A three-dimensional high throughput assay identifies novel antibacterial molecules with activity against intracellular Shigella

A three-dimensional high throughput assay identifies novel antibacterial molecules with activity against intracellular Shigella | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article A three-dimensional high throughput assay identifies novel antibacterial molecules with activity against intracellular Shigella Voong Vinh Phat, Andrew Lim, Cristina Cozar-Gallardo, Maria Isabel Castellote Alvaro, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5874972/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 13 You are reading this latest preprint version Abstract The Gram-negative bacterial species Shigella is the second leading cause of diarrhea among children in low and middle-income countries (LMICs) and is a World Health Organization (WHO) priority pathogen. Shigella infections are becoming increasing difficult to treat due to antimicrobial resistance (AMR), leading to an urgent for new antimicrobial agents with novel modes of action. Shigella pathogenesis is largely intracellular and antibacterial chemicals that preferentially work inside cells may be desirable to limit collateral AMR and block key components of the Shigella infection cycle. Aiming to facilitate the process of identifying antibacterial chemicals that kill intracellular Shigella , we developed a high-throughput screening (HTS) cell-based chemical screening assay. The three-dimensional (3-D) assay, incorporating Shigella invasion into Caco-2 cells on Cytodex 3 beads, was scaled into a 384 well platform for screening chemical compound libraries. Using this assay, we evaluated > 500,000 compounds, identifying 12 chemical hits that inhibit Shigella replication inside cells. This simple, efficient and HTS-compatible assays circumvents many of the limitations of traditional screening methods with cell monolayers and may be deployed for antibacterial compound screening for other intracellular pathogens. Biological sciences/Drug discovery Health sciences/Diseases/Gastrointestinal diseases Health sciences/Diseases/Infectious diseases Biological sciences/Microbiology Biological sciences/Microbiology/Antimicrobials drug discovery high-throughput screening Shigellosis caco-2 cell Figures Figure 1 Figure 2 Figure 3 Introduction Shigellosis is an acute enteric infection caused by organisms belonging to the genus Shigella . With 165 million cases per year globally and an estimated 1.1 million deaths, Shigella remains a significant public health challenge 1 . Data from the Global Enteric Multi Center Study (GEMS) found that Shigella was a major contributor to the global diarrhea burden and the most common aetiological agent in young children with diarrhea 2 . Currently, there is no licensed vaccine to provide protection against any Shigella species 3,4 . Consequently, antimicrobials are the mainstay for disease control and are commonly used to treat severe Shigella infections 1 . However, Shigella are highly adept at acquiring multi-drug resistance (MDR) plasmids from other Enterobacteriaceae, and antimicrobial resistance (AMR) mutations are commonly associated with successful lineages 5,6 . Treatment options for Shigella infection are rapidly diminishing due to resistance to the majority of commonly used antimicrobials, included those recommended by the WHO as empirical choices for Shigella- related diarrhea 7 . A universal increase in AMR necessitates the need for the discovery of new antibacterials, ideally with novel modes of action. However, antibacterial drug discovery and development is challenging 8 , as exemplified by various notable negative experiences in large pharmaceutical companies 9,10 . Such issues have also been compounded by the “innovation” gap where no new classes of antibacterial agents were discovered 11,12 . One of the key discovery issues is finding small molecules that can overcome the unique architecture of the bacterial cell wall. The bacterial cell wall of Gram-negative bacteria contains outer membrane proteins and porins which create a hydrophilic barrier. The hydrophilic barrier contrasts the lipophilic barrier that characterizes the mammalian cell membrane, which antibacterial compounds must overcome to target intracellular bacteria. Therefore, antibacterial compounds should ideally have physicochemical properties that allow permeation into mammalian and bacterial cells. Additionally, there is discussion regarding the physicochemical space that antibacterial agents occupy, which may differ to those occupied by alternative drug classes 13,14 . Phenotypic cellular approaches for screening new antibacterial agents have the advantage of being able to identify compounds with intracellular activity early in the process, although this approach has the risk of a lower hit rate due to the more restrictive physicochemical property requirements. Caco-2 cells are widely used in intestinal epithelial cell-based models to evaluate and predict absorption of compounds due to their simplicity and reproducibility 15 . However, experiments with Caco-2 cells are normally conducted on permeable filter inserts or collagen-coated surfaces and experimented 16,17 , which is low-throughput and not amenable to screening large numbers of compounds across a broad chemical space. Given the limitations of traditional methods, micro-carrier technologies are being exploited to grow various anchorage-dependent cell types that are not able to grow in a suspension or cells that are adherent but need to be differentiated for functionality 18 . Micro-carrier technology has multiple advantages over more traditional approaches, including enhancing the surface-volume ratio, reducing experimental time, increasing reliability and permitting the use of adherent cells like a suspension system and scale-up can be performed in bioreactors 19–21 . Notably, Caco-2 cells cultured on microcarrier beads have been applied for screening chemicals against a range of pathogens, but have not yet been deployed for Shigella 18,22–24 . Aiming to identify new compounds with intracellular killing activity against Shigella , we developed a three-dimensional Caco-2 cell platform to perform in vitro phenotypic high-throughput screening of > 500,000 compounds against Shigella . Caco-2 cells were cultured in high-yield on Cytodex 3 beads and infected by nanoluciferase-producing Shigella flexneri , the resulting infected cell culture was used for compound screening. We aimed to evaluate the efficiency and apply this new assay to early-stage drug discovery. Our results confirmed that this new assay is a comparatively simple method in which to identify compounds with antibacterial properties against intracellular Shigella without the limitations of existing monolayer models. Results High-throughput screening assay development We aimed to establish a high-throughput screening assay based on the Caco-2 cell model to identify potential drug targets that could inhibit the replication of Shigella flexneri inside cells (Fig. 1 ). Shigella flexneri infectivity was conducted using a 3-dimensional model of large intestine Caco-2 cells, which were grown in a bioreactor at the given cell and microcarrier concentrations in a humid atmosphere of 5% CO 2 at 37°C, 3-D cells were then harvested and transferred to cell-culture plates for experimenting or screening. The assay was optimized using a large-volume spinner flask (Wheaton, Spain), as an alternative to the rotating wall vessel (RWV) bioreactor as has been used in previous studies 19–21 . This vessel was used because it was facilitative to change medium during differentiation process, scale up culture volume to adapt HTS’s requirement 25 , and the integrity of the Caco-2 cells was not affected, as evidenced by the analysis of sucrase, ALP production, and ZO-1 formation, which are markers of Caco-2 functionality. At cell confluence, total activity of sucrase in the monolayer had increased by 2.8-fold from baseline, and the activity relative to the 3-D model increased by 3.5-fold ( Fig S1 . A and B ). ALP activity demonstrated an increasing trend from day 8 to day 21 in both the monolayer and 3-D model (5.5-fold) ( Fig S1 . C and D ). These increases were consistent with the induction of sucrase and ALP that has been reported for Caco-2 cells 26,27 . Additionally, ZO-1 proteins were expressed abundantly in the majority of Caco-2 cells in both the monolayer and 3-D models, forming a clear, narrow, punctate line along the apical surface of cells ( Figure S2 ). The presence of ZO-1 indicated the formation of tight junctions in cell-to-cell interaction 26,28,29 . The optimization of assay parameters, such as MOI, invasion incubation time, the optimal time and multiplicity of infection (MOI) for Shigella invasion were determined using the balance between the Z’ factor and percentage of bacterial coverage ( Table S2 ). A MOI of 150 and six hours of invasion were determined to be the optimal assay conditions, which ensured 100% bacterial growth in wells, while maintaining the robustness of the assay (Z’ > 0.4; and S/B value > 2) 30,31 . Cell detachment resulting in empty beads were observed after four hours of invasion, this phenomenon was observed to increase with time. At 4 and 6 hours of invasion, ~ 80–90% of beads had attached Caco-2 cells in comparison to pre-invaded beads. Induction was not suitable for screening after 6 hours of invasion. Other testing conditions did not meet validator standards. MOIs of 5 and 10 were suboptimal as they required too much time for invasion and the robustness of the assay could not be maintained. An MOI of 100 demonstrated improved performance with perfect coverage after six and eight hours of invasion; however, the Z’ values were not stable and fell out of standard range (Z’ < 0.4). To reduce the number of beads required for primary screening, three different bead concentrations (1,000, 2,000 and 4,000 beads/ml) were tested, but only the 4,000 beads/ml gave a suitable for a high-throughput screening assay (Z-score (mean Z’ = 0.57) and S/B values > 2-fold). We evaluated the invasive efficiency of Shigella flexneri by defining invasive efficiency as the percentage of intracellular bacteria (CFU/ml) to total number of bacteria (CFU/ml) (%) at the same duration of invasion ( Table S3 ). The S. flexneri SF_nanoluc strain, Shigella flexneri serotype 2a 2457T carrying reporter plasmid pMK-RQ_tac + nanoluc (Table S1 ), with an invasion efficiency of 0.083%, exhibited the highest invasion efficiency at six hours of invasion. This score resulted in approximately eight invading bacteria for every 10 4 CFU input. In terms of bacteria per cell or bead, we estimated seven intracellular Shigella flexneri for every 100 cells or ~ 15 Shigella flexneri for each bead. We next assessed dose-response potential of the assay by measuring the intra-assay and inter-assay of 11 different commercially available antimicrobials ( Table S4 ). IC 50 values were used to calculate coefficient of variances in percentage ( Table S7 ), then we determined the repeatability and reproducibility of this Caco-2 cell-based assay in case of maintained robustness of the assay. The intra-assay CV and the inter-assay CV across 11 drugs ranged from 0.16–8.53% and from 1.43–14.82%, respectively. The intra-assay CV (< 10%) demonstrated a high degree of reproducible in each batch, and the inter-assay CV (< 15%) demonstrated acceptable reproducibility between batches of the cellular assay. High-throughput screening process validations A validation set of chemicals, comprised of ~ 10,000 compounds and representing a wide diversity of chemotypes present within the GSK chemical collection, was used to evaluate the HTS assay configuration. The assay displayed an average Z’ factor of 0.48 and the obtained cut-off (average response of the sample distribution + 3SD) was 48.87%. The chemical hit rate was ~ 2.5% (i.e. 247 active compounds), but signal patterns were detected after analysing the dataset by