Tagged for thiolation: A novel glycoconjugation strategy utilizes a non‐natural olefin‐containing amino acid (homoallylglycine, Hag) as a “tag” for modification and a photoinitiated hydroglycothiolation reaction that is selective only for the Hag olefinic “tag”. Application of this method to a number of model proteins allowed complete and precise site‐selective glycosylation generating glycoconjugates that include, for example, virus‐like particles displaying up to 180 glycans at preselected positions (see scheme).

Ligand polyvalency is a powerful modulator of protein–receptor interactions. Host–pathogen infection interactions are often mediated by glycan ligand–protein interactions, yet its interrogation with very high copy number ligands has been limited to heterogenous systems. Here we report that through the use of nested layers of multivalency we are able to assemble the most highly valent glycodendrimeric constructs yet seen (bearing up to 1,620 glycans). These constructs are pure and well-defined single entities that at diameters of up to 32 nm are capable of mimicking pathogens both in size and in their highly glycosylated surfaces. Through this mimicry these glyco-dendri-protein-nano-particles are capable of blocking (at picomolar concentrations) a model of the infection of T-lymphocytes and human dendritic cells by Ebola virus. The high associated polyvalency effects (β>106, β/N ~102–103) displayed on an unprecedented surface area by precise clusters suggest a general strategy for modulation of such interactions.

One advantage of detecting amide H/2 H exchange by mass spectrometry instead of NMR is that the more rapidly exchanging surface amides are still detectable. In this study, we present quench-flow amide H/ 2 H exchange experiments to probe how rapidly the surfaces of two different proteins exchange. We compared the amide H/ 2 H exchange behavior of thrombin, a globular protein, and IB␣, a nonglobular protein, to explore any differences in the determinants of amide H/ 2 H exchange rates for each class of protein. The rates of exchange of only a few of the surface amides were as rapid as the "intrinsic" exchange rates measured for amides in unstructured peptides. Most of the surface amides exchanged at a slower rate, despite the fact that they were not seen to be hydrogen bonded to another protein group in the crystal structure. To elucidate the influence of the surface environment on amide H/ 2 H exchange, we compared exchange data with the number of amides participating in hydrogen bonds with other protein groups and with the solvent accessible surface area. The best correlation with amide H/ 2 H exchange was found with the total solvent accessible surface area, including side chains. In the case of the globular protein, the correlation was modest, whereas it was well correlated for the nonglobular protein. The nonglobular protein also showed a correlation between amide exchange and hydrogen bonding. These data suggest that other factors, such as complex dynamic behavior and surface burial, may alter the expected exchange rates in globular proteins more than in nonglobular proteins where all of the residues are near the

Thrombomodulin (TM) forms a 1:1 complex with thrombin. Whereas thrombin alone cleaves fibrinogen to make the fibrin clot, the thrombin-TM complex cleaves protein C to initiate the anticoagulant pathway. Crystallographic investigations of the complex between thrombin and TMEGF456 did not show any changes in the thrombin active site. Therefore, research has focused recently on how TM may provide a docking site for the protein C substrate. Previous work, however, showed that when the thrombin active site was occupied with substrate analogues labeled with fluorophores, the fluorophores responded differently to active (TMEGF1-6) versus inactive (TMEGF56) fragments of TM. To investigate this further, we have carried out amide H/(2)H exchange experiments on thrombin in the presence of active (TMEGF45) and inactive (TMEGF56) fragments of TM. Both on-exchange and off-exchange experiments show changes in the thrombin active site loops, some of which are observed only when the active TM fragment is bound. These results are consistent with the previously observed fluorescence changes and point to a mechanism by which TM changes the thrombin substrate specificity in favor of protein C cleavage.