Artificial permutations have also been
Artificial permutations have also been exploited by protein engineering to manipulate protein scaffolds, in order to improve catalytic activity, alter substrate or ligand binding affinity, reduce proteolytic susceptibility, increase stability, generate different aggregation states and improve fluorescence properties, as well as in the design of novel biosensors and biocatalysts [, , ,, , , , ]. This experimental approach has also proved valuable to determine the role of topology in folding cooperativity , as well as to explore the protein folding jak inhibitor landscape to gain insights into mechanisms underlying protein stability and folding . However, their impact on the evolution of enzyme function within a protein superfamily has been poorly explored. Therefore, since both ATP and ADP-dependent enzymes from the ribokinase superfamily are almost identical in terms of the three-dimensional structures of their large domain, the question arises as to whether the large rearrangement of the folding topology has been crucial for enabling changes in nucleotide specificity or kinetic features during the evolution of this superfamily. To address this issue, we have determined the effect of the introduction of an ATP-dependent topology in an ADP-dependent homologous enzyme via a non-cyclic permutation. Specifically, we have used the glucokinase from Thermococcus litoralis (TlGK), since this enzyme has been extensively characterized regarding its kinetic mechanism and the ligand-induced conformational changes correlating with enzyme function . Structural evidence demonstrates that rewiring the topology of TlGK leads to an active and soluble enzyme without modifications on its three-dimensional architecture. Although this non-cyclic permutation does not affect nucleotide preference, it leads to a striking change in substrate binding order that mimics the one observed for ATP-dependent homologs. Our results demonstrate that rearrangement of the protein topology of the β-meander region of enzymes from the ribokinase superfamily determines the binding order of their substrates.
Materials and methods
Acknowledgments This research was supported by Fondo Nacional de Desarrollo Científico y Tecnológico from Chile (Grant 1150460, 3160373) and the Spanish Ministry of Economy, Industry and Competitiveness (CTQ2015-66206-C2-2-R), co-funded with European Union ERDF funds (European Regional Development Fund), and CSIC (PIE-201620E064) to M.C.V. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. SAXS measurements were supported by LNLS (proposal SAXS1-16983). We gratefully acknowledge access to the beamlines PROXIMA1 and PROXIMA2A at the SOLEIL synchrotron source, Paris, France, and the staff for excellent support. CD spectroscopy was performed in equipment supported by Fondequip EQM140151.
Introduction Historically, eukaryotic protein glycosylation was thought to occur exclusively in the endoplasmic reticulum and Golgi apparatus as part of the secretory pathway, which produces a vast array of diverse membrane glycoproteins. In the mid-1980s, however, Hart et al. found O-linked β-N-acetylglucosamine (O-GlcNAc) on nuclear and cytoplasmic proteins (Figure 1a) . The O-GlcNAc modification is dynamic, and its addition and removal are governed by a single pair of enzymes, O-GlcNAc transferase (OGT) and O-GlcNAcase (OGA) (Figure 1b) [2, 3, 4]. Thousands of nucleocytoplasmic proteins are substrates of these O-GlcNAc-cycling enzymes. Because O-GlcNAc levels change substantially in response to nutrient availability and multiple forms of environmental stress (e.g. hypoxia, oxidative stress, thermal stress), it is thought that O-GlcNAc cycling serves to maintain cell homeostasis by impacting cell signaling, gene expression, and proteostasis, among other processes [5,6]. Dysregulated O-GlcNAc abundance has been linked to several human diseases, including diabetes, cardiovascular disease, cancer, and neurodegenerative diseases, and it has been speculated that OGT and OGA may be therapeutic targets [7, 8, 9, 10]. In addition, mutations in OGT have been connected to X-linked intellectual disability . While the importance of O-GlcNAc cycling in metazoan physiology is by now indisputable, the functional significance of O-GlcNAc on individual substrates is extraordinarily challenging to decipher because there are so many O-GlcNAc substrates, and the rules governing substrate selection are still unclear. Therefore, methods to selectively manipulate the cellular repertoire of O-GlcNAc are currently limiting. For OGT, the challenge is compounded by the recent discovery that this enzyme uses the same active site to attach O-GlcNAc and to effect another physiologically relevant modification, the cleavage of the essential cell cycle regulator, HCF-1 (Figure 1b) [12,13]. Progress in deconvoluting the functions of the O-GlcNAc cycling enzymes depends on having structural information to guide cellular experiments. A number of major advances have been made on this front in the past five years. This review will summarize key findings of structural studies on human OGT and OGA, with our apologies for the many omissions made due to space limitations. Information about structures mentioned in the text is provided in Figure 1.