using ActivityBase software, signifying the errors and noise of dispense patterns caused by unevenly dispensed beads. The noisy plates were re-run. Aiming to identify hits with novel modes of action, we performed structural screening, discarding 43 compounds that belonged to known drug lines, including dihydrofolate reductases (DHFR), quinolones, and bacterial topoisomerase (BTIs). We then performed a dose-response assay on the 204 remaining compounds. Finally, four compounds gave dose response to the targeted screening S. flexneri organisms, three of them were active in both replicates and had IC 50 < 10µM and the remaining chemical was active in one of two replicates with an IC 50 = 36.31µM. These compounds were further analysed for cytotoxicity in HepG2 cells and subjected to physical characterization ( Table S5 ). HTS primary screening campaign The primary screening campaign of 518,282 compounds was performed in different 49 runs utilising 1,521 individual plates, the workflow is presented in Fig. 2 . The optimal conditions for HTS assay, which using the 3-D Caco-2 cells along with Shigella flexneri SF_nano strain for infectivity with a MOI of 150 in six hours, following by gentamicin protection and diluted to 4,000beads/ml, were applied before 25µl invasive mixture was dispensed into prepared plates for screening. Overall, 45 plates were re-run because their Z’ values were < 0.4 ( Figure S3 ). The robustness of the assay was maintained with an average Z’ factor of 0.55. Compounds with inhibition rates greater than the robust cutoff value of 41.48% were considered active, with 14,451 compounds selected. Considering false signal patterns, 14,451 primary hits were screened again for confirmation at 10µM in the same assay format, representing a hit rate of 0.44% (2,283 compounds). After structural screening, we discarded 221 non-novel antimicrobial compounds, including BTIs, macrolides, quinolones, and DHFRs. The final list of 2,062 compounds were progressed into a dose-response assay, removing those with IC 50 > 25µM. As a result, 225 compounds exhibited inhibition with IC 50 ≤ 25µM, but 15 compounds were not accessible. Therefore, to factor out compounds that may be false positives due to nano-luciferase inhibition, 210 available preliminary hits advanced from the intracellular survival screen were tested in a nanoluciferase-interference assay in dose-response assays starting at 100µM. Ultimately, we identified15 compounds with a pIC50 1, meaning no extracellular-nanoluciferase activity were identified to have intracellular activity against S. flexneri in this assay format (Fig. 2 ). After chemical review, the 15 hit compounds were reorganized based on structure similarity into 13 clusters. Two chemicals were considered as a series and the remaining 11 clusters were singletons (Fig. 3 ). To ensure that the compounds were suitable starting points for qualify hits and to determine the toxic potential of the new chemical entities, an inspection of the compounds for structural features that may impact toxicity was conducted, followed by in vitro cytotoxicity testing on HepG2 cells. Four of the 15 structures carried an NO 2 moiety, including two hits of Series 1, one hit of Series 2, and Singleton 6. Although the antibacterial activity of nitro-containing molecules is known to be broad, nitro groups have been extensively associated with mutagenicity and genotoxicity. Therefore, the compounds that were representative of the entire clusters were additionally tested on HepG2 cells, four compounds (1, 2, 8, and 11) were found to exhibit HepG2 cytotoxicity. Series 1 (compounds 1 and 2) and singleton 4 (compound 8) demonstrated some concert of progression considering its selectivity (Caco-2 pIC50 vs HepG2) and its structure alert ( Table S6) . The remaining 12 compounds (Table 1 ), represented by Series 2 and Singletons 1–3, 5, 6, and 8–11, were of interest as starting points for further optimization toward the drug discovery for treatment of shigellosis. Discussion Pathogenic bacteria have been engaged in a sustained ‘arms race’ with antimicrobials since their introduction and antimicrobial resistant Shigella has emerged as a significant global threat to public health 7 . Novel treatments and approaches are urgently to combat infections caused by MDR Shigella . To aid in addressing this issue we developed, validated and exploited a HTS Caco-2 cell-based system to screen a large compound library against Shigella flexneri . Compared to traditional cell monolayer screening, this assay system retained the disease-relevant cellular architecture, which improves the success of translating hits from the screen to cell-based or in vivo infection models. Additionally, this approach simultaneously has the common advantages of all HTS systems, including, scalability, amenable to automation, simple and rapid transfer of reagents. Our project was initiated by establishing an HTS assay using a spinner flask, and experiments were performed in a 384-well format. We demonstrated that there was no significant difference in the structural integrity of the Caco-2 cells between microcarrier beads (Cytodex 3 ) and the monolayer format; Shigella flexneri still demonstrated invasion and spread from cell-to-cell in both systems. One of the main advantages of a bead-based system is a large surface area to grow cells to high density. This approach facilitates scale-up and reduces the floor space and incubator volume required for a given-sized manufacturing operation and eases the precise process control in a large-scale bioreactor. Consequently, the number of compounds screened per batch can be easily scaled up. However, we found that even after optimising conditions, bacterial invasion rates were substantially lower in comparison to the monolayer format 32,33 . Notably, not every Caco-2 cell was invaded by a Shigella organism and not every microcarrier bead possessed an invaded cell. Therefore, we compensated for this lower rate of invasion by increasing the number of beads to ensure each well contained intracellular Shigella . In generating this new high throughput approach we observed a high false discovery rate or chemicals that did not elaborate a dose-response. After the primary screen, 2.79% of chemical were retained, with only 0.44% remaining after rescreening. This discrepancy may have been caused by false signal patterns recognized throughout the screen, such as edge effects (caused by evaporation), patterns on a plate in general, and errors in adding assay components (dispenser distribution). In addition, there were compounds which were latterly found to inhibit the luciferase reporting system. Nonetheless, we employed various orthogonal screenings so to filter out false positives. We next analysed what chemical scaffolds had been elucidated by our screening effort. To ensure that the compounds were suitable starting points for qualifying hits, we characterized the potential toxicity of the new chemical entities. As a result, 12 compounds represented by Series 2 and Singletons 1–3, 5, 6, and 8–11 were prioritized. Considering their overall profile in terms of physchem properties (most of them with predicted high permeability and different range of solubility) and initial toxicity profile, although the potency of these compounds was low (pIC50 values between 4.6 and 5.8), they could be considered as novel starting points for drug development against Shigella (Table 1 ). Additional research is currently being performed to assess these hits and convert them into developable hit series, these compounds may be developed into tool compounds for antibacterial agents with preferential activity against intracellular bacteria. The validation of these types of tool compound will aid in the future design of novel antibacterial agents with intracellular activity. Additionally, various studies can be planned in the future using such tool compounds to better understand the mode of action of the new chemicals. In summary, we have shown that it is possible to integrate three-dimensional cell culture into an assay system that is amenable to HTS using microcarriers. We have demonstrated the capability of this system in an HTS of 500,000 compounds and have identified novel hits that have activity against intracellular Shigella flexneri . The experience in developing this assay system has enabled us to gain insight on assay optimization issues of which we have demonstrated steps to mitigate them. We think that the microcarrier-based assay system has utility in HTS to drive the hit discovery program, and we envisage this assay system to be used in other intracellular bacterial infection models to pioneer hit discovery against gastrointestinal pathogens. Materials and methods Bacterial strains, Cell lines, Growth media and Reagents Th bacterial strains, cell lines and plasmids used in this study are listed in Table S1 . Shigella flexneri isolates were routinely cultured on Luria-Bertani (LB) broth or LB agar. Where required, the growth medium was supplemented with antimicrobials at the following concentration: Kanamycin (50ug ml − 1 ) for Shigella flexneri serotype 2a 2457T (ATCC® 700930™) carrying plasmid pMK-RQ_tac + nanoluc (SF_nanoluc), and ampicillin (100ug ml − 1 ) for Shigella flexneri ATCC 12022GFP. Caco-2 cells were grown in 5-layer flasks (Nunc/ThermalFisher) with basic EMEM (Sigma-Aldrich) supplemented with 10% FBS (Gibco), 1% L-glutamine in a humidified atmosphere of 5% CO 2 at 37 o C. The medium was changed every 72 hours until 80–90% confluent layers were generated. Generation of the reporter strain The synthetic gene Tac + Nanoluc (593 bp) was assembled from synthetic oligonucleotides and the fragment was inserted into the pMK-RQ vector, yielding the 18ADIM2C_Tac + Nanoluc_pMK-RQ (KanR) construct (hereafter referred to as pMK-RQ-nl). Escherichia coli K12 DH10B™ T1R was used for the propagation of the plasmid. Plasmid DNA was purified from transformed colonies and sequenced (Dynamimed, Spain) for verification. This process was conducted by contract to Invitrogen (Thermo Fisher Scientific, USA). Electroporation of pMK-RQ-nl into S. flexneri 2457T was performed according to a standard protocol (NEB Electroporation protocol, C2986) yielding the S. flexneri serotype 2a 2457T-nl (or SF_nanoluc) used in this study. High-throughput Screening Assay Shigella flexneri SF_nanoluc was grown overnight on a 50mgL − 1 kanamycin LB plate at 37 o C. Bacteria were scraped from the plate surface and sub-cultured in warm supplemented EMEM medium for 2 hours (at 37 o C and 200rpm) and diluted to get OD around 0.3 at 625nm. Induction was performed in 500ml or 1,000ml siliconized spinner flasks under closed culture conditions. 