Effect of Monovalent Cations on the Structure and Dynamics of Multimodal Chromatographic Surfaces.

While multimodal (MM) chromatography is a promising approach for purifying proteins, the lack of a fundamental understanding of how ion-ligand interactions govern selectivity limits its use in the biopharmaceutical industry. This study uses molecular dynamics simulations to study the interactions between simple monovalent cations and two commonly used structurally similar multimodal chromatography ligands, the Capto ligand and Nuvia cPrime, immobilized on the surface. On the Capto ligand surface, ion presence and type play a key role in modulating the formation of phenyl rings and carboxylate clusters. The flexible linkage attaching the Capto ligand to the self-assembled monolayer (SAM) surface allowed multiple ligands to form interactions with the small cations, while large cations interacted less strongly, following the order Li+ > Na+ > K+ > Cs+. Thus, smaller cations resulted in greater ordering on the surface and lower ion diffusivities, while larger cations resulted in less ordering and higher ion diffusivities, following the order Li+ < Na+ < K+ < Cs+. In contrast, due to the rigid attachment of Nuvia cPrime to the SAM surfaces, the cations bound less strongly and had a much smaller effect on ligand clustering or ordering. Additionally, ions in the presence of the Nuvia cPrime surface had generally greater diffusivities than those in the presence of the Capto ligand. Overall, the interaction of cations with the multimodal ligands can lead to unique configurations on the SAM that likely contribute to differential behavior in biological separations.

Autonomous, multiproperty-driven molecular discovery: From predictions to measurements and back.

A closed-loop, autonomous molecular discovery platform driven by integrated machine learning tools was developed to accelerate the design of molecules with desired properties. We demonstrated two case studies on dye-like molecules, targeting absorption wavelength, lipophilicity, and photooxidative stability. In the first study, the platform experimentally realized 294 unreported molecules across three automatic iterations of molecular design-make-test-analyze cycles while exploring the structure-function space of four rarely reported scaffolds. In each iteration, the property prediction models that guided exploration learned the structure-property space of diverse scaffold derivatives, which were realized with multistep syntheses and a variety of reactions. The second study exploited property models trained on the explored chemical space and previously reported molecules to discover nine top-performing molecules within a lightly explored structure-property space.

Probing IgG1 FC-Multimodal Nanoparticle Interactions: A Combined Nuclear Magnetic Resonance and Molecular Dynamics Simulations Approach.

In this study, NMR and molecular dynamics simulations were employed to study IgG1 FC binding to multimodal surfaces. Gold nanoparticles functionalized with two multimodal cation-exchange ligands (Capto and Nuvia) were synthesized and employed to carry out solution-phase NMR experiments with the FC. Experiments with perdeuterated 15N-labeled FC and the multimodal surfaces revealed micromolar residue-level binding affinities as compared to millimolar binding affinities with these ligands in free solution, likely due to cooperativity and avidity effects. The binding of FC with the Capto ligand nanoparticles was concentrated near an aliphatic cluster in the CH2/CH3 interface, which corresponded to a focused hydrophobic region. In contrast, binding with the Nuvia ligand nanoparticles was more diffuse and corresponded to a large contiguous positive electrostatic potential region on the side face of the FC. Results with lower-ligand-density nanoparticles indicated a decrease in binding affinity for both systems. For the Capto ligand system, several aliphatic residues on the FC that were important for binding to the higher-density surface did not interact with the lower-density nanoparticles. In contrast, no significant difference was observed in the interacting residues on the FC to the high- and low-ligand density Nuvia surfaces. The binding affinities of FC to both multimodal-functionalized nanoparticles decreased in the presence of salt due to the screening of multiple weak interactions of polar and positively charged residues. For the Capto ligand nanoparticle system, this resulted in an even more focused hydrophobic binding region in the interface of the CH2 and CH3 domains. Interestingly, for the Nuvia ligand nanoparticles, the presence of salt resulted in a large transition from a diffuse binding region to the same focused binding region determined for Capto nanoparticles at 150 mM salt. Molecular dynamics simulations corroborated the NMR results and provided important insights into the molecular basis of FC binding to these different multimodal systems containing clustered (observed at high-ligand densities) and nonclustered ligand surfaces. This combined biophysical and simulation approach provided significant insights into the interactions of FC with multimodal surfaces and sets the stage for future analyses with even more complex biotherapeutics.

