed multiple structures for peptides of identical sequence. This may reflect variations in sample environment or similar nucleation barriers for multiple folds. Here we investigate the influence of aggregating surface and during fibril formation from monomeric amyloid {beta} peptide (A{beta}42). The fibril structures formed in identical solutions in glass vials / bottles and PEGylated polystyrene 96-well plates display very different monomer fold within the fibrils. This is a clear indication of extremely slow homogeneous primary nucleation. Instead, heterogeneous primary nucleation at vessel walls or at the three-line interface between air, water and vessel wall dictate the outcome in each solution, where the initially formed structure is propagated through elongation and secondary nucleation. The results further imply that the conversion towards the thermodynamically stable fibril structure one is slow for A{beta}42.

biochemistry↗