200mg autoclaved Cytodex 3 beads (GE Healthcare), rehydrated with cation-free PBS (Sigma), were washed with supplemented EMEM medium and inoculated with 1.7 million Caco-2 cells/ml in half of the desired culture volume of the medium. The spinner vessel was kept static for 4 hours in a humidified atmosphere of 5% CO 2 at 37 o C, allowing Caco-2 cells to bind to beads. The remaining medium was added into the vessel and stirred at 32.5rpm. The medium was changed every 72 hours, and the induction was used on day 17 to ensure Caco-2 cells were fully differentiated. Cell concentration induction was measured by NucleoCounter® NC-100™ (ChemoMetec Version 2.4). Cellular invasion was initiated by adding the bacterial stock into cell induction with optimal MOI, and the vessel was stationary for an hour before stirring at 22.5rpm for 5 hours in a humid atmosphere of 5% CO 2 at 37 o C. After incubation, culture media was discarded by centrifugation; and a gentamicin protection assay was performed, in which cell beads were washed twice with a warm medium and EMEM containing gentamicin (100ug ml -1 ) was added to the chamber of the system. The cell culture was incubated for 1 hour with stirring and then washed three times with a warm medium before 25µl of the invasive mixture was dispensed into each well. Plates were incubated for two days in a humid atmosphere of 5% CO 2 at 37 o C before proceeding to develop the luminescent signal by adding 10µl mixture of nanoluciferase substrate (Nano-Glo® Luciferase Assay System, N1130 - Promega, Spain) into each well. After standing at room temperature for 30 minutes, luminescent signals were quantified by the Envision Reader (PerkinElmer) in focus luminescent mode. For the bacterial tracking, we used the green fluorescent protein in S. flexneri ATCC 12022GFP to monitor the invasion. A confocal HCS microscope (Opera by PerkinElmer) was used to confirm the invasion of S. flexneri into Caco-2 beads (Fig. 1 ). Compounds and assay plate preparation The HTS assay was performed in 384-well black clear bottom plates (Greiner). Screening compounds were prepared with an appropriate DMSO concentration (0.5% in cell-based assay). Every assay plate contained 16 wells of DMSO as negative controls and 16 wells of moxifloxacin at a final concentration of 20µM as positive controls. These controls were used to monitor assay quality through determination of Z’ as well as normalizing the data on a per-plate basis. To assess the quality of the dose-response assay, 11 commercially available antimicrobials were used to evaluate the reliability of the HTS assay. The co-efficiency of variance (CV%) was calculated by accessing the IC50 (µM) deviation of selected positive controls. Intra-assay reproducibility was determined by comparing the IC50 that were generated in the same run of the two replicates. Inter-assay reproducibility was assessed by comparing the IC50 generated by two replicates of each day over two days. Intra-assay %CV should be < 10%, while inter-assay %CV should be < 15%. To validate the HTS assay configuration, a set of 10 thousand compounds was assayed in 384-well plates at a final concentration of 10µM. Assay were run in duplicate and confirmed if needed, a validation data was assessed to analyze the pros and cons of the whole procedure. The primary screening campaign of 520,000 compounds was screened at the single shot of 10µM final concentration; active compounds were confirmed at the same concentration (10µM). Following screening, all non-novel structures were removed, and the remaining compounds were subjected to a dose response (DR) assay. Those compounds that had DR to targeted organism of ≤ 25µM were progressed to toxicity and structure analysis. To select the most drug-like compounds, Fasted State Simulated Intestinal Fluid (FaSSIF) is used to test the dissolution and permeability of compounds, as well as their stability in the gastrointestinal tract 34 . Human ether-à-go-go-related gene (hERG) is the gene that encodes a potassium channel protein, which is essential for normal heart rhythm. It is important to screen compounds for hERG blockade 35 . Optimizing invasive conditions To determine the optimal condition (Multiplicity of infection - MOI and time invasion) for the invasive process, the induction products were infected with SF_nanoluc at an MOI ranging from 5–150 (5, 10, 100, 150) for five time points (4 hours, 6 hours, 8 hours, 10 hours and > 16 hours). After gentamicin protection, 25µl of invasion was dispensed into prepared 384-well plates, the number of beads > 5,000 beads/ml. The experiments were repeated three times. Results were assessed by Z’ factors (Z’ ≥ 0.4) 31 , percentage of bacterial growth in wells, and bead coverage observed by microscopy. If the coverage of beads with cells was < 70%, the experiment was cancelled (Data do not show). To increase the number of plates per run, the number of beads per well was optimized with three different bead concentrations (1,000, 2,000, and 4,000 beads/ml). 10µl homogenized invasive mixture was dropped on each counting grid of a Neubauer chamber, and the number of beads (with or without attached cells) were enumerated by microscopy. We repeated three times and calculated the average before dispensing into each well of prepared plates. The effects of bead concentration on Z’ values were considered as a selective indicator. Invasive efficiency evaluation A bacterial quantification assay was adapted and modified from a previous study to calculate the invasive rate of Shigella into Caco-2 cells 3 >. After inoculation, 50ml homogenized invasion was withdrawn at three time points (2 hours, 4 hours, and 6 hours) and equally divided into two tubes (A and B) at each time point. Tube A went directly to cell lysis. After being washed with fresh medium and centrifuged at 8, 000rpm for 10 minutes twice to collect all beads and bacterial pellets, the mixture was resuspended in 5ml of PBS and divided into five different 1.5ml tubes (1ml/tube). One was used for cell counting and another for bead counts. The remaining three tubes were lysed with 0.1% saponin in one hour and the dilutions of lysate were plated on 50mgL − 1 kanamycin LB plates for quantification. Tube B was used to determine bacterial invasion; extracellular bacteria were killed with gentamicin (100mgL − 1 ) for 1 hour. Then cultured beads were washed twice with PBS at 800rpm and re-suspended in 5ml PBS. Then, the same process of tube A was applied. The experiments were repeated four times. Invasive rates were calculated as the percentage of alive Shigella after gentamicin protection compared to the total number of Shigella before being treated with gentamicin. Sucrase assay Sucrase is an important digestive enzyme secreted in the small intestine on the brush border and catalyses the hydrolysis of sucrose to its subunits of fructose and glucose. Sucrase is increased during differentiation of the enterocytes and is considered a reliable indicator of Caco-2 cell differentiation in vitro 37,3 >. We monitored sucrose production at day 8, 12, 15 19 and 21 after seeding, in the cytodex 3 beads system as well as in monolayers. Measurements were conducted by the fluorescent Amplex® Red Glucose/Glucose Oxidase Assay Kit (A22189, Invitrogen, Molecular probes) following the manufacturer’s instructions. Alkaline phosphatase (ALP) activity The intestinal alkaline phosphatase encodes a digestive brush-border enzyme, which is highly upregulated during small intestinal epithelial cell differentiation 3 >. ALP activity was measured in the supernatant of cytodex 3 and monolayer Caco-2 cells during differentiation 8, 12, 15, 19 and 21 days after seeding. Measurements were performed by Alkaline Phosphatase Assay Kit Colorimetric (Abcam, ab83369) according to the manufacturer's instructions. Tight junctions of Caco-2 cells The primary function of tight junctions (TJ) in epithelial cells is to create a regulated barrier in the extracellular space serve to regulate the passage of ions and molecules between cells and mark the division between apical and basolateral surfaces of cells in cellular differentiation 4 >. TJs are essential for the polarization of epithelial cells. Intracellular transduction pathways can regulate the integrity of the tight junctions. We identified and characterized ZO-1 as a peripheral membrane protein specifically associated with the cytoplasmic surface of tight junctions. We used Immunofluorescence of Human Caco-2 cells stained with Mouse anti-ZO-1 Monoclonal Antibody - Alexa Fluor® 488 (Product #339188, Thermo Fisher Scientific). DNA is counter-stained with blue Hoechst 33258 (Product #H3569, Thermo Fisher Scientific). Immunofluorescence analysis of ZO-1 was performed using 90% /100% confluent log phase Caco-2 cells in the beads system. Cells were fixed with 4% paraformaldehyde for 30 minutes, permeabilized with 0.1% Triton™ X-100 for 10 minutes and blocked with 1% BSA for 30 minutes at room temperature. The cells were labelled with ZO-1 Monoclonal Antibody (ZO1-1A12), Alexa Fluor 488 at 5µg/mL in 0.1% BSA and incubated overnight at room temperature. Nuclei were stained with DAPI at 1ug/mL concentration without washing. Interference with nanoluciferase reporter assay. The main disadvantage of luciferase-based assays is interference. Therefore, an interference assay was established to identify compounds that might have interference with the nanoluciferase reporter system or luminescence readout. A 3-fold serial dilution, starting at 100µM, of each compound was dispended into 1,536-well plates. A volume of 5µl of fresh bacterial broth was added into each well to obtain 10 6 CFU/ml, which were incubated at 37 o C overnight. Subsequently, 5µl of nanoluciferase substrate solution (Promega) was added to each well, which were place in the dark for 30 minutes before reading the luminescence by Envision device. The pIC50 ratio between the Caco-2 cell assay and interference assay had to be > 1. Compounds with no extracellular-nanoluciferase activity (pIC50 < 4) were considered non-interference compounds. Results were compared with those obtained from an extracellular assay using the same organism but using resazurin as a readout (surrogate of bacterial viability) to assess if activity was dependant on bacterial extracellular or enzymatic inhibition. Results were compared against data from intracellular-THP1 screening, to calculate shift between both luminescence assays. Computational and statistical analysis The raw data of the Envision device was imported and analysed in ActivityBase and Spotfire version 10.3.1.18 (TIBCO). pIC50 values (pIC50 = -log10(IC50 (M)) were obtained using the Activity Base XE nonlinear regression bundle. The Cut-off values (Mean ± 3SD), Z’ factor and signal-to-base were produced automatically by ActivityBase, or manually calculated with equations in table S7. Other data were analysed in Excel (Microsoft Office) and plots were generated using Prism version 6.07 (Graphpad) or IC 50 calculator tool (ATT Bioquest, https://www.aatbio.com/tools/ic50-calculator ). Abbreviations 3-D three-dimension HTS high-throughput screening GEMS Global Enteric Multicentre Study RWV rotating wall vessel MOI Multiplicity of infection SD standard deviation IFI Inhibition Frequency Index IC inhibition concentration CV co-efficiency of variance S/B Signal-to-base PFI Property Forecast Index FaSSIF Fasted State Simulated Intestinal Fluid hERG Human ether-à-go-go-related gene Fig Figure Declarations Ethics approval and consent to participate Not required Consent for publication Not required Competing interests The authors declare no competing interests. Funding This project TC239 was co-funded by the Tres Cantos Open Lab Foundation. Stephen Baker is supported by a Wellcome senior research fellowship (215515/Z/19/Z). The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Author Contribution Conceptualization: Stephen BakerFormal analysis: Voong Vinh Phat, Andrew Lim, Sonia Lozano-AriasProvided supervision: Stephen BakerMethodology: Voong Vinh Phat, Andrew Lim, Sonia Lozano-AriasWriting original draft: Voong Vinh Phat, Andrew Lim, Sonia Lozano-AriasReview and editing: All authorsRead and approved the final version of the manuscript: All authors Acknowledgements We would like to express our sincere thanks to Dr. Janneth Rodrigues at OpenLab Foundation for her kind help. We also thank Ms. Ana Isabel Bardera for her help with our experiments. Data Availability All data are presented in the manuscript and raw data are freely available upon request from the corresponding author. References Kotloff, K. L., Riddle, M. S., Platts-Mills, J. A., Pavlinac, P. & Zaidi, A. K. M. Shigellosis. Lancet 391 , 801–812 (2018). Kotloff, K. L. et al. The Global Enteric Multicenter Study (GEMS) of Diarrheal Disease in Infants and Young Children in Developing Countries: Epidemiologic and Clinical Methods of the Case/Control Study. Clin. Infect. Dis. 55 , S232–S245 (2012). Raso, M. M., Arato, V., Gasperini, G. & Micoli, F. Toward a Shigella Vaccine: Opportunities and Challenges to Fight an Antimicrobial-Resistant Pathogen. Int. J. Mol. Sci. 24 , (2023). Mani, S., Wierzba, T. & Walker, R. I. Status of vaccine research and development for Shigella. Vaccine 34 , 2887–2894 (2016). Baker, K. S. et al. Travel- and Community-Based Transmission of Multidrug-Resistant Shigella sonnei Lineage among International Orthodox Jewish Communities. Emerg. Infect. Dis. 22 , 1545–1553 (2016). Chung The, H. et al. Dissecting the molecular evolution of fluoroquinolone-resistant Shigella sonnei. Nat. Commun. 