A perfusion host-microbe bioreactor (HMB) system that captures dynamic interactions of secreted metabolites between epithelial cells cocultured with a human gut anaerobe.

The human microbiota impacts a variety of diseases and responses to therapeutics. Due to a lack of robust in vitro models, detailed mechanistic explanations of host-microbiota interactions cannot often be recapitulated. We describe the design and development of a novel, versatile and modular in vitro system that enables indirect coculture of human epithelial cells with anaerobic bacteria for the characterization of host-microbe secreted metabolite interactions. This system was designed to compartmentalize anaerobes and human cells in separate chambers conducive to each organism's requisite cell growth conditions. Using perfusion, fluidic mixing, and automated sample collection, the cells continuously received fresh media, while in contact with their corresponding compartments conditioned supernatant. Supernatants from each chamber were collected in a cell-free time-resolved fashion. The system sustained low oxygen conditions in the anaerobic chamber, while also supporting the growth of a representative anaerobe (Bacteroides thetaiotaomicron) and a human colonic epithelial cell line (Caco-2) in the aerobic chamber. Caco-2 global gene expression changes in response to coculture with B. thetaiotaomicron was characterized using RNA sequencing. Extensive, targeted metabolomics analysis of over 150 central carbon metabolites was performed on the serially collected supernatants. We observed broad metabolite changes in host-microbe coculture, compared to respective mono-culture controls. These effects were dependent both on sampling time and the compartment probed (apical vs. basolateral). Coculturing resulted in the depletion of several important metabolites, including guanine, uridine 5'-monophosphate, asparagine, and thiamine. Additionally, while Caco-2 cells cultured alone predominantly affected the basolateral metabolite milieu, increased abundance of 2,3-dihydroxyisovalerate and thymine on the basolateral side, occurred when the cells were cocultured with B. thetaiotaomicron. Thus, our system can capture the dynamic, competitive and cooperative processes between host cells and gut microbes.

A high-throughput microfluidic mechanoporation platform to enable intracellular delivery of cyclic peptides in cell-based assays.

Cyclic peptides are poised to target historically difficult to drug intracellular protein-protein interactions, however, their general cell impermeability poses a challenge for characterizing function. Recent advances in microfluidics have enabled permeabilization of the cytoplasmic membrane by physical cell deformation (i.e., mechanoporation), resulting in intracellular delivery of impermeable macromolecules in vector- and electrophoretic-free approaches. However, the number of payloads (e.g., peptides) and/or concentrations delivered via microfluidic mechanoporation is limited by having to pre-mix cells and payloads, a manually intensive process. In this work, we show that cells are momentarily permeable (t 1/2 = 1.1-2.8 min) after microfluidic vortex shedding (µVS) and that lower molecular weight macromolecules can be cytosolically delivered upon immediate exposure after cells are processed/permeabilized. To increase the ability to screen peptides, we built a system, dispensing-microfluidic vortex shedding (DµVS), that integrates a µVS chip with inline microplate-based dispensing. To do so, we synced an electronic pressure regulator, flow sensor, on/off dispense valve, and an x-y motion platform in a software-driven feedback loop. Using this system, we were able to deliver low microliter-scale volumes of transiently mechanoporated cells to hundreds of wells on microtiter plates in just several minutes (e.g., 96-well plate filled in <2.5 min). We validated the delivery of an impermeable peptide directed at MDM2, a negative regulator of the tumor suppressor p53, using a click chemistry- and NanoBRET-based cell permeability assay in 96-well format, with robust delivery across the full plate. Furthermore, we demonstrated that DµVS could be used to identify functional, low micromolar, cellular activity of otherwise cell-inactive MDM2-binding peptides using a p53 reporter cell assay in 96- and 384-well format. Overall, DµVS can be combined with downstream cell assays to investigate intracellular target engagement in a high-throughput manner, both for improving structure-activity relationship efforts and for early proof-of-biology of non-optimized peptide (or potentially other macromolecular) tools.

Functional Diversity of Gram-Negative Permeability Barriers Reflected in Antibacterial Activities and Intracellular Accumulation of Antibiotics.

Gram-negative bacteria are notoriously more resistant to antibiotics than Gram-positive bacteria, primarily due to the presence of the outer membrane and a plethora of active efflux pumps. However, the potency of antibiotics also varies dramatically between different Gram-negative pathogens, suggesting major mechanistic differences in how antibiotics penetrate permeability barriers. Two approaches are used broadly to analyze how permeability barriers affect intracellular accumulation of antibiotics. One compares the antibacterial activities of compounds, while the other measures the total intracellular concentrations of compounds in nongrowing cells, with both approaches using strains harboring wild-type or genetically modified efflux systems and permeability barriers. Whether the two assays provide similar mechanistic insights remains unclear. In this study, we analyzed the intracellular accumulation and antibacterial activities of antibiotics representative of major clinical classes in three Gram-negative pathogens of high clinical importance, Pseudomonas aeruginosa, Escherichia coli, and Acinetobacter baumannii. We found that both assays are informative about properties of permeability barriers, but there is no quantitative agreement between the assays. Our results show that the three pathogens differ dramatically in their permeability barriers, with the outer membrane playing the dominant role in E. coli and P. aeruginosa but efflux dominating in A. baumannii. However, even compounds of the same chemotype may use different permeation pathways depending on small chemical modifications. Accordingly, a classification analysis revealed limited conservation of molecular properties that define compound penetration into the three bacteria.

