Reinforcement learning-guided control strategies for CAR T-cell activation and expansion.

Reinforcement learning (RL), a subset of machine learning (ML), could optimize and control biomanufacturing processes, such as improved production of therapeutic cells. Here, the process of CAR T-cell activation by antigen-presenting beads and their subsequent expansion is formulated in silico. The simulation is used as an environment to train RL-agents to dynamically control the number of beads in culture to maximize the population of robust effector cells at the end of the culture. We make periodic decisions of incremental bead addition or complete removal. The simulation is designed to operate in OpenAI Gym, enabling testing of different environments, cell types, RL-agent algorithms, and state inputs to the RL-agent. RL-agent training is demonstrated with three different algorithms (PPO, A2C, and DQN), each sampling three different state input types (tabular, image, mixed); PPO-tabular performs best for this simulation environment. Using this approach, training of the RL-agent on different cell types is demonstrated, resulting in unique control strategies for each type. Sensitivity to input-noise (sensor performance), number of control step interventions, and advantages of pre-trained RL-agents are also evaluated. Therefore, we present an RL framework to maximize the population of robust effector cells in CAR T-cell therapy production.

Computational prediction of the effect of amino acid changes on the binding affinity between SARS-CoV-2 spike RBD and human ACE2.

The association of the receptor binding domain (RBD) of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein with human angiotensin-converting enzyme 2 (hACE2) represents the first required step for cellular entry. SARS-CoV-2 has continued to evolve with the emergence of several novel variants, and amino acid changes in the RBD have been implicated with increased fitness and potential for immune evasion. Reliably predicting the effect of amino acid changes on the ability of the RBD to interact more strongly with the hACE2 can help assess the implications for public health and the potential for spillover and adaptation into other animals. Here, we introduce a two-step framework that first relies on 48 independent 4-ns molecular dynamics (MD) trajectories of RBD-hACE2 variants to collect binding energy terms decomposed into Coulombic, covalent, van der Waals, lipophilic, generalized Born solvation, hydrogen bonding, π-π packing, and self-contact correction terms. The second step implements a neural network to classify and quantitatively predict binding affinity changes using the decomposed energy terms as descriptors. The computational base achieves a validation accuracy of 82.8% for classifying single-amino acid substitution variants of the RBD as worsening or improving binding affinity for hACE2 and a correlation coefficient of 0.73 between predicted and experimentally calculated changes in binding affinities. Both metrics are calculated using a fivefold cross-validation test. Our method thus sets up a framework for screening binding affinity changes caused by unknown single- and multiple-amino acid changes offering a valuable tool to predict host adaptation of SARS-CoV-2 variants toward tighter hACE2 binding.

De novo design and Rosetta-based assessment of high-affinity antibody variable regions (Fv) against the SARS-CoV-2 spike receptor binding domain (RBD).

The continued emergence of new SARS-CoV-2 variants has accentuated the growing need for fast and reliable methods for the design of potentially neutralizing antibodies (Abs) to counter immune evasion by the virus. Here, we report on the de novo computational design of high-affinity Ab variable regions (Fv) through the recombination of VDJ genes targeting the most solvent-exposed hACE2-binding residues of the SARS-CoV-2 spike receptor binding domain (RBD) protein using the software tool OptMAVEn-2.0. Subsequently, we carried out computational affinity maturation of the designed variable regions through amino acid substitutions for improved binding with the target epitope. Immunogenicity of designs was restricted by preferring designs that match sequences from a 9-mer library of "human Abs" based on a human string content score. We generated 106 different antibody designs and reported in detail on the top five that trade-off the greatest computational binding affinity for the RBD with human string content scores. We further describe computational evaluation of the top five designs produced by OptMAVEn-2.0 using a Rosetta-based approach. We used Rosetta SnugDock for local docking of the designs to evaluate their potential to bind the spike RBD and performed "forward folding" with DeepAb to assess their potential to fold into the designed structures. Ultimately, our results identified one designed Ab variable region, P1.D1, as a particularly promising candidate for experimental testing. This effort puts forth a computational workflow for the de novo design and evaluation of Abs that can quickly be adapted to target spike epitopes of emerging SARS-CoV-2 variants or other antigenic targets.

Evaluation of strategies to narrow the product chain-length distribution of microbially synthesized free fatty acids.

