Justin A. Lemkul

Carbohydrate-digesting enzymes, including pancreatic α-amylase and maltase-glucoamylase (MGAM), are important therapeutic targets for type 2 diabetes due to their role in regulating glucose production. In this study, molecular dynamics (MD) simulations were performed to investigate the stability and interaction patterns of inhibitors identified through SILCS-based virtual screening. Representative ligands for the N-terminal and C-terminal domains of MGAM (NtMGAM, CtMGAM), and α-amylase were parametrized using both the nonpolarizable CHARMM and polarizable Drude force fields based on quantum mechanical (QM) reference data. MD simulations of the apo enzymes show that their overall structures remain stable, with flexibility primarily localized to regions surrounding the catalytic pockets. Simulations of the enzyme–ligand complexes demonstrate that the ligands remain associated with the binding regions throughout the trajectories. Comparison between the two force fields reveals distinct interaction patterns: CHARMM interactions are largely driven by a small number of dominant contacts, whereas the Drude model captures a more extensive and cooperative interaction network. Notably, the Drude model enables enhanced engagement with catalytic residues in NtMGAM, whereas ligand binding in CtMGAM and α-amylase remains more flexible and less directly coupled to the catalytic machinery. These results highlight the importance of polarization in describing enzyme–ligand interactions and suggest that polarization can influence the balance and persistence of interactions in a system-dependent manner. These findings provide further computational support for the candidate inhibitors identified through computational screening.

Highly charged enzyme active sites present a challenge for conventional Site Identification by Ligand Competitive Saturation (SILCS) because strongly interacting charged probes can suppress sampling of neutral fragments, while additive force fields do not explicitly account for electronic polarization. To address these limitations, we developed a comprehensive SILCS-based computational framework and applied it to the N-terminal and C-terminal domains of maltase–glucoamylase and to α-amylase. Both the conventional mixed-probe SILCS workflow and a modified neutral–charged protocol were evaluated using the CHARMM additive and Drude polarizable force fields. FragMap convergence analysis demonstrated reliable sampling across all systems, with the neutral–charged protocol alleviating competitive exclusion effects observed in conventional simulations. In highly charged catalytic pockets, the conventional workflow was dominated by cationic probe sampling, suppressing neutral and hydrogen-bonding features. Separation of charged and neutral probes restored chemically diverse interaction patterns, improving representation of hydrogen-bond donor, acceptor, and hydrophobic hotspots. Incorporation of the Drude polarizable force field further enhanced resolution of electrostatic and hydrogen-bonding interactions. Energy-based SILCS-Pharmacophore models were generated from the neutral–charged FragMaps and used for virtual screening of the Maybridge and ChemBridge libraries. Hits were refined using SILCS-MC, ranked by ligand grid free energy (LGFE) and ligand efficiency (LE), and analyzed through scaffold- and fingerprint-based clustering. Integration of energetic and structural analyses enabled identification of favorable scaffolds and functional group patterns associated with strong binding efficiency. Together, this study develops, evaluates, and applies an improved SILCS workflow to three enzymes associated with type 2 diabetes (T2D), demonstrating its utility for characterizing highly charged active sites and prioritizing candidate inhibitors.

The growing interest in nucleic acids as therapeutic targets has prompted the development of novel computational methods to facilitate drug discovery. In this study, we extend the Site Identification by Ligand Competitive Saturation (SILCS) methodology to characterize ligand–nucleic acid interactions using the Drude polarizable force field. We demonstrate the ability of the Drude force field to better model solute–nucleic acid interactions, resulting in improved identification of known binding sites and ligand binding favorability predictions across a diverse set of nucleic acid structures. This new workflow addresses limitations in previous SILCS studies by exploiting the enhanced sampling of solutes in the original SILCS-RNA workflow, accurately modeling the interactions of charged species, and improving solute sampling in minor groove binding sites. These results establish this Drude-based SILCS workflow as a valuable tool for structure-based drug design targeting nucleic acids and offer insights into solute preferences that can guide rational ligand design.

Intrinsically disordered proteins (IDPs) play critical roles in cellular signaling and regulation, yet their dynamic conformational landscapes make them difficult to characterize experimentally and computationally. Phosphorylation, one of the most common post-translational modifications, frequently occurs within intrinsically disordered regions and can modulate protein structure and function. Enhanced sampling molecular dynamics methods offer a potential route to more efficiently explore the diverse conformations accessible to IDPs, but systematic comparisons of their performance sampling the conformational landscape of such systems remain limited. Here, we evaluate the conformational sampling of the intrinsically disordered β-catenin_17-48 peptide in both its nonphosphorylated and phosphorylated states using three enhanced sampling approaches: Gaussian-accelerated molecular dynamics (GaMD), metadynamics (METAD), and weighted ensemble simulations (WESTPA), in comparison with conventional molecular dynamics simulations. Two collective variables (CVs) were explored to guide sampling: the ϕ dihedral angles of the phosphorylation sites Ser33 and Ser37 and the end-to-end distance of the peptide. We found that different enhanced sampling methods explored distinct regions of conformational space rather than converging to a single ensemble, with GaMD largely overlapping with unbiased simulations. Notably, METAD and WESTPA more readily accessed conformational regions not observed in unbiased simulations. Analysis of the combined conformational ensembles identified intermediate conformations connecting the nonphosphorylated and phosphorylated states, which are preferentially sampled in simulations employing adaptive strategies. Additionally, the Ser33/Ser37 ϕ angle CV more effectively captures phosphorylation-dependent conformational shifts than the end-to-end distance metric. Together, these results highlight how both the choice of enhanced sampling strategy and the selection of collective variables influence the exploration of IDP conformational landscapes.