Development of a fully automated platform for agar-based measurement of viable bacterial growth.

Dynamic in vitro antibacterial studies provide valuable insight on effective dosing strategies prior to translating to in vivo models. Frequent sampling is required to monitor the pharmacodynamics (PD) of these studies, leading to significant work when quantifying the bacterial load of the samples. Spreading a bacterial suspension on agar to allow colony counting is a proven process for measuring very low levels of growth, but commercial automation equipment to handle agar plating and colony counting at scale is not readily available. We describe a process to greatly decrease the hands-on time required for PD assays by utilizing general-purpose liquid handling robots to plate bacteria and a custom-made plate imager to automate colony counting. The platform developed handles the biological assay from beginning to end as well as sample tracking at each step of the process. The process relies heavily on custom automation scheduling software to enable dynamic process decisions and coordinate data flow throughout. Using the described platform, we can efficiently quantify >100 PD samples per day while maintaining the necessary dynamic range of the assay. Alleviating the main bottleneck in the dynamic antibacterial studies has allowed us to accelerate the rate of experiments to provide antibacterial dosing data within shorter timelines.

Development of a microneutralization assay for HSV-2.

BACKGROUND: Plaque Reduction Neutralization Test (PRNT) is the standard assay used for measuring neutralizing antibody responses to Herpes simplex virus type-2 (HSV-2). The PRNT is a cumbersome, time-consuming and laborious assay. The development of a faster, high throughput microneutralization assay (MNA) for HSV-2 viruses carried out in a 96-well format will allow for rapid testing of large numbers of samples for drug and vaccine development. METHODS: We describe the generation of a MNA that utilizes a pair of anti-HSV human monoclonal antibodies (mAbs) for virus detection in HSV-2 infected Vero cells. Antibodies were generated by B-cell cloning from PBMC's isolated from HSV-1 negative/HSV-2 positive donors. We describe the selection and characterization of the antibodies used for virus detection by ELISA with purified, recombinant anti-HSV glycoproteins, antibody binding in infected cells, and Western Blot. We determine the anti-HSV-2 neutralizing titers of immune sera from mice by MNA and PRNT and compare these results by linear regression analysis. RESULTS: We show that neutralization titers for HSV-2, determined by the 96-well MNA correlate with titers determined by a PRNT completed in 24-well plates in both the absence (R2 = 0.8250) and presence (R2 = 0.7075) of complement. CONCLUSIONS: We have successfully developed an MNA that can be used in place of the burdensome PRNT to determine anti-HSV-2 neutralizing activity in serum. This MNA has much greater throughput than the PRNT, allowing many more samples to be processed in a shorter time saving ∼90 % of the time required by the laboratory scientist to complete the task as compared to the traditional PRNT.

In Vitro Pharmacokinetic/Pharmacodynamic Modeling of HIV Latency Reversal by Novel HDAC Inhibitors Using an Automated Platform.

Antiretroviral therapy is able to effectively control but not eradicate HIV infection, which can persist, leading to the need for lifelong therapy. The existence of latently HIV-infected cells is a major barrier to the eradication of chronic HIV infection. Histone deacetylase inhibitors (HDACis), small molecules licensed for oncology indications, have shown the ability to produce HIV transcripts in vitro and in vivo. The pharmacologic parameters that drive optimal HIV latency reversal in vivo are unknown and could be influenced by such factors as the HDACi binding kinetics, concentration of compound, and duration of exposure. This study evaluates how these parameters affect HIV latency reversal for a series of novel HDACis that differ in their enzymatic on and off rates. Varying cellular exposure, using automated washout methods of HDACi in a Jurkat cell model of HIV latency, led to the investigation of the relationship between pharmacokinetic (PK) properties, target engagement (TE), and pharmacodynamic (PD) responses. Using an automated robotic platform enabled miniaturization of a suspension cell-based washout assay that required multiple manipulations over the 48 h duration of the assay. Quantification of histone acetylation (TE) revealed that HDACis showed early peaks and differences in the durability of response between different investigated HDACis. By expanding the sample times, the shift between TE and PD, as measured by green fluorescent protein, could be fully characterized. The comprehensive data set generated by automating the assays described here was used to establish a PK/PD model for HDACi-induced HIV latency reversal.

High-Throughput Agonist Shift Assay Development for the Analysis of M1-Positive Allosteric Modulators.