Behavior of Water Near Multimodal Chromatography Ligands and Its Consequences for Modulating Protein-Ligand Interactions.

Multimodal chromatography is a powerful approach for purifying proteins that uses ligands containing multiple modes of interaction. Recent studies have shown that selectivity in multimodal chromatographic separations is a function of the ligand structure and geometry. Here, we performed molecular dynamics simulations to explore how the ligand structure and geometry affect ligand-water interactions and how these differences in solution affect the nature of protein-ligand interactions. Our investigation focused on three chromatography ligands: Capto MMC, Nuvia cPrime, and Prototype 4, a structural variant of Nuvia cPrime. First, the solvation characteristics of each ligand were quantified via three metrics: average water density, fluctuations, and residence time. We then explored how solvation was perturbed when the ligand was bound to the protein surface and found that the probability of the phenyl ring dewetting followed the order: Capto MMC > Prototype 4 > Nuvia cPrime. To explore how these differences in dewetting affect protein-ligand interactions, we calculated the probability of each ligand binding to different types of residues on the protein surface and found that the probability of binding to a hydrophobic residue followed the same order as the dewetting behavior. This study illustrates the role that wetting and dewetting play in modulating protein-ligand interactions.

Ion Solvation and Transport in Narrow Carbon Nanotubes: Effects of Polarizability, Cation-π Interaction, and Confinement.

Understanding ion solvation and transport under confinement is critical for a wide range of emerging technologies, including water desalination and energy storage. While molecular dynamics (MD) simulations have been widely used to study the behavior of confined ions, considerable deviations between simulation results depending on the specific treatment of intermolecular interactions remain. In the following, we present a systematic investigation of the structure and dynamics of two representative solutions, that is, KCl and LiCl, confined in narrow carbon nanotubes (CNTs) with a diameter of 1.1 and 1.5 nm, using a combination of first-principles and classical MD simulations. Our simulations show that the inclusion of both polarization and cation-π interactions is essential for the description of ion solvation under confinement, particularly for large ions with weak hydration energies. Beyond the variation in ion solvation, we find that cation-π interactions can significantly influence the transport properties of ions in CNTs, particularly for KCl, where our simulations point to a strong correlation between ion dehydration and diffusion. Our study highlights the complex interplay between nanoconfinement and specific intermolecular interactions that strongly control the solvation and transport properties of ions.

Fast Permeation of Small Ions in Carbon Nanotubes.

Simulations and experiments have revealed enormous transport rates through carbon nanotube (CNT) channels when a pressure gradient drives fluid flow, but comparatively little attention has been given to concentration-driven transport despite its importance in many fields. Here, membranes are fabricated with a known number of single-walled CNTs as fluid transport pathways to precisely quantify the diffusive flow through CNTs. Contrary to early experimental studies that assumed bulk or hindered diffusion, measurements in this work indicate that the permeability of small ions through single-walled CNT channels is more than an order of magnitude higher than through the bulk. This flow enhancement scales with the ion free energy of transfer from bulk solutions to a nanoconfined, lower-dielectric environment. Reported results suggest that CNT membranes can unlock dialysis processes with unprecedented efficiency.

Atomistic Characterization of Gramicidin Channel Formation.

We investigated gramicidin A (gA) subunit dimerization in lipid bilayers using microsecond-long replica-exchange umbrella sampling simulations, millisecond-long unbiased molecular dynamics simulations, and machine learning. Our simulations led to a dimer structure that is indistinguishable from the experimentally determined gA channel structures, with the two gA subunits joined by six hydrogen bonds (6HB). The simulations also uncovered two additional dimer structures, with different gA-gA stacking orientations that were stabilized by four or two hydrogen bonds (4HB or 2HB). When examining the temporal evolution of the dimerization, we found that two bilayer-inserted gA subunits can form the 6HB dimer directly, with no discernible intermediate states, as well as through paths that involve the 2HB and 4HB dimers.

Identification of preferred multimodal ligand-binding regions on IgG1 FC using nuclear magnetic resonance and molecular dynamics simulations.