The dominant strategy for tailoring the chain-length distribution of free fatty acids (FFA) synthesized by heterologous hosts is expression of a selective acyl-acyl carrier protein (ACP) thioesterase. However, few of these enzymes can generate a precise (greater than 90% of a desired chain-length) product distribution when expressed in a microbial or plant host. The presence of alternative chain-lengths can complicate purification in situations where blends of fatty acids are not desired. We report the assessment of several strategies for improving the dodecanoyl-ACP thioesterase from the California bay laurel to exhibit more selective production of medium-chain free fatty acids to near exclusivity. We demonstrated that matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-ToF MS) was an effective library screening technique for identification of thioesterase variants with favorable shifts in chain-length specificity. This strategy proved to be a more effective screening technique than several rational approaches discussed herein. With this data, we isolated four thioesterase variants which exhibited a more selective FFA distribution over wildtype when expressed in the fatty acid accumulating E. coli strain, RL08. We then combined mutations from the MALDI isolates to generate BTE-MMD19, a thioesterase variant capable of producing free fatty acids consisting of 90% of C12 products. Of the four mutations which conferred a specificity shift, we noted that three affected the shape of the binding pocket, while one occurred on the positively charged acyl carrier protein landing pad. Finally, we fused the maltose binding protein (MBP) from E. coli to the N - terminus of BTE-MMD19 to improve enzyme solubility and achieve a titer of 1.9 g per L of twelve-carbon fatty acids in a shake flask.

A single amino acid change led to structural and functional differentiation of PvHd1 to control flowering in switchgrass.

Switchgrass, a forage and bioenergy crop, occurs as two main ecotypes with different but overlapping ranges of adaptation. The two ecotypes differ in a range of characteristics, including flowering time. Flowering time determines the duration of vegetative development and therefore biomass accumulation, a key trait in bioenergy crops. No causal variants for flowering time differences between switchgrass ecotypes have, as yet, been identified. In this study, we mapped a robust flowering time quantitative trait locus (QTL) on chromosome 4K in a biparental F2 population and characterized the flowering-associated transcription factor gene PvHd1, an ortholog of CONSTANS in Arabidopsis and Heading date 1 in rice, as the underlying causal gene. Protein modeling predicted that a serine to glycine substitution at position 35 (p.S35G) in B-Box domain 1 greatly altered the global structure of the PvHd1 protein. The predicted variation in protein compactness was supported in vitro by a 4 °C shift in denaturation temperature. Overexpressing the PvHd1-p.35S allele in a late-flowering CONSTANS-null Arabidopsis mutant rescued earlier flowering, whereas PvHd1-p.35G had a reduced ability to promote flowering, demonstrating that the structural variation led to functional divergence. Our findings provide us with a tool to manipulate the timing of floral transition in switchgrass cultivars and, potentially, expand their cultivation range.

Inactivation Mechanism and Efficacy of Grape Seed Extract for Human Norovirus Surrogate.

Proper disinfection of harvested food and water is critical to minimize infectious disease. Grape seed extract (GSE), a commonly used health supplement, is a mixture of plant-derived polyphenols. Polyphenols possess antimicrobial and antifungal properties, but antiviral effects are not well-known. Here we show that GSE outperformed chemical disinfectants (e.g., free chlorine and peracetic acids) in inactivating Tulane virus, a human norovirus surrogate. GSE induced virus aggregation, a process that correlated with a decrease in virus titers. This aggregation and disinfection were not reversible. Molecular docking simulations indicate that polyphenols potentially formed hydrogen bonds and strong hydrophobic interactions with specific residues in viral capsid proteins. Together, these data suggest that polyphenols physically associate with viral capsid proteins to aggregate viruses as a means to inhibit virus entry into the host cell. Plant-based polyphenols like GSE are an attractive alternative to chemical disinfectants to remove infectious viruses from water or food. IMPORTANCE Human noroviruses are major food- and waterborne pathogens, causing approximately 20% of all cases of acute gastroenteritis cases in developing and developed countries. Proper sanitation or disinfection are critical strategies to minimize human norovirus-caused disease until a reliable vaccine is created. Grape seed extract (GSE) is a mixture of plant-derived polyphenols used as a health supplement. Polyphenols are known for antimicrobial, antifungal, and antibiofilm activities, but antiviral effects are not well-known. In studies presented here, plant-derived polyphenols outperformed chemical disinfectants (i.e., free chlorine and peracetic acids) in inactivating Tulane virus, a human norovirus surrogate. Based on data from molecular assays and molecular docking simulations, the current model is that the polyphenols in GSE bind to the Tulane virus capsid, an event that triggers virion aggregation. It is thought that this aggregation prevents Tulane virus from entering host cells.

Rapid fabrication of precise high-throughput filters from membrane protein nanosheets.