10 , (2019). Williams, P. C. M. & Berkley, J. A. Guidelines for the treatment of dysentery (shigellosis): a systematic review of the evidence. Paediatr. Int. Child Health 38 , S50–S65 (2018). Silver, L. L. Challenges of Antibacterial Discovery. Clin. Microbiol. Rev. 24 , 71–109 (2011). Payne, D. J., Gwynn, M. N., Holmes, D. J. & Pompliano, D. L. Drugs for bad bugs: confronting the challenges of antibacterial discovery. Nat. Rev. Drug Discov. 6 , 29–40 (2007). Tommasi, R., Brown, D. G., Walkup, G. K., Manchester, J. I. & Miller, A. A. ESKAPEing the labyrinth of antibacterial discovery. Nat. Rev. Drug Discov. 14 , 529–542 (2015). Walsh, C. Where will new antibiotics come from? Nat. Rev. Microbiol. 1 , 65–70 (2003). Fischbach, M. A. & Walsh, C. T. Antibiotics for Emerging Pathogens. Science (80-. ). 325 , 1089–1093 (2009). O’Shea, R. & Moser, H. E. Physicochemical Properties of Antibacterial Compounds: Implications for Drug Discovery. J. Med. Chem. 51 , 2871–2878 (2008). Mugumbate, G. & Overington, J. P. The relationship between target-class and the physicochemical properties of antibacterial drugs. Bioorg. Med. Chem. 23 , 5218–5224 (2015). Artursson, P., Palm, K. & Luthman, K. Caco-2 monolayers in experimental and theoretical predictions of drug transport. Adv. Drug Deliv. Rev. 64 , 280–289 (2012). Delie, F. & Rubas, W. A Human Colonic Cell Line Sharing Similarities With Enterocytes as a Model to Examine Oral Absorption: Advantages and Limitations of the Caco-2 Model. Crit. Rev. Ther. Drug Carr. Syst. 14 , 66 (1997). Lamson, N. G., Ball, R. L., Fein, K. C. & Whitehead, K. A. Thrifty, Rapid Intestinal Monolayers (TRIM) Using Caco-2 Epithelial Cells for Oral Drug Delivery Experiments. Pharm. Res. 36 , 1–12 (2019). Jakob, P. H. et al. A 3-D cell culture system to study epithelia functions using microcarriers. Cytotechnology 68 , 1813–1825 (2016). Kato, D. The design of polymer microcarrier surfaces for enhanced cell growth. Biomaterials 24 , 4253–4264 (2003). WU, S., LIU, C. & LIAN, W. Optimization of microcarrier cell culture process for the inactivated enterovirus type 71 vaccine development. Vaccine 22 , 3858–3864 (2004). Chen, A. K.-L., Reuveny, S. & Oh, S. K. W. Application of human mesenchymal and pluripotent stem cell microcarrier cultures in cellular therapy: Achievements and future direction. Biotechnol. Adv. 31 , 1032–1046 (2013). Drummond, C. G., Nickerson, C. A. & Coyne, C. B. A Three-Dimensional Cell Culture Model To Study Enterovirus Infection of Polarized Intestinal Epithelial Cells. mSphere 1 , 1–17 (2016). Straub, T. M. et al. In vitro cell culture infectivity assay for human noroviruses. Emerg. Infect. Dis. 13 , 396–403 (2007). Straub, T. M. et al. Human norovirus infection of Caco-2 cells grown as a three-dimensional tissue structure. J. Water Health 9 , 225–240 (2011). Khoshnood, N. & Zamanian, A. A comprehensive review on scaffold-free bioinks for bioprinting. Bioprinting 19 , e00088 (2020). Ferruzza, S., Rossi, C., Scarino, M. L. & Sambuy, Y. A protocol for in situ enzyme assays to assess the differentiation of human intestinal Caco-2 cells. Toxicol. Vitr. 26 , 1247–1251 (2012). Chen, L. et al. Mechanistic studies of the transport of peimine in the Caco-2 cell model. Acta Pharm. Sin. B 6 , 125–131 (2016). Valenzano, M. C. et al. Remodeling of Tight Junctions and Enhancement of Barrier Integrity of the CACO-2 Intestinal Epithelial Cell Layer by Micronutrients. PLoS One 10 , e0133926 (2015). Liu, W. et al. Intestinal Alkaline Phosphatase Regulates Tight Junction Protein Levels. J. Am. Coll. Surg. 222 , 1009–1017 (2016). Inglese, J. et al. High-throughput screening assays for the identification of chemical probes. Nat. Chem. Biol. 3 , 466–479 (2007). Bar, H. & Zweifach, A. Z’ Does Not Need to Be > 0.5. SLAS Discov. Adv. Sci. Drug Discov. 247255522094276 (2020) doi:10.1177/2472555220942764. Honma, Y. Effect of erythromycin on Shigella infection of Caco-2 cells. FEMS Immunol. Med. Microbiol. 27 , 139–145 (2000). Mounier, J., Vasselon, T., Hellio, R., Lesourd, M. & Sansonetti, P. J. Shigella flexneri enters human colonic Caco-2 epithelial cells through the basolateral pole. Infect. Immun. 60 , 237–248 (1992). Amidon, G. L., Lennernäs, H., Shah, V. P. & Crison, J. R. A theoretical basis for a biopharmaceutic drug classification: the correlation of in vitro drug product dissolution and in vivo bioavailability. Pharm. Res. 12 , 413–20 (1995). Vargas, H. M. et al. Time for a Fully Integrated Nonclinical–Clinical Risk Assessment to Streamline QT Prolongation Liability Determinations: A Pharma Industry Perspective. Clin. Pharmacol. Ther. 109 , 310–318 (2021). Luck, S. N., Bennett-Wood, V., Poon, R., Robins-Browne, R. M. & Hartland, E. L. Invasion of Epithelial Cells by Locus of Enterocyte Effacement-Negative Enterohemorrhagic Escherichia coli. Infect. Immun. 73 , 3063–3071 (2005). Galand, G. Brush border membrane sucrase-isomaltase, maltase-glucoamylase and trehalase in mammals. Comparative development, effects of glucocorticoids, molecular mechanisms, and phylogenetic implications. Comp. Biochem. Physiol. Part B Comp. Biochem. 94 , 1–11 (1989). Van Beers, E. H. et al. Lactase and sucrase-isomaltase gene expression during Caco-2 cell differentiation. Biochem. J. 308 , 769–775 (1995). Olsen, L., Bressendorff, S., Troelsen, J. T. & Olsen, J. Differentiation-dependent activation of the human intestinal alkaline phosphatase promoter by HNF-4 in intestinal cells. Am. J. Physiol. Liver Physiol. 289 , G220–G226 (2005). Dunagan, M., Chaudhry, K., Samak, G. & Rao, R. K. Acetaldehyde disrupts tight junctions in Caco-2 cell monolayers by a protein phosphatase 2A-dependent mechanism. Am. J. Physiol. Liver Physiol. 303 , G1356–G1364 (2012). Tables Table 1 is available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files SupplementaryFiguresandTables.pdf floatimage1.png Table 1. Complete profiles of the selected 12 hits. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 25 Feb, 2025 Reviews received at journal 20 Feb, 2025 Reviews received at journal 20 Feb, 2025 Reviews received at journal 10 Feb, 2025 Reviewers agreed at journal 06 Feb, 2025 Reviewers agreed at journal 04 Feb, 2025 Reviewers agreed at journal 28 Jan, 2025 Reviewers agreed at journal 28 Jan, 2025 Reviewers agreed at journal 28 Jan, 2025 Reviewers invited by journal 27 Jan, 2025 Editor assigned by journal 24 Jan, 2025 Submission checks completed at journal 23 Jan, 2025 First submitted to journal 21 Jan, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5874972","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":407914544,"identity":"6fc2bf33-343b-4d8c-86bb-b49c882600a9","order_by":0,"name":"Voong Vinh Phat","email":"","orcid":"","institution":"Oxford University","correspondingAuthor":false,"prefix":"","firstName":"Voong","middleName":"Vinh","lastName":"Phat","suffix":""},{"id":407914545,"identity":"bf2419f3-001c-437f-9873-d7b925ccbd26","order_by":1,"name":"Andrew Lim","email":"","orcid":"","institution":"GSK Global Health","correspondingAuthor":false,"prefix":"","firstName":"Andrew","middleName":"","lastName":"Lim","suffix":""},{"id":407914546,"identity":"57fea4cd-2ff1-46ad-ac89-9db156f14604","order_by":2,"name":"Cristina Cozar-Gallardo","email":"","orcid":"","institution":"GSK Global Health","correspondingAuthor":false,"prefix":"","firstName":"Cristina","middleName":"","lastName":"Cozar-Gallardo","suffix":""},{"id":407914547,"identity":"c75e2e06-6277-48d2-9715-c99b0ffd35bf","order_by":3,"name":"Maria Isabel Castellote Alvaro","email":"","orcid":"","institution":"GSK, Tres Cantos","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"Isabel Castellote","lastName":"Alvaro","suffix":""},{"id":407914548,"identity":"2bc7c104-af81-475c-8603-16d06cbb6f8e","order_by":4,"name":"Demetrio Muñoz Alvarez","email":"","orcid":"","institution":"GSK Global Health","correspondingAuthor":false,"prefix":"","firstName":"Demetrio","middleName":"Muñoz","lastName":"Alvarez","suffix":""},{"id":407914549,"identity":"c6f7db23-de43-4177-b6ea-3fc7d6ecaa99","order_by":5,"name":"Elena Fernandez Alvaro","email":"","orcid":"","institution":"GSK Global Health","correspondingAuthor":false,"prefix":"","firstName":"Elena","middleName":"Fernandez","lastName":"Alvaro","suffix":""},{"id":407914550,"identity":"77a55fb7-29dc-4ae5-b17b-cba7d811252f","order_by":6,"name":"Lluis Ballell-Pages","email":"","orcid":"","institution":"GSK Global Health","correspondingAuthor":false,"prefix":"","firstName":"Lluis","middleName":"","lastName":"Ballell-Pages","suffix":""},{"id":407914551,"identity":"fad95c15-35cf-4ff6-b5ff-dada044bd20c","order_by":7,"name":"Sonia Lozano-Arias","email":"","orcid":"","institution":"GSK Global Health","correspondingAuthor":false,"prefix":"","firstName":"Sonia","middleName":"","lastName":"Lozano-Arias","suffix":""},{"id":407914552,"identity":"dda62b43-4766-4d6f-b395-28eb5f05a6c3","order_by":8,"name":"Stephen Baker","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvUlEQVRIiWNgGAWjYFACHgbjPxVwXgJxWgp4zpCq5QNvGyla5PvXHtwgOc9OXreB+eEHxrY0wloMbrxLNjDclmy47QCbsQRjWw4RWiTOmBkkbjvAuO0AgxkDY1sFYS3yM86Y/zg454D9tgPs34jTwnC+x8CwseEA0CIekC3EOOwGX4Ixw7Hk5G2HeYolEs4R4X35/rMHjBlq7Gy3HW/f+OFDWTIRDpNIgDKYGYiMSAb+A0QpGwWjYBSMgpEMAAMzOh9Saw11AAAAAElFTkSuQmCC","orcid":"","institution":"STAR Infectious Diseases Labs (A*STAR IDL), Agency for Science, Technology and Research (A*STAR)","correspondingAuthor":true,"prefix":"","firstName":"Stephen","middleName":"","lastName":"Baker","suffix":""}],"badges":[],"createdAt":"2025-01-21 17:38:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5874972/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5874972/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":75312123,"identity":"d3b8716b-f4c1-473b-b832-a67e4c07c987","added_by":"auto","created_at":"2025-02-03 09:12:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":827882,"visible":true,"origin":"","legend":"\u003cp\u003eHigh-throughput screening of Caco-2 cells.\u003c/p\u003e\n\u003cp\u003eLight microscope images of Caco-2 cell grown on Cytodex\u003csup\u003e3\u003c/sup\u003e beads (A) Induced cells on beads, (B) Cell damaged and detached off beads. Cell imaging by confocal microscope Opera Phenix (C) Non-invasive in Caco-2 cells, (D) Invasive Shigella inside Caco-2 cells.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-5874972/v1/3e89b79dbb69e25c4aea0c0c.png"},{"id":75310623,"identity":"031a2736-c046-4e42-be83-eed7c8aa2149","added_by":"auto","created_at":"2025-02-03 09:04:22","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":32193,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart outlining the screening process leading to confirm 12 potential hits against \u003cem\u003eShigella\u003c/em\u003einvasion.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-5874972/v1/22e81861a93c526f2a09794f.png"},{"id":75310625,"identity":"fec3bea9-cda9-44b2-bdca-9fab3af27290","added_by":"auto","created_at":"2025-02-03 09:04:22","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":88465,"visible":true,"origin":"","legend":"\u003cp\u003ePromising hits from the HTS campaign:\u003c/p\u003e\n\u003cp\u003eTwo chemical families, and eleven potent singletons were selected for complete profiling.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-5874972/v1/0dcc58e89d79c904dccea406.png"},{"id":75312863,"identity":"5a417420-d0ff-4667-bb4d-c8c6712a923a","added_by":"auto","created_at":"2025-02-03 09:20:23","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1774507,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5874972/v1/931bc5c0-c89e-4100-aea7-9145c45296b3.pdf"},{"id":75310633,"identity":"2255599c-85b3-4294-beda-15bcfb816b24","added_by":"auto","created_at":"2025-02-03 09:04:22","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3268478,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFiguresandTables.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5874972/v1/849c120866e18e1809583fa2.pdf"},{"id":75310626,"identity":"b2804d5a-deac-47b5-95b9-bb4ed4f8b06f","added_by":"auto","created_at":"2025-02-03 09:04:22","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":97785,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTable 1. \u0026nbsp;\u0026nbsp;\u003c/strong\u003eComplete profiles of the selected 12 hits.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-5874972/v1/5c5d7eca71a5cc1ca6b4f93f.png"}],"financialInterests":"No competing interests reported.","formattedTitle":"A three-dimensional high throughput assay identifies novel antibacterial molecules with activity against intracellular Shigella","fulltext":[{"header":"Introduction","content":"\u003cp\u003eShigellosis is an acute enteric infection caused by organisms belonging to the genus \u003cem\u003eShigella\u003c/em\u003e. With 165\u0026nbsp;million cases per year globally and an estimated 1.1\u0026nbsp;million deaths, \u003cem\u003eShigella\u003c/em\u003e remains a significant public health challenge \u003csup\u003e1\u003c/sup\u003e. Data from the Global Enteric Multi Center Study (GEMS) found that \u003cem\u003eShigella\u003c/em\u003e was a major contributor to the global diarrhea burden and the most common aetiological agent in young children with diarrhea \u003csup\u003e2\u003c/sup\u003e. Currently, there is no licensed vaccine to provide protection against any \u003cem\u003eShigella\u003c/em\u003e species \u003csup\u003e3,4\u003c/sup\u003e. Consequently, antimicrobials are the mainstay for disease control and are commonly used to treat severe \u003cem\u003eShigella\u003c/em\u003e infections \u003csup\u003e1\u003c/sup\u003e. However, \u003cem\u003eShigella\u003c/em\u003e are highly adept at acquiring multi-drug resistance (MDR) plasmids from other Enterobacteriaceae, and antimicrobial resistance (AMR) mutations are commonly associated with successful lineages \u003csup\u003e5,6\u003c/sup\u003e. Treatment options for \u003cem\u003eShigella\u003c/em\u003e infection are rapidly diminishing due to resistance to the majority of commonly used antimicrobials, included those recommended by the WHO as empirical choices for \u003cem\u003eShigella-\u003c/em\u003erelated diarrhea \u003csup\u003e7\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eA universal increase in AMR necessitates the need for the discovery of new antibacterials, ideally with novel modes of action. However, antibacterial drug discovery and development is challenging \u003csup\u003e8\u003c/sup\u003e, as exemplified by various notable negative experiences in large pharmaceutical companies \u003csup\u003e9,10\u003c/sup\u003e. Such issues have also been compounded by the \u0026ldquo;innovation\u0026rdquo; gap where no new classes of antibacterial agents were discovered \u003csup\u003e11,12\u003c/sup\u003e. One of the key discovery issues is finding small molecules that can overcome the unique architecture of the bacterial cell wall. The bacterial cell wall of Gram-negative bacteria contains outer membrane proteins and porins which create a hydrophilic barrier. The hydrophilic barrier contrasts the lipophilic barrier that characterizes the mammalian cell membrane, which antibacterial compounds must overcome to target intracellular bacteria. Therefore, antibacterial compounds should ideally have physicochemical properties that allow permeation into mammalian and bacterial cells. Additionally, there is discussion regarding the physicochemical space that antibacterial agents occupy, which may differ to those occupied by alternative drug classes \u003csup\u003e13,14\u003c/sup\u003e. Phenotypic cellular approaches for screening new antibacterial agents have the advantage of being able to identify compounds with intracellular activity early in the process, although this approach has the risk of a lower hit rate due to the more restrictive physicochemical property requirements.\u003c/p\u003e \u003cp\u003eCaco-2 cells are widely used in intestinal epithelial cell-based models to evaluate and predict absorption of compounds due to their simplicity and reproducibility \u003csup\u003e15\u003c/sup\u003e. However, experiments with Caco-2 cells are normally conducted on permeable filter inserts or collagen-coated surfaces and experimented \u003csup\u003e16,17\u003c/sup\u003e, which is low-throughput and not amenable to screening large numbers of compounds across a broad chemical space. Given the limitations of traditional methods, micro-carrier technologies are being exploited to grow various anchorage-dependent cell types that are not able to grow in a suspension or cells that are adherent but need to be differentiated for functionality \u003csup\u003e18\u003c/sup\u003e. Micro-carrier technology has multiple advantages over more traditional approaches, including enhancing the surface-volume ratio, reducing experimental time, increasing reliability and permitting the use of adherent cells like a suspension system and scale-up can be performed in bioreactors \u003csup\u003e19\u0026ndash;21\u003c/sup\u003e. Notably, Caco-2 cells cultured on microcarrier beads have been applied for screening chemicals against a range of pathogens, but have not yet been deployed for \u003cem\u003eShigella\u003c/em\u003e \u003csup\u003e18,22\u0026ndash;24\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAiming to identify new compounds with intracellular killing activity against \u003cem\u003eShigella\u003c/em\u003e, we developed a three-dimensional Caco-2 cell platform to perform \u003cem\u003ein vitro\u003c/em\u003e phenotypic high-throughput screening of \u0026gt;\u0026thinsp;500,000 compounds against \u003cem\u003eShigella\u003c/em\u003e. Caco-2 cells were cultured in high-yield on Cytodex\u003csup\u003e3\u003c/sup\u003e beads and infected by nanoluciferase-producing \u003cem\u003eShigella flexneri\u003c/em\u003e, the resulting infected cell culture was used for compound screening. We aimed to evaluate the efficiency and apply this new assay to early-stage drug discovery. Our results confirmed that this new assay is a comparatively simple method in which to identify compounds with antibacterial properties against intracellular \u003cem\u003eShigella\u003c/em\u003e without the limitations of existing monolayer models.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eHigh-throughput screening assay development\u003c/h2\u003e \u003cp\u003eWe aimed to establish a high-throughput screening assay based on the Caco-2 cell model to identify potential drug targets that could inhibit the replication of \u003cem\u003eShigella flexneri\u003c/em\u003e inside cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). \u003cem\u003eShigella flexneri\u003c/em\u003e infectivity was conducted using a 3-dimensional model of large intestine Caco-2 cells, which were grown in a bioreactor at the given cell and microcarrier concentrations in a humid atmosphere of 5% CO\u003csub\u003e2\u003c/sub\u003e at 37\u0026deg;C, 3-D cells were then harvested and transferred to cell-culture plates for experimenting or screening. The assay was optimized using a large-volume spinner flask (Wheaton, Spain), as an alternative to the rotating wall vessel (RWV) bioreactor as has been used in previous studies \u003csup\u003e19\u0026ndash;21\u003c/sup\u003e. This vessel was used because it was facilitative to change medium during differentiation process, scale up culture volume to adapt HTS\u0026rsquo;s requirement \u003csup\u003e25\u003c/sup\u003e, and the integrity of the Caco-2 cells was not affected, as evidenced by the analysis of sucrase, ALP production, and ZO-1 formation, which are markers of Caco-2 functionality. At cell confluence, total activity of sucrase in the monolayer had increased by 2.8-fold from baseline, and the activity relative to the 3-D model increased by 3.5-fold (\u003cb\u003eFig \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e. A and B\u003c/b\u003e). ALP activity demonstrated an increasing trend from day 8 to day 21 in both the monolayer and 3-D model (5.5-fold) (\u003cb\u003eFig \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e. C and D\u003c/b\u003e). These increases were consistent with the induction of sucrase and ALP that has been reported for Caco-2 cells \u003csup\u003e26,27\u003c/sup\u003e. Additionally, ZO-1 proteins were expressed abundantly in the majority of Caco-2 cells in both the monolayer and 3-D models, forming a clear, narrow, punctate line along the apical surface of cells (\u003cb\u003eFigure S2\u003c/b\u003e). The presence of ZO-1 indicated the formation of tight junctions in cell-to-cell interaction \u003csup\u003e26,28,29\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe optimization of assay parameters, such as MOI, invasion incubation time, the optimal time and multiplicity of infection (MOI) for \u003cem\u003eShigella\u003c/em\u003e invasion were determined using the balance between the Z\u0026rsquo; factor and percentage of bacterial coverage (\u003cb\u003eTable S2\u003c/b\u003e). A MOI of 150 and six hours of invasion were determined to be the optimal assay conditions, which ensured 100% bacterial growth in wells, while maintaining the robustness of the assay (Z\u0026rsquo; \u0026gt; 0.4; and S/B value\u0026thinsp;\u0026gt;\u0026thinsp;2) \u003csup\u003e30,31\u003c/sup\u003e. Cell detachment resulting in empty beads were observed after four hours of invasion, this phenomenon was observed to increase with time. At 4 and 6 hours of invasion, ~\u0026thinsp;80\u0026ndash;90% of beads had attached Caco-2 cells in comparison to pre-invaded beads. Induction was not suitable for screening after 6 hours of invasion. Other testing conditions did not meet validator standards. MOIs of 5 and 10 were suboptimal as they required too much time for invasion and the robustness of the assay could not be maintained. An MOI of 100 demonstrated improved performance with perfect coverage after six and eight hours of invasion; however, the Z\u0026rsquo; values were not stable and fell out of standard range (Z\u0026rsquo; \u0026lt; 0.4). To reduce the number of beads required for primary screening, three different bead concentrations (1,000, 2,000 and 4,000 beads/ml) were tested, but only the 4,000 beads/ml gave a suitable for a high-throughput screening assay (Z-score (mean Z\u0026rsquo; = 0.57) and S/B values\u0026thinsp;\u0026gt;\u0026thinsp;2-fold).\u003c/p\u003e \u003cp\u003eWe evaluated the invasive efficiency of \u003cem\u003eShigella flexneri\u003c/em\u003e by defining invasive efficiency as the percentage of intracellular bacteria (CFU/ml) to total number of bacteria (CFU/ml) (%) at the same duration of invasion (\u003cb\u003eTable S3\u003c/b\u003e). The \u003cem\u003eS. flexneri\u003c/em\u003e SF_nanoluc strain, \u003cem\u003eShigella flexneri\u003c/em\u003e serotype 2a 2457T carrying reporter plasmid pMK-RQ_tac\u0026thinsp;+\u0026thinsp;nanoluc (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e), with an invasion efficiency of 0.083%, exhibited the highest invasion efficiency at six hours of invasion. This score resulted in approximately eight invading bacteria for every 10\u003csup\u003e4\u003c/sup\u003e CFU input. In terms of bacteria per cell or bead, we estimated seven intracellular \u003cem\u003eShigella flexneri\u003c/em\u003e for every 100 cells or ~\u0026thinsp;15 \u003cem\u003eShigella flexneri\u003c/em\u003e for each bead.\u003c/p\u003e \u003cp\u003eWe next assessed dose-response potential of the assay by measuring the intra-assay and inter-assay of 11 different commercially available antimicrobials (\u003cb\u003eTable S4\u003c/b\u003e). IC\u003csub\u003e50\u003c/sub\u003e values were used to calculate coefficient of variances in percentage (\u003cb\u003eTable S7\u003c/b\u003e), then we determined the repeatability and reproducibility of this Caco-2 cell-based assay in case of maintained robustness of the assay. The intra-assay CV and the inter-assay CV across 11 drugs ranged from 0.16\u0026ndash;8.53% and from 1.43\u0026ndash;14.82%, respectively. The intra-assay CV (\u0026lt;\u0026thinsp;10%) demonstrated a high degree of reproducible in each batch, and the inter-assay CV (\u0026lt;\u0026thinsp;15%) demonstrated acceptable reproducibility between batches of the cellular assay.