In this study, the binding of multimodal chromatographic ligands to the IgG1 FC domain were studied using nuclear magnetic resonance and molecular dynamics simulations. Nuclear magnetic resonance experiments carried out with chromatographic ligands and a perdeuterated 15 N-labeled FC domain indicated that while single-mode ion exchange ligands interacted very weakly throughout the FC surface, multimodal ligands containing negatively charged and aromatic moieties interacted with specific clusters of residues with relatively high affinity, forming distinct binding regions on the FC . The multimodal ligand-binding sites on the FC were concentrated in the hinge region and near the interface of the CH 2 and CH 3 domains. Furthermore, the multimodal binding sites were primarily composed of positively charged, polar, and aliphatic residues in these regions, with histidine residues exhibiting some of the strongest binding affinities with the multimodal ligand. Interestingly, comparison of protein surface property data with ligand interaction sites indicated that the patch analysis on FC corroborated molecular-level binding information obtained from the nuclear magnetic resonance experiments. Finally, molecular dynamics simulation results were shown to be qualitatively consistent with the nuclear magnetic resonance results and to provide further insights into the binding mechanisms. An important contribution to multimodal ligand-FC binding in these preferred regions was shown to be electrostatic interactions and π-π stacking of surface-exposed histidines with the ligands. This combined biophysical and simulation approach has provided a deeper molecular-level understanding of multimodal ligand-FC interactions and sets the stage for future analyses of even more complex biotherapeutics.

Quantifying orthogonality and separability: A method for optimizing resin selection and design.

Recent studies have shown that by combining orthogonal, non-affinity chromatography steps, it is possible to rapidly develop efficient purification processes for molecules of interest. Here, we build upon previous work to develop a flexible framework for identifying resins that remove optimally orthogonal sets of impurities for a wide variety of products. Our approach involves screening a library of proteins with diverse properties (pI ranging from 5.0-11.4 and varying hydrophobicity measured by retention in a HIC gradient) on a library of resins and quantifying each resin's ability to separate every protein pair in the library. Orthogonality is then defined as the degree to which two resins separate mutually exclusive sets of protein pairs. We applied this approach to a library of model proteins and a series of strong, salt tolerant, and multimodal ion exchangers and evaluated which resin combinations performed well and which performed poorly. In particular, we found that strong cation and strong anion exchangers were orthogonal, while strong and salt tolerant anion exchangers were not orthogonal. Interestingly, salt tolerant and multimodal cation exchangers were found to be orthogonal and the best resin combination included a multimodal cation exchange resin and a tentacular anion exchange resin. This approach for quantifying orthogonality is valuable in that it can be used both as a criteria for resin design as well as process design. We envision that, using this framework, it will be possible to design a set of next generation chromatography ligands that are explicitly engineered to optimize separability and orthogonality.

Predicting Small Molecule Transfer Free Energies by Combining Molecular Dynamics Simulations and Deep Learning.

Accurately predicting small molecule partitioning and hydrophobicity is critical in the drug discovery process. There are many heterogeneous chemical environments within a cell and entire human body. For example, drugs must be able to cross the hydrophobic cellular membrane to reach their intracellular targets, and hydrophobicity is an important driving force for drug-protein binding. Atomistic molecular dynamics (MD) simulations are routinely used to calculate free energies of small molecules binding to proteins, crossing lipid membranes, and solvation but are computationally expensive. Machine learning (ML) and empirical methods are also used throughout drug discovery but rely on experimental data, limiting the domain of applicability. We present atomistic MD simulations calculating 15,000 small molecule free energies of transfer from water to cyclohexane. This large data set is used to train ML models that predict the free energies of transfer. We show that a spatial graph neural network model achieves the highest accuracy, followed closely by a 3D-convolutional neural network, and shallow learning based on the chemical fingerprint is significantly less accurate. A mean absolute error of ∼4 kJ/mol compared to the MD calculations was achieved for our best ML model. We also show that including data from the MD simulation improves the predictions, tests the transferability of each model to a diverse set of molecules, and show multitask learning improves the predictions. This work provides insight into the hydrophobicity of small molecules and ML cheminformatics modeling, and our data set will be useful for designing and testing future ML cheminformatics methods.

The Role of Ligand-Ligand Interactions in Multimodal Ligand Conformational Equilibria and Surface Pattern Formation.

