Structural and computational examination of the Arabidopsis profilin-Poly-P complex reveals mechanistic details in profilin-regulated actin assembly.

Profilins are abundant cytosolic proteins that are universally expressed in eukaryotes and that regulate actin filament elongation by binding to both monomeric actin (G-actin) and formin proteins. The atypical profilin Arabidopsis AtPRF3 has been reported to cooperate with canonical profilin isoforms in suppressing formin-mediated actin polymerization during plant innate immunity responses. AtPRF3 has a 37-amino acid-long N-terminal extension (NTE), and its suppressive effect on actin assembly is derived from enhanced interaction with the polyproline (Poly-P) of the formin AtFH1. However, the molecular mechanism remains unclear. Here, we solved the crystal structures of AtPRF3Δ22 and AtPRF3Δ37, as well as AtPRF2 apo form and in complex with AtFH1 Poly-P at 1.5-3.6 Å resolutions. By combining these structures with molecular modeling, we found that AtPRF3Δ22 NTE has high plasticity, with a primary "closed" conformation that can adopt an open conformation that enables Poly-P binding. Furthermore, using molecular dynamics simulation and free-energy calculations of protein-protein binding, along with experimental validation, we show that the AtPRF3Δ22 binds to Poly-P in an adaptive manner, thereby enabling different binding modes that maintain the interaction through disordered sequences. Together, our structural and simulation results suggest that the dynamic conformational changes of the AtPRF3 NTE upon Poly-P binding modulate their interactions to fine-tune formin-mediated actin assembly.

Plasticity of the 340-Loop in Influenza Neuraminidase Offers New Insight for Antiviral Drug Development.

The recently discovered 340-cavity in influenza neuraminidase (NA) N6 and N7 subtypes has introduced new possibilities for rational structure-based drug design. However, the plasticity of the 340-loop (residues 342-347) and the role of the 340-loop in NA activity and substrate binding have not been deeply exploited. Here, we investigate the mechanism of 340-cavity formation and demonstrate for the first time that seven of nine NA subtypes are able to adopt an open 340-cavity over 1.8 µs total molecular dynamics simulation time. The finding that the 340-loop plays a role in the sialic acid binding pathway suggests that the 340-cavity can function as a druggable pocket. Comparing the open and closed conformations of the 340-loop, the side chain orientation of residue 344 was found to govern the formation of the 340-cavity. Additionally, the conserved calcium ion was found to substantially influence the stability of the 340-loop. Our study provides dynamical evidence supporting the 340-cavity as a druggable hotspot at the atomic level and offers new structural insight in designing antiviral drugs.

DeepBindPoc: a deep learning method to rank ligand binding pockets using molecular vector representation.

Accurate identification of ligand-binding pockets in a protein is important for structure-based drug design. In recent years, several deep learning models were developed to learn important physical-chemical and spatial information to predict ligand-binding pockets in a protein. However, ranking the native ligand binding pockets from a pool of predicted pockets is still a hard task for computational molecular biologists using a single web-based tool. Hence, we believe, by using closer to real application data set as training and by providing ligand information, an enhanced model to identify accurate pockets can be obtained. In this article, we propose a new deep learning method called DeepBindPoc for identifying and ranking ligand-binding pockets in proteins. The model is built by using information about the binding pocket and associated ligand. We take advantage of the mol2vec tool to represent both the given ligand and pocket as vectors to construct a densely fully connected layer model. During the training, important features for pocket-ligand binding are automatically extracted and high-level information is preserved appropriately. DeepBindPoc demonstrated a strong complementary advantage for the detection of native-like pockets when combined with traditional popular methods, such as fpocket and P2Rank. The proposed method is extensively tested and validated with standard procedures on multiple datasets, including a dataset with G-protein Coupled receptors. The systematic testing and validation of our method suggest that DeepBindPoc is a valuable tool to rank near-native pockets for theoretically modeled protein with unknown experimental active site but have known ligand. The DeepBindPoc model described in this article is available at GitHub (https://github.com/haiping1010/DeepBindPoc) and the webserver is available at (http://cbblab.siat.ac.cn/DeepBindPoc/index.php).