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eHigh-throughput screening process validations\u003c/h3\u003e\n\u003cp\u003eA validation set of chemicals, comprised of ~\u0026thinsp;10,000 compounds and representing a wide diversity of chemotypes present within the GSK chemical collection, was used to evaluate the HTS assay configuration. The assay displayed an average Z\u0026rsquo; factor of 0.48 and the obtained cut-off (average response of the sample distribution\u0026thinsp;+\u0026thinsp;3SD) was 48.87%. The chemical hit rate was ~\u0026thinsp;2.5% (i.e. 247 active compounds), but signal patterns were detected after analysing the dataset by using ActivityBase software, signifying the errors and noise of dispense patterns caused by unevenly dispensed beads. The noisy plates were re-run.\u003c/p\u003e \u003cp\u003eAiming to identify hits with novel modes of action, we performed structural screening, discarding 43 compounds that belonged to known drug lines, including dihydrofolate reductases (DHFR), quinolones, and bacterial topoisomerase (BTIs). We then performed a dose-response assay on the 204 remaining compounds. Finally, four compounds gave dose response to the targeted screening \u003cem\u003eS. flexneri\u003c/em\u003e organisms, three of them were active in both replicates and had IC\u003csub\u003e50\u003c/sub\u003e\u0026thinsp;\u0026lt;\u0026thinsp;10\u0026micro;M and the remaining chemical was active in one of two replicates with an IC\u003csub\u003e50\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;36.31\u0026micro;M. These compounds were further analysed for cytotoxicity in HepG2 cells and subjected to physical characterization (\u003cb\u003eTable S5\u003c/b\u003e).\u003c/p\u003e\n\u003ch3\u003eHTS primary screening campaign\u003c/h3\u003e\n\u003cp\u003eThe primary screening campaign of 518,282 compounds was performed in different 49 runs utilising 1,521 individual plates, the workflow is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. The optimal conditions for HTS assay, which using the 3-D Caco-2 cells along with \u003cem\u003eShigella flexneri\u003c/em\u003e SF_nano strain for infectivity with a MOI of 150 in six hours, following by gentamicin protection and diluted to 4,000beads/ml, were applied before 25\u0026micro;l invasive mixture was dispensed into prepared plates for screening. Overall, 45 plates were re-run because their Z\u0026rsquo; values were \u0026lt;\u0026thinsp;0.4 (\u003cb\u003eFigure S3\u003c/b\u003e). The robustness of the assay was maintained with an average Z\u0026rsquo; factor of 0.55. Compounds with inhibition rates greater than the robust cutoff value of 41.48% were considered active, with 14,451 compounds selected. Considering false signal patterns, 14,451 primary hits were screened again for confirmation at 10\u0026micro;M in the same assay format, representing a hit rate of 0.44% (2,283 compounds). After structural screening, we discarded 221 non-novel antimicrobial compounds, including BTIs, macrolides, quinolones, and DHFRs. The final list of 2,062 compounds were progressed into a dose-response assay, removing those with IC\u003csub\u003e50\u003c/sub\u003e\u0026thinsp;\u0026gt;\u0026thinsp;25\u0026micro;M. As a result, 225 compounds exhibited inhibition with IC\u003csub\u003e50\u003c/sub\u003e\u0026thinsp;\u0026le;\u0026thinsp;25\u0026micro;M, but 15 compounds were not accessible. Therefore, to factor out compounds that may be false positives due to nano-luciferase inhibition, 210 available preliminary hits advanced from the intracellular survival screen were tested in a nanoluciferase-interference assay in dose-response assays starting at 100\u0026micro;M. Ultimately, we identified15 compounds with a pIC50\u0026thinsp;\u0026lt;\u0026thinsp;4 and a pIC50 ratio between the Caco-2 cell assay and interference assay\u0026thinsp;\u0026gt;\u0026thinsp;1, meaning no extracellular-nanoluciferase activity were identified to have intracellular activity against \u003cem\u003eS. flexneri\u003c/em\u003e in this assay format (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAfter chemical review, the 15 hit compounds were reorganized based on structure similarity into 13 clusters. Two chemicals were considered as a series and the remaining 11 clusters were singletons (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). To ensure that the compounds were suitable starting points for qualify hits and to determine the toxic potential of the new chemical entities, an inspection of the compounds for structural features that may impact toxicity was conducted, followed by \u003cem\u003ein vitro\u003c/em\u003e cytotoxicity testing on HepG2 cells. Four of the 15 structures carried an NO\u003csub\u003e2\u003c/sub\u003e moiety, including two hits of Series 1, one hit of Series 2, and Singleton 6. Although the antibacterial activity of nitro-containing molecules is known to be broad, nitro groups have been extensively associated with mutagenicity and genotoxicity. Therefore, the compounds that were representative of the entire clusters were additionally tested on HepG2 cells, four compounds (1, 2, 8, and 11) were found to exhibit HepG2 cytotoxicity. Series 1 (compounds 1 and 2) and singleton 4 (compound 8) demonstrated some concert of progression considering its selectivity (Caco-2 pIC50 vs HepG2) and its structure alert (\u003cb\u003eTable S6)\u003c/b\u003e. The remaining 12 compounds (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e1\u003c/span\u003e), represented by Series 2 and Singletons 1\u0026ndash;3, 5, 6, and 8\u0026ndash;11, were of interest as starting points for further optimization toward the drug discovery for treatment of shigellosis.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003ePathogenic bacteria have been engaged in a sustained \u0026lsquo;arms race\u0026rsquo; with antimicrobials since their introduction and antimicrobial resistant \u003cem\u003eShigella\u003c/em\u003e has emerged as a significant global threat to public health \u003csup\u003e7\u003c/sup\u003e. Novel treatments and approaches are urgently to combat infections caused by MDR \u003cem\u003eShigella\u003c/em\u003e. To aid in addressing this issue we developed, validated and exploited a HTS Caco-2 cell-based system to screen a large compound library against \u003cem\u003eShigella flexneri\u003c/em\u003e. Compared to traditional cell monolayer screening, this assay system retained the disease-relevant cellular architecture, which improves the success of translating hits from the screen to cell-based or \u003cem\u003ein vivo\u003c/em\u003e infection models. Additionally, this approach simultaneously has the common advantages of all HTS systems, including, scalability, amenable to automation, simple and rapid transfer of reagents.\u003c/p\u003e \u003cp\u003eOur project was initiated by establishing an HTS assay using a spinner flask, and experiments were performed in a 384-well format. We demonstrated that there was no significant difference in the structural integrity of the Caco-2 cells between microcarrier beads (Cytodex\u003csup\u003e3\u003c/sup\u003e) and the monolayer format; \u003cem\u003eShigella flexneri\u003c/em\u003e still demonstrated invasion and spread from cell-to-cell in both systems. One of the main advantages of a bead-based system is a large surface area to grow cells to high density. This approach facilitates scale-up and reduces the floor space and incubator volume required for a given-sized manufacturing operation and eases the precise process control in a large-scale bioreactor. Consequently, the number of compounds screened per batch can be easily scaled up. However, we found that even after optimising conditions, bacterial invasion rates were substantially lower in comparison to the monolayer format \u003csup\u003e32,33\u003c/sup\u003e. Notably, not every Caco-2 cell was invaded by a \u003cem\u003eShigella\u003c/em\u003e organism and not every microcarrier bead possessed an invaded cell. Therefore, we compensated for this lower rate of invasion by increasing the number of beads to ensure each well contained intracellular \u003cem\u003eShigella\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eIn generating this new high throughput approach we observed a high false discovery rate or chemicals that did not elaborate a dose-response. After the primary screen, 2.79% of chemical were retained, with only 0.44% remaining after rescreening. This discrepancy may have been caused by false signal patterns recognized throughout the screen, such as edge effects (caused by evaporation), patterns on a plate in general, and errors in adding assay components (dispenser distribution). In addition, there were compounds which were latterly found to inhibit the luciferase reporting system. Nonetheless, we employed various orthogonal screenings so to filter out false positives. We next analysed what chemical scaffolds had been elucidated by our screening effort. To ensure that the compounds were suitable starting points for qualifying hits, we characterized the potential toxicity of the new chemical entities. As a result, 12 compounds represented by Series 2 and Singletons 1\u0026ndash;3, 5, 6, and 8\u0026ndash;11 were prioritized. Considering their overall profile in terms of physchem properties (most of them with predicted high permeability and different range of solubility) and initial toxicity profile, although the potency of these compounds was low (pIC50 values between 4.6 and 5.8), they could be considered as novel starting points for drug development against \u003cem\u003eShigella\u003c/em\u003e (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Additional research is currently being performed to assess these hits and convert them into developable hit series, these compounds may be developed into tool compounds for antibacterial agents with preferential activity against intracellular bacteria. The validation of these types of tool compound will aid in the future design of novel antibacterial agents with intracellular activity. Additionally, various studies can be planned in the future using such tool compounds to better understand the mode of action of the new chemicals.\u003c/p\u003e \u003cp\u003eIn summary, we have shown that it is possible to integrate three-dimensional cell culture into an assay system that is amenable to HTS using microcarriers. We have demonstrated the capability of this system in an HTS of 500,000 compounds and have identified novel hits that have activity against intracellular \u003cem\u003eShigella flexneri\u003c/em\u003e. The experience in developing this assay system has enabled us to gain insight on assay optimization issues of which we have demonstrated steps to mitigate them. We think that the microcarrier-based assay system has utility in HTS to drive the hit discovery program, and we envisage this assay system to be used in other intracellular bacterial infection models to pioneer hit discovery against gastrointestinal pathogens.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eBacterial strains, Cell lines, Growth media and Reagents\u003c/h2\u003e \u003cp\u003eTh bacterial strains, cell lines and plasmids used in this study are listed in \u003cb\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e. \u003cem\u003eShigella flexneri\u003c/em\u003e isolates were routinely cultured on Luria-Bertani (LB) broth or LB agar. Where required, the growth medium was supplemented with antimicrobials at the following concentration: Kanamycin (50ug ml\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) for \u003cem\u003eShigella flexneri\u003c/em\u003e serotype 2a 2457T (ATCC\u0026reg; 700930\u0026trade;) carrying plasmid pMK-RQ_tac\u0026thinsp;+\u0026thinsp;nanoluc (SF_nanoluc), and ampicillin (100ug ml\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) for \u003cem\u003eShigella flexneri\u003c/em\u003e ATCC 12022GFP.