Multimodal chromatography uses multiple modes of interaction such as charge, hydrophobic, or hydrogen bonding to separate proteins. Recently, we used molecular dynamics (MD) simulations to show that ligands immobilized on surfaces can interact and associate with neighboring ligands to form hydrophobic and charge patches, which may have important implications for the nature of protein-surface interactions. Here, we study interfacial systems of increasing complexity-from a single immobilized multimodal ligand to high density surfaces-to better understand how ligand behavior is affected by the presence of a surface and the presence of other ligands in the vicinity, and how this behavior scales to larger systems. We find that tethering a ligand to a surface restricts its conformations to a subset of those observed in free solution, yet the ligand maintains flexibility in the plane of the surface and can form contacts with neighboring ligands. We find that although the formation of a contact between two neighboring ligands is slightly unfavorable, three neighboring ligands exhibit a preference for the formation of a fully connected cluster. To explore how these trends in ligand association extend to a larger surface with high density of ligands, we performed coarse-grained Monte Carlo (MC) simulations of a 132-ligand surface using ligand interactions parametrized based on free energies obtained from the three-ligand MD simulations. Despite their simplicity, the coarse-grained simulations qualitatively capture the cluster size distribution of ligands observed in detailed MD simulations. Quantitative differences between the two suggest opportunities for improvements in the coarse-grained energy function for efficient predictions of cluster and pattern formations. Our approach presents a promising route to the engineering of multimodal patterns for future chromatographic resin design.

Formation of Ligand Clusters on Multimodal Chromatographic Surfaces.

Multimodal chromatography is a powerful tool which uses multiple modes of interaction, such as charge and hydrophobicity, to purify protein-based therapeutics. In this work, we performed molecular dynamics simulations of a series of multimodal cation-exchange ligands immobilized on a hydrophilic self-assembled monolayer surface at the commercially relevant surface density (1 ligand/nm2). We found that ligands that were flexible and terminated in a hydrophobic group had a propensity to aggregate on the surface, while less flexible ligands containing a hydrophobic group closer to the surface did not aggregate. For aggregating ligands, this resulted in the formation of a surface pattern that contained relatively large patches of hydrophobicity and charge whose sizes exceeded the length scale of the individual ligands. On the other hand, lowering the surface density to 1 ligand/3 nm2 reduced or eliminated this aggregation behavior. In addition, the introduction of a flexible linker (corresponding to the commercially available ligand) enhanced cluster formation and allowed aggregation to occur at lower surface densities. Further, the use of flexible linkers enabled hydrophobic groups to collapse to the surface, reducing their accessibility. Finally, we developed an approach for quantifying differences in the observed surface patterns by calculating distributions of the patch size and patch length. This clustering phenomenon is likely to play a key role in governing protein-surface interactions in multimodal chromatography. This new understanding of multimodal surfaces has important implications for developing improved predictive models and designing new classes of multimodal separation materials.

Conformational Equilibria of Multimodal Chromatography Ligands in Water and Bound to Protein Surfaces.

Multimodal chromatography uses small ligands with multiple modes of interaction, e.g., charged, hydrophobic or hydrogen bonding, to separate proteins from complex mixtures. The mechanism by which multimodal ligands interact with proteins is expected to be affected by ligand conformations, among other factors. Here, we study conformational equilibria of two commercially used multimodal cation exchange ligands, Capto MMC and Nuvia cPrime, in a range of solvents, a Lennard-Jones (LJ) liquid, ethanol, and water, using molecular dynamics (MD) simulations. By mapping ligand conformations onto two key torsion angles, ω and φ, in these solvents and in low and high dielectric media, we quantify the relative importance of intramolecular and solvent-mediated interactions. In a high dielectric medium, Capto MMC preferentially samples three conformations, which are stabilized by a combination of an intramolecular torsion potential (on ω) and LJ interactions. In an LJ liquid, solvent molecules compete with intramolecular interactions while simultaneously providing an osmotic force, stabilizing both closer and farther distances between ligand sites. This has the overall effect of "flattening out" the conformational landscape. Interestingly, in ethanol and water, hydrogen bonding between the amide hydrogen and solvent molecules stabilizes two additional conformations of Capto MMC in which ω takes on less favorable cis-like configurations. MD simulations of ligands in free solution with three therapeutic antibody fragments show that ligand conformational equilibria remain effectively unchanged upon binding to proteins. Although, there is 20-30% dehydration of the overall ligand upon binding, the hydrogen-bonding sites are dehydrated to a much smaller extent, particularly in cis-like configurations. Conformational preferences of Nuvia cPrime are similar to that of Capto MMC, except for the effect of symmetry arising from the absence of an alkyl thiol tail. Characterizing the conformational equilibria of these two ligands in free solution and bound to a protein provides a foundation for developing a mechanistic understanding of protein-multimodal ligand interactions.