De novo design of protein interactions with learned surface fingerprints.

Physical interactions between proteins are essential for most biological processes governing life1. However, the molecular determinants of such interactions have been challenging to understand, even as genomic, proteomic and structural data increase. This knowledge gap has been a major obstacle for the comprehensive understanding of cellular protein-protein interaction networks and for the de novo design of protein binders that are crucial for synthetic biology and translational applications2-9. Here we use a geometric deep-learning framework operating on protein surfaces that generates fingerprints to describe geometric and chemical features that are critical to drive protein-protein interactions10. We hypothesized that these fingerprints capture the key aspects of molecular recognition that represent a new paradigm in the computational design of novel protein interactions. As a proof of principle, we computationally designed several de novo protein binders to engage four protein targets: SARS-CoV-2 spike, PD-1, PD-L1 and CTLA-4. Several designs were experimentally optimized, whereas others were generated purely in silico, reaching nanomolar affinity with structural and mutational characterization showing highly accurate predictions. Overall, our surface-centric approach captures the physical and chemical determinants of molecular recognition, enabling an approach for the de novo design of protein interactions and, more broadly, of artificial proteins with function.

Cholesterol Localization around the Metabotropic Glutamate Receptor 2.

The metabotropic glutamate receptor (mGluR) 2 plays a key role in the central nervous system. mGluR2 has been shown to be regulated by its surrounding lipid environment, especially by cholesterol, by an unknown mechanism. Here, using a combination of biochemical approaches, photo-cross-linking experiments, and molecular dynamics simulations we show the interaction of cholesterol with at least two, but potentially five more, preferential sites on the mGluR2 transmembrane domain. Our simulations demonstrate that surface matching, rather than electrostatic interactions with specific amino acids, is the main factor defining cholesterol localization. Moreover, the cholesterol localization observed here is similar to the sterol-binding pattern previously described in silico for other members of the mGluR family. Biochemical assays suggest little influence of cholesterol on trafficking or dimerization of mGluR2. Nevertheless, simulations revealed a significant reduction of residue-residue contacts together with an alteration in the internal mechanical stress at the cytoplasmic side of the helical bundle when cholesterol was present in the membrane. These alterations may be related to destabilization of the basal state of mGluR2. Due to the high sequence conservation of the transmembrane domains of mGluRs, the molecular interaction of cholesterol and mGluR2 described here is also likely to be relevant for other members of the mGLuR family.