\u003c/p\u003e \u003cp\u003eCaco-2 cells were grown in 5-layer flasks (Nunc/ThermalFisher) with basic EMEM (Sigma-Aldrich) supplemented with 10% FBS (Gibco), 1% L-glutamine in a humidified atmosphere of 5% CO\u003csub\u003e2\u003c/sub\u003e at 37\u003csup\u003eo\u003c/sup\u003eC. The medium was changed every 72 hours until 80\u0026ndash;90% confluent layers were generated.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eGeneration of the reporter strain\u003c/h3\u003e\n\u003cp\u003eThe synthetic gene Tac\u0026thinsp;+\u0026thinsp;Nanoluc (593 bp) was assembled from synthetic oligonucleotides and the fragment was inserted into the pMK-RQ vector, yielding the 18ADIM2C_Tac\u0026thinsp;+\u0026thinsp;Nanoluc_pMK-RQ (KanR) construct (hereafter referred to as pMK-RQ-nl). \u003cem\u003eEscherichia coli\u003c/em\u003e K12 DH10B\u0026trade; T1R was used for the propagation of the plasmid. Plasmid DNA was purified from transformed colonies and sequenced (Dynamimed, Spain) for verification. This process was conducted by contract to Invitrogen (Thermo Fisher Scientific, USA). Electroporation of pMK-RQ-nl into \u003cem\u003eS. flexneri\u003c/em\u003e 2457T was performed according to a standard protocol (NEB Electroporation protocol, C2986) yielding the \u003cem\u003eS. flexneri\u003c/em\u003e serotype 2a 2457T-nl (or SF_nanoluc) used in this study.\u003c/p\u003e\n\u003ch3\u003eHigh-throughput Screening Assay\u003c/h3\u003e\n\u003cp\u003e \u003cem\u003eShigella flexneri\u003c/em\u003e SF_nanoluc was grown overnight on a 50mgL\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e kanamycin LB plate at 37\u003csup\u003eo\u003c/sup\u003eC. Bacteria were scraped from the plate surface and sub-cultured in warm supplemented EMEM medium for 2 hours (at 37\u003csup\u003eo\u003c/sup\u003eC and 200rpm) and diluted to get OD around 0.3 at 625nm. Induction was performed in 500ml or 1,000ml siliconized spinner flasks under closed culture conditions. 200mg autoclaved Cytodex\u003csup\u003e3\u003c/sup\u003e beads (GE Healthcare), rehydrated with cation-free PBS (Sigma), were washed with supplemented EMEM medium and inoculated with 1.7\u0026nbsp;million Caco-2 cells/ml in half of the desired culture volume of the medium. The spinner vessel was kept static for 4 hours in a humidified atmosphere of 5% CO\u003csub\u003e2\u003c/sub\u003e at 37\u003csup\u003eo\u003c/sup\u003eC, allowing Caco-2 cells to bind to beads. The remaining medium was added into the vessel and stirred at 32.5rpm. The medium was changed every 72 hours, and the induction was used on day 17 to ensure Caco-2 cells were fully differentiated.\u003c/p\u003e \u003cp\u003eCell concentration induction was measured by NucleoCounter\u0026reg; NC-100\u0026trade; (ChemoMetec Version 2.4). Cellular invasion was initiated by adding the bacterial stock into cell induction with optimal MOI, and the vessel was stationary for an hour before stirring at 22.5rpm for 5 hours in a humid atmosphere of 5% CO\u003csub\u003e2\u003c/sub\u003e at 37\u003csup\u003eo\u003c/sup\u003eC. After incubation, culture media was discarded by centrifugation; and a gentamicin protection assay was performed, in which cell beads were washed twice with a warm medium and EMEM containing gentamicin (100ug ml\u003csup\u003e-1\u003c/sup\u003e) was added to the chamber of the system. The cell culture was incubated for 1 hour with stirring and then washed three times with a warm medium before 25\u0026micro;l of the invasive mixture was dispensed into each well. Plates were incubated for two days in a humid atmosphere of 5% CO\u003csub\u003e2\u003c/sub\u003e at 37\u003csup\u003eo\u003c/sup\u003eC before proceeding to develop the luminescent signal by adding 10\u0026micro;l mixture of nanoluciferase substrate (Nano-Glo\u0026reg; Luciferase Assay System, N1130 - Promega, Spain) into each well. After standing at room temperature for 30 minutes, luminescent signals were quantified by the Envision Reader (PerkinElmer) in focus luminescent mode. For the bacterial tracking, we used the green fluorescent protein in \u003cem\u003eS. flexneri\u003c/em\u003e ATCC 12022GFP to monitor the invasion. A confocal HCS microscope (Opera by PerkinElmer) was used to confirm the invasion of \u003cem\u003eS. flexneri\u003c/em\u003e into Caco-2 beads (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eCompounds and assay plate preparation\u003c/h2\u003e \u003cp\u003eThe HTS assay was performed in 384-well black clear bottom plates (Greiner). Screening compounds were prepared with an appropriate DMSO concentration (0.5% in cell-based assay). Every assay plate contained 16 wells of DMSO as negative controls and 16 wells of moxifloxacin at a final concentration of 20\u0026micro;M as positive controls. These controls were used to monitor assay quality through determination of Z\u0026rsquo; as well as normalizing the data on a per-plate basis.\u003c/p\u003e \u003cp\u003eTo assess the quality of the dose-response assay, 11 commercially available antimicrobials were used to evaluate the reliability of the HTS assay. The co-efficiency of variance (CV%) was calculated by accessing the IC50 (\u0026micro;M) deviation of selected positive controls. Intra-assay reproducibility was determined by comparing the IC50 that were generated in the same run of the two replicates. Inter-assay reproducibility was assessed by comparing the IC50 generated by two replicates of each day over two days. Intra-assay %CV should be \u0026lt;\u0026thinsp;10%, while inter-assay %CV should be \u0026lt;\u0026thinsp;15%. To validate the HTS assay configuration, a set of 10 thousand compounds was assayed in 384-well plates at a final concentration of 10\u0026micro;M. Assay were run in duplicate and confirmed if needed, a validation data was assessed to analyze the pros and cons of the whole procedure.\u003c/p\u003e \u003cp\u003eThe primary screening campaign of 520,000 compounds was screened at the single shot of 10\u0026micro;M final concentration; active compounds were confirmed at the same concentration (10\u0026micro;M). Following screening, all non-novel structures were removed, and the remaining compounds were subjected to a dose response (DR) assay. Those compounds that had DR to targeted organism of \u0026le;\u0026thinsp;25\u0026micro;M were progressed to toxicity and structure analysis.\u003c/p\u003e \u003cp\u003eTo select the most drug-like compounds, Fasted State Simulated Intestinal Fluid (FaSSIF) is used to test the dissolution and permeability of compounds, as well as their stability in the gastrointestinal tract \u003csup\u003e34\u003c/sup\u003e. Human ether-\u0026agrave;-go-go-related gene (hERG) is the gene that encodes a potassium channel protein, which is essential for normal heart rhythm. It is important to screen compounds for hERG blockade \u003csup\u003e35\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eOptimizing invasive conditions\u003c/h2\u003e \u003cp\u003eTo determine the optimal condition (Multiplicity of infection - MOI and time invasion) for the invasive process, the induction products were infected with SF_nanoluc at an MOI ranging from 5\u0026ndash;150 (5, 10, 100, 150) for five time points (4 hours, 6 hours, 8 hours, 10 hours and \u0026gt;\u0026thinsp;16 hours). After gentamicin protection, 25\u0026micro;l of invasion was dispensed into prepared 384-well plates, the number of beads\u0026thinsp;\u0026gt;\u0026thinsp;5,000 beads/ml. The experiments were repeated three times. Results were assessed by Z\u0026rsquo; factors (Z\u0026rsquo; \u0026ge; 0.4) \u003csup\u003e31\u003c/sup\u003e, percentage of bacterial growth in wells, and bead coverage observed by microscopy. If the coverage of beads with cells was \u0026lt;\u0026thinsp;70%, the experiment was cancelled (Data do not show).\u003c/p\u003e \u003cp\u003eTo increase the number of plates per run, the number of beads per well was optimized with three different bead concentrations (1,000, 2,000, and 4,000 beads/ml). 10\u0026micro;l homogenized invasive mixture was dropped on each counting grid of a Neubauer chamber, and the number of beads (with or without attached cells) were enumerated by microscopy. We repeated three times and calculated the average before dispensing into each well of prepared plates. The effects of bead concentration on Z\u0026rsquo; values were considered as a selective indicator.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eInvasive efficiency evaluation\u003c/h2\u003e \u003cp\u003eA bacterial quantification assay was adapted and modified from a previous study to calculate the invasive rate of \u003cem\u003eShigella\u003c/em\u003e into Caco-2 cells \u003csup\u003e3\u003c/sup\u003e\u0026gt;. After inoculation, 50ml homogenized invasion was withdrawn at three time points (2 hours, 4 hours, and 6 hours) and equally divided into two tubes (A and B) at each time point. Tube A went directly to cell lysis. After being washed with fresh medium and centrifuged at 8, 000rpm for 10 minutes twice to collect all beads and bacterial pellets, the mixture was resuspended in 5ml of PBS and divided into five different 1.5ml tubes (1ml/tube). One was used for cell counting and another for bead counts. The remaining three tubes were lysed with 0.1% saponin in one hour and the dilutions of lysate were plated on 50mgL\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e kanamycin LB plates for quantification. Tube B was used to determine bacterial invasion; extracellular bacteria were killed with gentamicin (100mgL\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) for 1 hour. Then cultured beads were washed twice with PBS at 800rpm and re-suspended in 5ml PBS. Then, the same process of tube A was applied. The experiments were repeated four times. Invasive rates were calculated as the percentage of alive \u003cem\u003eShigella\u003c/em\u003e after gentamicin protection compared to the total number of \u003cem\u003eShigella\u003c/em\u003e before being treated with gentamicin.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eSucrase assay\u003c/h2\u003e \u003cp\u003eSucrase is an important digestive enzyme secreted in the small intestine on the brush border and catalyses the hydrolysis of sucrose to its subunits of fructose and glucose. Sucrase is increased during differentiation of the enterocytes and is considered a reliable indicator of Caco-2 cell differentiation \u003cem\u003ein vitro\u003c/em\u003e \u003csup\u003e37,3\u003c/sup\u003e\u0026gt;. We monitored sucrose production at day 8, 12, 15 19 and 21 after seeding, in the cytodex\u003csup\u003e3\u003c/sup\u003e beads system as well as in monolayers. Measurements were conducted by the fluorescent Amplex\u0026reg; Red Glucose/Glucose Oxidase Assay Kit (A22189, Invitrogen, Molecular probes) following the manufacturer\u0026rsquo;s instructions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eAlkaline phosphatase (ALP) activity\u003c/h2\u003e \u003cp\u003eThe intestinal alkaline phosphatase encodes a digestive brush-border enzyme, which is highly upregulated during small intestinal epithelial cell differentiation\u003csup\u003e3\u003c/sup\u003e\u0026gt;. ALP activity was measured in the supernatant of cytodex\u003csup\u003e3\u003c/sup\u003e and monolayer Caco-2 cells during differentiation 8, 12, 15, 19 and 21 days after seeding. Measurements were performed by Alkaline Phosphatase Assay Kit Colorimetric (Abcam, ab83369) according to the manufacturer's instructions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eTight junctions of Caco-2 cells\u003c/h2\u003e \u003cp\u003eThe primary function of tight junctions (TJ) in epithelial cells is to create a regulated barrier in the extracellular space serve to regulate the passage of ions and molecules between cells and mark the division between apical and basolateral surfaces of cells in cellular differentiation \u003csup\u003e4\u003c/sup\u003e\u0026gt;. TJs are essential for the polarization of epithelial cells. Intracellular transduction pathways can regulate the integrity of the tight junctions. We identified and characterized ZO-1 as a peripheral membrane protein specifically associated with the cytoplasmic surface of tight junctions. We used Immunofluorescence of Human Caco-2 cells stained with Mouse anti-ZO-1 Monoclonal Antibody - Alexa Fluor\u0026reg; 488 (Product #339188, Thermo Fisher Scientific). DNA is counter-stained with blue Hoechst 33258 (Product #H3569, Thermo Fisher Scientific). Immunofluorescence analysis of ZO-1 was performed using 90% /100% confluent log phase Caco-2 cells in the beads system. Cells were fixed with 4% paraformaldehyde for 30 minutes, permeabilized with 0.1% Triton\u0026trade; X-100 for 10 minutes and blocked with 1% BSA for 30 minutes at room temperature. The cells were labelled with ZO-1 Monoclonal Antibody (ZO1-1A12), Alexa Fluor 488 at 5\u0026micro;g/mL in 0.1% BSA and incubated overnight at room temperature. Nuclei were stained with DAPI at 1ug/mL concentration without washing.