Acceptors stability modulates the efficiency of post-translational protein N-glycosylation.

N-glycosylation is the most common protein modification in the eukaryotic secretory pathway. It involves the attachment a high mannose glycan to Asn residues in the context of Asn-X-Ser/Thr/Cys, a motif known as N-glycosylation sequon. This process is mediated by STT3A and STT3B, the catalytic subunits of the oligosaccharyltransferase complexes. STT3A forms part of complexes associated with the SEC61 translocon and functions co-translationally. Vacant sequons have another opportunity for glycosylation by complexes carrying STT3B. Local sequence information plays an important role in determining N-glycosylation efficiency, but non-local factors can also have a significant impact. For instance, certain proteins associated with human genetic diseases exhibit abnormal N-glycosylation levels despite having wild-type acceptor sites. Here, we investigated the effect of protein stability on this process. To this end, we generated a family of 40 N-glycan acceptors based on superfolder GFP, and we measured their efficiency in HEK293 cells and in two derived cell lines lacking STT3B or STT3A. Sequon occupancy was highly dependent on protein stability, improving as the thermodynamic stability of the acceptor proteins decreases. This effect is mainly due to the activity of the STT3B-based OST complex. These findings can be integrated into a simple kinetic model that distinguishes local information within sequons from global information of the acceptor proteins.

N-glycosylation Triggers a Dual Selection Pressure in Eukaryotic Secretory Proteins.

Nearly one third of the eukaryotic proteome traverses the secretory pathway and most of these proteins are N-glycosylated in the lumen of the endoplasmic reticulum. N-glycans fulfill multiple structural and biological functions, and are crucial for productive folding of many glycoproteins. N-glycosylation involves the attachment of an oligosaccharide to selected asparagine residues in the sequence N-X-S/T (X ≠ P), a motif known as an N-glycosylation'sequon'. Mutations that create novel sequons can cause disease due to the destabilizing effect of a bulky N-glycan. Thus, an analogous process must have occurred during evolution, whenever ancestrally cytosolic proteins were recruited to the secretory pathway. Here, we show that during evolution N-glycosylation triggered a dual selection pressure on secretory pathway proteins: while sequons were positively selected in solvent exposed regions, they were almost completely eliminated from buried sites. This process is one of the sharpest evolutionary signatures of secretory pathway proteins, and was therefore critical for the evolution of an efficient secretory pathway.

Protein fibrillation lag times during kinetic inhibition.

Protein aggregation is linked to more than 30 human pathologies, including Alzheimer's and Parkinson's diseases. Since small oligomers that form at the beginning of the fibrillation process probably are the most toxic elements, therapeutic strategies involving fibril fragmentation could be detrimental. An alternative approach, named kinetic inhibition, aims to prevent fibril formation by using small ligands that stabilize the parent protein. The factors that govern fibrillation lag times during kinetic inhibition are largely unknown, notwithstanding their importance for designing effective long-term therapies. Inhibitor-bound species are not likely to be incorporated into the core of mature fibrils, although their presence could alter the kinetics of the fibrillation process. For instance, inhibitor-bound species may act as capping elements that impair the nucleation process and/or fibril growth. Here, we address this issue by studying the effect of two natural inhibitors on the fibrillation behavior of lysozyme at neutral pH. We analyzed a set of 79 fibrillation curves obtained in lysozyme alone and a set of 37 obtained in the presence of inhibitors. We calculated the concentrations of the relevant species at the beginning of the curves using the inhibitor-binding constants measured under the same experimental conditions. We found that inhibitor-bound protein species do not affect fibrillation onset times, which are mainly determined by the concentration of unbound protein species present in equilibrium. In this system, knowledge of the fibrillation kinetics and inhibitor affinities suffices to predict the effect of kinetic inhibitors on fibrillation lag times. In addition, we developed a new methodology to better estimate fibrillation lag times from experimental curves.