Agonist shift assays feature cross-titrations of allosteric modulators and orthosteric ligands. Information generated in agonist shift assays can include a modulator's effect on the orthosteric agonist's potency (alpha) and efficacy (beta), as well as direct agonist activity of the allosteric ligand (tauB) and the intrinsic binding affinity of the modulator to the unoccupied receptor (KB). Because of the heavy resource demand and complex data handling, these allosteric parameters are determined infrequently during the course of a drug discovery program and on a relatively small subset of compounds. Automation of agonist shift assays enables this data-rich analysis to evaluate a larger number of compounds, offering the potential to differentiate compound classes earlier and prospectively prioritize based on desired molecular pharmacology. A high-throughput calcium-imaging agonist shift assay was pursued to determine the allosteric parameters of over 1000 positive allosteric modulator (PAM) molecules for the human muscarinic acetylcholine receptor 1 (M1). Control compounds were run repeatedly to demonstrate internal consistency. Comparisons between potency measurements and the allosteric parameter results demonstrate that these different types of measurements do not necessarily correlate, highlighting the importance of fully characterizing and understanding the allosteric properties of leads.

Prospective Assessment of Virtual Screening Heuristics Derived Using a Novel Fusion Score.

High-throughput screening (HTS) is a widespread method in early drug discovery for identifying promising chemical matter that modulates a target or phenotype of interest. Because HTS campaigns involve screening millions of compounds, it is often desirable to initiate screening with a subset of the full collection. Subsequently, virtual screening methods prioritize likely active compounds in the remaining collection in an iterative process. With this approach, orthogonal virtual screening methods are often applied, necessitating the prioritization of hits from different approaches. Here, we introduce a novel method of fusing these prioritizations and benchmark it prospectively on 17 screening campaigns using virtual screening methods in three descriptor spaces. We found that the fusion approach retrieves 15% to 65% more active chemical series than any single machine-learning method and that appropriately weighting contributions of similarity and machine-learning scoring techniques can increase enrichment by 1% to 19%. We also use fusion scoring to evaluate the tradeoff between screening more chemical matter initially in lieu of replicate samples to prevent false-positives and find that the former option leads to the retrieval of more active chemical series. These results represent guidelines that can increase the rate of identification of promising active compounds in future iterative screens.

Iterative Focused Screening with Biological Fingerprints Identifies Selective Asc-1 Inhibitors Distinct from Traditional High Throughput Screening.

N-methyl-d-aspartate receptors (NMDARs) mediate glutamatergic signaling that is critical to cognitive processes in the central nervous system, and NMDAR hypofunction is thought to contribute to cognitive impairment observed in both schizophrenia and Alzheimer's disease. One approach to enhance the function of NMDAR is to increase the concentration of an NMDAR coagonist, such as glycine or d-serine, in the synaptic cleft. Inhibition of alanine-serine-cysteine transporter-1 (Asc-1), the primary transporter of d-serine, is attractive because the transporter is localized to neurons in brain regions critical to cognitive function, including the hippocampus and cortical layers III and IV, and is colocalized with d-serine and NMDARs. To identify novel Asc-1 inhibitors, two different screening approaches were performed with whole-cell amino acid uptake in heterologous cells stably expressing human Asc-1: (1) a high-throughput screen (HTS) of 3 M compounds measuring 35S l-cysteine uptake into cells attached to scintillation proximity assay beads in a 1536 well format and (2) an iterative focused screen (IFS) of a 45 000 compound diversity set using a 3H d-serine uptake assay with a liquid scintillation plate reader in a 384 well format. Critically important for both screening approaches was the implementation of counter screens to remove nonspecific inhibitors of radioactive amino acid uptake. Furthermore, a 15 000 compound expansion step incorporating both on- and off-target data into chemical and biological fingerprint-based models for selection of additional hits enabled the identification of novel Asc-1-selective chemical matter from the IFS that was not identified in the full-collection HTS.

Development of a Platform to Enable Fully Automated Cross-Titration Experiments.

In the triage of hits from a high-throughput screening campaign or during the optimization of a lead compound, it is relatively routine to test compounds at multiple concentrations to determine potency and maximal effect. Additional follow-up experiments, such as agonist shift, can be quite valuable in ascertaining compound mechanism of action (MOA). However, these experiments require cross-titration of a test compound with the activating ligand of the receptor requiring 100-200 data points, severely limiting the number tested in MOA assays in a screening triage. We describe a process to enhance the throughput of such cross-titration experiments through the integration of Hewlett Packard's D300 digital dispenser onto one of our robotics platforms to enable on-the-fly cross-titration of compounds in a 1536-well plate format. The process handles all the compound management and data tracking, as well as the biological assay. The process relies heavily on in-house-built software and hardware, and uses our proprietary control software for the platform. Using this system, we were able to automate the cross-titration of compounds for both positive and negative allosteric modulators of two different G protein-coupled receptors (GPCRs) using two distinct assay detection formats, IP1 and Ca2+ detection, on nearly 100 compounds for each target.