Coevolution proceeds through the evolution of traits that mediate ecological interactions and evolutionary outcomes. In the arms race between toxic Pacific newts ( Taricha ) and their garter snake predators ( Thamnophis ), this interface involves tetrodotoxin (TTX), an antipredator defense that inhibits nerve and muscle function by blocking voltage-gated sodium channels. In response, snakes have evolved TTX-resistant channels, in some cases leading to snake populations that are nearly invulnerable to TTX. For decades, newt TTX has been treated as a single defensive trait, yet TTX occurs as a family of structurally related analogs that may represent alternative defenses against snakes. Here, we characterize TTX analog diversity in all four species of Taricha and evaluate how these compounds interact with sodium channels in coevolved garter snakes. Using LC–MS analysis of newt skin secretions, we detected a diverse suite of TTX analogs previously unrecognized in Pacific newts. We then used molecular docking models to evaluate interactions between various TTX analogs and variants of the skeletal muscle channel (Na _V 1.4) that span the range of TTX resistance in garter snakes. We found that some TTX analogs docked better than canonical TTX in resistant snake channels. Notably, we show that 11-deoxy-4-epi-TTX and 11-deoxy-TTX have favorable interactions with hydrophobic amino-acid substitutions in extremely resistant garter snake sodium channels, potentially circumventing predator resistance to canonical TTX. Our results suggest a complex arms race involving multiple newt TTX analogs and multiple snake sodium channel variants. As such, newts may keep pace with snakes by diversifying their arsenal of chemical weapons. Significance: In western North America, some newts have enough toxin in their skin to kill most predators, but coevolution has led to toxin resistance in their garter snake predators that in some cases is so extreme that snakes should be able to withstand more toxin than newts could possibly possess. Here, we explore the possibility that newts may be able to fight back against these resistant predators through the by evolving a diverse repertoire of toxins. We use chemical analyses and computational modeling to show that newts have a wide variety of toxins, some of which are likely to be more effective against resistant snakes, suggesting that coevolution may continue to occur even when snakes become very resistant.

Amylose and cellulose are important biopolymers with diverse applications in biotechnology and materials science. Understanding their structural, dynamic, and solvation properties at the molecular level is critical for harnessing their potential. This study investigates the electronic and structural properties of single-chain cellulose and single- and double-chain amylose in aqueous solution using molecular dynamics simulations with both nonpolarizable (CHARMM) and polarizable (Drude) force fields. CHARMM simulations show stable hydrogen bonding between amylose and water, higher glucose ring dipole moments, increased rigidity, adoption of chair conformations, and less variation in dihedral angles. In contrast, Drude simulations captured dynamic electronic polarization, enhanced conformational flexibility, and resulted in heterogeneous inter- and intramolecular hydrogen bonds. For cellulose, structural and solvation behaviors were largely similar between CHARMM and Drude. These findings highlight molecular interactions and solvation dynamics of amylose and cellulose, with potential relevance in materials science and biotechnology.

The growing interest in nucleic acids as therapeutic targets has prompted the devel-opment of novel computational methods to facilitate drug discovery. In this study, we extend the Site Identification by Ligand Competitive Saturation (SILCS) methodology to characterize ligand–nucleic acid interactions using the Drude polarizable force field. We demonstrate the ability of the Drude force field to better model solute-nucleic acid interactions, resulting in improved identification of known binding sites and ligand binding favorability predictions across a diverse set of nucleic acid structures. This new workflow addresses limitations in previous SILCS studies by exploiting the en-hanced sampling of solutes in the original SILCS-RNA workflow, accurately modeling the interactions of charged species, and improving solute sampling in minor groove binding sites. These results establish this Drude-based SILCS workflow as a valuable tool for structure-based drug design targeting nucleic acids and offer insights into solute preferences that can guide rational ligand design.

Enhanced sampling methods have become powerful techniques to investigate complex systems and rare molecular events by facilitating greater access to configurational space. In parallel, the continuous development of polarizable force fields augments such techniques, allowing the investigation of the role of electronic polarization in transitional barriers that might be overlooked in nonpolarizable simulations. Together, improved sampling methods and polarizable force fields should accelerate computational discoveries. Here, we present an implementation of a weighted ensemble strategy for the Drude polarizable force field using the WESTPA software package in conjunction with the OpenMM simulation engine. We demonstrate its use for backbone conformational sampling in a model system (alanine dipeptide) and in studying functional sidechain rotations in an enzyme (Abl1 kinase). We further discuss possible consequences related to the inclusion of explicit electronic polarization in the model. This software implementation and validation should facilitate the use of WESTPA‐Drude strategies throughout the simulation community.