Biological membranes are ideal for separations as they provide high permeability while maintaining high solute selectivity due to the presence of specialized membrane protein (MP) channels. However, successful integration of MPs into manufactured membranes has remained a significant challenge. Here, we demonstrate a two-hour organic solvent method to develop 2D crystals and nanosheets of highly packed pore-forming MPs in block copolymers (BCPs). We then integrate these hybrid materials into scalable MP-BCP biomimetic membranes. These MP-BCP nanosheet membranes maintain the molecular selectivity of the three types of ß-barrel MP channels used, with pore sizes of 0.8 nm, 1.3 nm, and 1.5 nm. These biomimetic membranes demonstrate water permeability that is 20-1,000 times greater than that of commercial membranes and 1.5-45 times greater than that of the latest research membranes with comparable molecular exclusion ratings. This approach could provide high performance alternatives in the challenging sub-nanometre to few-nanometre size range.

7 Log Virus Removal in a Simple Functionalized Sand Filter.

Viral contamination of drinking water due to fecal contamination is difficult to detect and treat effectively, leading to frequent outbreaks worldwide. The purpose of this paper is to report on the molecular mechanism for unprecedented high virus removal from a practical sand filter. Sand filters functionalized using a water extract of Moringa oleifera (MO) seeds, functionalized sand (f-sand) filters, achieved a ∼7 log10 virus removal. These tests were conducted with MS2 bacteriophage, a recognized surrogate for pathogenic norovirus and rotavirus. We studied the molecular mechanism of this high removal since it can have important implications for sand filtration, the most common water treatment technology worldwide. Our data reveal that the virus removal activity of f-sand is due to the presence of a chitin-binding protein, M. oleifera chitin-binding protein (MoCBP) on f-sand. Standard column experiments were supported by proteomic analysis and molecular docking simulations. Our simulations show that MoCBP binds preferentially to MS2 capsid proteins demonstrating that specific molecular interactions are responsible for enhanced virus removal. In addition, we simplified the process of making f-sand and evinced how it could be regenerated using saline water. At present, no definitive solution exists for the challenge of treating fecally contaminated drinking and irrigation water for viruses without using technologies that demand high energy or chemical consumption. We propose functionalized sand (f-sand) filters as a highly effective, energy-efficient, and practical technology for virus removal applicable to both developing and developed countries.

Identification and biophysical characterization of epitope atlas of Porcine Reproductive and Respiratory Syndrome Virus.

Porcine Reproductive and Respiratory Syndrome Virus (PRRSV) have been a critical threat to swine health since 1987 due to its high mutation rate and substantial economic loss over half a billion dollar in USA. The rapid mutation rate of PRRSV presents a significant challenge in developing an effective vaccine. Even though surveillance and intervention studies have recently (2019) unveiled utilization of PRRSV glycoprotein 5 (GP5; encoded by ORF5 gene) to induce immunogenic reaction and production of neutralizing antibodies in porcine populations, the future viral generations can accrue escape mutations. In this study we identify 63 porcine-PRRSV protein-protein interactions which play primary or ancillary roles in viral entry and infection. Using genome-proteome annotation, protein structure prediction, multiple docking experiments, and binding energy calculations, we identified a list of 75 epitope locations on PRRSV proteins crucial for infection. Additionally, using machine learning-based diffusion model, we designed 56 stable immunogen peptides that contain one or more of these epitopes with their native tertiary structures stabilized through optimized N- and C-terminus flank sequences and interspersed with appropriate linker regions. Our workflow successfully identified numerous known interactions and predicted several novel PRRSV-porcine interactions. By leveraging the structural and sequence insights, this study paves the way for more effective, high-avidity, multi-valent PRRSV vaccines, and leveraging neural networks for immunogen design.

Inhibiting the cGAS-STING Pathway in Ulcerative Colitis with Programmable Micelles.

Ulcerative colitis is a chronic condition in which a dysregulated immune response contributes to the acute intestinal inflammation of the colon. Current clinical therapies often exhibit limited efficacy and undesirable side effects. Here, programmable nanomicelles were designed for colitis treatment and loaded with RU.521, an inhibitor of the cyclic GMP-AMP synthase-stimulator of interferon genes (cGAS-STING) pathway. STING-inhibiting micelles (SIMs) comprise hyaluronic acid-stearic acid conjugates and include a reactive oxygen species (ROS)-responsive thioketal linker. SIMs were designed to selectively accumulate at the site of inflammation and trigger drug release in the presence of ROS. Our in vitro studies in macrophages and in vivo studies in a murine model of colitis demonstrated that SIMs leverage HA-CD44 binding to target sites of inflammation. Oral delivery of SIMs to mice in both preventive and delayed therapeutic models ameliorated colitis's severity by reducing STING expression, suppressing the secretion of proinflammatory cytokines, enabling bodyweight recovery, protecting mice from colon shortening, and restoring colonic epithelium. In vivo end points combined with metabolomics identified key metabolites with a therapeutic role in reducing intestinal and mucosal inflammation. Our findings highlight the significance of programmable delivery platforms that downregulate inflammatory pathways at the intestinal mucosa for managing inflammatory bowel diseases.