\u003c/p\u003e \u003cp\u003e \u003cem\u003eInterference with nanoluciferase reporter assay.\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe main disadvantage of luciferase-based assays is interference. Therefore, an interference assay was established to identify compounds that might have interference with the nanoluciferase reporter system or luminescence readout. A 3-fold serial dilution, starting at 100\u0026micro;M, of each compound was dispended into 1,536-well plates. A volume of 5\u0026micro;l of fresh bacterial broth was added into each well to obtain 10\u003csup\u003e6\u003c/sup\u003eCFU/ml, which were incubated at 37\u003csup\u003eo\u003c/sup\u003eC overnight. Subsequently, 5\u0026micro;l of nanoluciferase substrate solution (Promega) was added to each well, which were place in the dark for 30 minutes before reading the luminescence by Envision device. The pIC50 ratio between the Caco-2 cell assay and interference assay had to be \u0026gt;\u0026thinsp;1. Compounds with no extracellular-nanoluciferase activity (pIC50\u0026thinsp;\u0026lt;\u0026thinsp;4) were considered non-interference compounds. Results were compared with those obtained from an extracellular assay using the same organism but using resazurin as a readout (surrogate of bacterial viability) to assess if activity was dependant on bacterial extracellular or enzymatic inhibition. Results were compared against data from intracellular-THP1 screening, to calculate shift between both luminescence assays.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eComputational and statistical analysis\u003c/h2\u003e \u003cp\u003eThe raw data of the Envision device was imported and analysed in ActivityBase and Spotfire version 10.3.1.18 (TIBCO). pIC50 values (pIC50 = -log10(IC50 (M)) were obtained using the Activity Base XE nonlinear regression bundle. The Cut-off values (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;3SD), Z\u0026rsquo; factor and signal-to-base were produced automatically by ActivityBase, or manually calculated with equations in table S7. Other data were analysed in Excel (Microsoft Office) and plots were generated using Prism version 6.07 (Graphpad) or IC\u003csub\u003e50\u003c/sub\u003e calculator tool (ATT Bioquest, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.aatbio.com/tools/ic50-calculator\u003c/span\u003e\u003cspan address=\"https://www.aatbio.com/tools/ic50-calculator\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003e3-D\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ethree-dimension\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eHTS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ehigh-throughput screening\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eGEMS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eGlobal Enteric Multicentre Study\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eRWV\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003erotating wall vessel\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eMOI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eMultiplicity of infection\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003estandard deviation\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eIFI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eInhibition Frequency Index\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eIC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003einhibition concentration\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCV\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eco-efficiency of variance\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eS/B\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eSignal-to-base\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePFI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eProperty Forecast Index\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eFaSSIF\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eFasted State Simulated Intestinal Fluid\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ehERG\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eHuman ether-\u0026agrave;-go-go-related gene\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eFig\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eFigure\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":" \u003cp\u003e \u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e \u003cp\u003eNot required\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent for publication\u003c/strong\u003e \u003cp\u003eNot required\u003c/p\u003e \u003c/p\u003e\u003cp\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis project TC239 was co-funded by the Tres Cantos Open Lab Foundation. Stephen Baker is supported by a Wellcome senior research fellowship (215515/Z/19/Z). \u003cem\u003eThe funders\u003c/em\u003e had \u003cem\u003eno role\u003c/em\u003e in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eConceptualization: Stephen BakerFormal analysis: Voong Vinh Phat, Andrew Lim, Sonia Lozano-AriasProvided supervision: Stephen BakerMethodology: Voong Vinh Phat, Andrew Lim, Sonia Lozano-AriasWriting original draft: Voong Vinh Phat, Andrew Lim, Sonia Lozano-AriasReview and editing: All authorsRead and approved the final version of the manuscript: All authors\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eWe would like to express our sincere thanks to Dr. Janneth Rodrigues at OpenLab Foundation for her kind help. We also thank Ms. Ana Isabel Bardera for her help with our experiments.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eAll data are presented in the manuscript and raw data are freely available upon request from the corresponding author.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eKotloff, K. L., Riddle, M. S., Platts-Mills, J. A., Pavlinac, P. \u0026amp; Zaidi, A. K. M. Shigellosis. \u003cem\u003eLancet\u003c/em\u003e\u003cstrong\u003e391\u003c/strong\u003e, 801\u0026ndash;812 (2018).\u003c/li\u003e\n\u003cli\u003eKotloff, K. L. \u003cem\u003eet al.\u003c/em\u003e The Global Enteric Multicenter Study (GEMS) of Diarrheal Disease in Infants and Young Children in Developing Countries: Epidemiologic and Clinical Methods of the Case/Control Study. \u003cem\u003eClin. Infect. Dis.\u003c/em\u003e\u003cstrong\u003e55\u003c/strong\u003e, S232\u0026ndash;S245 (2012).\u003c/li\u003e\n\u003cli\u003eRaso, M. M., Arato, V., Gasperini, G. \u0026amp; Micoli, F. Toward a Shigella Vaccine: Opportunities and Challenges to Fight an Antimicrobial-Resistant Pathogen. \u003cem\u003eInt. J. Mol. Sci.\u003c/em\u003e\u003cstrong\u003e24\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eMani, S., Wierzba, T. \u0026amp; Walker, R. I. Status of vaccine research and development for Shigella. \u003cem\u003eVaccine\u003c/em\u003e\u003cstrong\u003e34\u003c/strong\u003e, 2887\u0026ndash;2894 (2016).\u003c/li\u003e\n\u003cli\u003eBaker, K. S. \u003cem\u003eet al.\u003c/em\u003e Travel- and Community-Based Transmission of Multidrug-Resistant Shigella sonnei Lineage among International Orthodox Jewish Communities. \u003cem\u003eEmerg. Infect. Dis.\u003c/em\u003e\u003cstrong\u003e22\u003c/strong\u003e, 1545\u0026ndash;1553 (2016).\u003c/li\u003e\n\u003cli\u003eChung The, H. \u003cem\u003eet al.\u003c/em\u003e Dissecting the molecular evolution of fluoroquinolone-resistant Shigella sonnei. \u003cem\u003eNat. Commun.\u003c/em\u003e\u003cstrong\u003e10\u003c/strong\u003e, (2019).\u003c/li\u003e\n\u003cli\u003eWilliams, P. C. M. \u0026amp; Berkley, J. A. Guidelines for the treatment of dysentery (shigellosis): a systematic review of the evidence. \u003cem\u003ePaediatr. Int. Child Health\u003c/em\u003e\u003cstrong\u003e38\u003c/strong\u003e, S50\u0026ndash;S65 (2018).\u003c/li\u003e\n\u003cli\u003eSilver, L. L. Challenges of Antibacterial Discovery. \u003cem\u003eClin. Microbiol. Rev.\u003c/em\u003e\u003cstrong\u003e24\u003c/strong\u003e, 71\u0026ndash;109 (2011).\u003c/li\u003e\n\u003cli\u003ePayne, D. J., Gwynn, M. N., Holmes, D. J. \u0026amp; Pompliano, D. L. Drugs for bad bugs: confronting the challenges of antibacterial discovery. \u003cem\u003eNat. Rev. Drug Discov.\u003c/em\u003e\u003cstrong\u003e6\u003c/strong\u003e, 29\u0026ndash;40 (2007).\u003c/li\u003e\n\u003cli\u003eTommasi, R., Brown, D. G., Walkup, G. K., Manchester, J. I. \u0026amp; Miller, A. A. ESKAPEing the labyrinth of antibacterial discovery. \u003cem\u003eNat. Rev. Drug Discov.\u003c/em\u003e\u003cstrong\u003e14\u003c/strong\u003e, 529\u0026ndash;542 (2015).\u003c/li\u003e\n\u003cli\u003eWalsh, C. Where will new antibiotics come from? \u003cem\u003eNat. Rev. Microbiol.\u003c/em\u003e\u003cstrong\u003e1\u003c/strong\u003e, 65\u0026ndash;70 (2003).\u003c/li\u003e\n\u003cli\u003eFischbach, M. A. \u0026amp; Walsh, C. T. Antibiotics for Emerging Pathogens. \u003cem\u003eScience (80-. ).\u003c/em\u003e\u003cstrong\u003e325\u003c/strong\u003e, 1089\u0026ndash;1093 (2009).\u003c/li\u003e\n\u003cli\u003eO\u0026rsquo;Shea, R. \u0026amp; Moser, H. E. Physicochemical Properties of Antibacterial Compounds: Implications for Drug Discovery. \u003cem\u003eJ. Med. Chem.\u003c/em\u003e\u003cstrong\u003e51\u003c/strong\u003e, 2871\u0026ndash;2878 (2008).\u003c/li\u003e\n\u003cli\u003eMugumbate, G. \u0026amp; Overington, J. P. The relationship between target-class and the physicochemical properties of antibacterial drugs. \u003cem\u003eBioorg. Med. Chem.\u003c/em\u003e\u003cstrong\u003e23\u003c/strong\u003e, 5218\u0026ndash;5224 (2015).\u003c/li\u003e\n\u003cli\u003eArtursson, P., Palm, K. \u0026amp; Luthman, K. Caco-2 monolayers in experimental and theoretical predictions of drug transport. \u003cem\u003eAdv. Drug Deliv. Rev.\u003c/em\u003e\u003cstrong\u003e64\u003c/strong\u003e, 280\u0026ndash;289 (2012).\u003c/li\u003e\n\u003cli\u003eDelie, F. \u0026amp; Rubas, W. A Human Colonic Cell Line Sharing Similarities With Enterocytes as a Model to Examine Oral Absorption: Advantages and Limitations of the Caco-2 Model. \u003cem\u003eCrit. Rev. Ther. Drug Carr. Syst.\u003c/em\u003e\u003cstrong\u003e14\u003c/strong\u003e, 66 (1997).\u003c/li\u003e\n\u003cli\u003eLamson, N. G., Ball, R. L., Fein, K. C. \u0026amp; Whitehead, K. A. Thrifty, Rapid Intestinal Monolayers (TRIM) Using Caco-2 Epithelial Cells for Oral Drug Delivery Experiments. \u003cem\u003ePharm. Res.\u003c/em\u003e\u003cstrong\u003e36\u003c/strong\u003e, 1\u0026ndash;12 (2019).\u003c/li\u003e\n\u003cli\u003eJakob, P. H. \u003cem\u003eet al.\u003c/em\u003e A 3-D cell culture system to study epithelia functions using microcarriers. \u003cem\u003eCytotechnology\u003c/em\u003e\u003cstrong\u003e68\u003c/strong\u003e, 1813\u0026ndash;1825 (2016).\u003c/li\u003e\n\u003cli\u003eKato, D. The design of polymer microcarrier surfaces for enhanced cell growth. \u003cem\u003eBiomaterials\u003c/em\u003e\u003cstrong\u003e24\u003c/strong\u003e, 4253\u0026ndash;4264 (2003).\u003c/li\u003e\n\u003cli\u003eWU, S., LIU, C. \u0026amp; LIAN, W. Optimization of microcarrier cell culture process for the inactivated enterovirus type 71 vaccine development. \u003cem\u003eVaccine\u003c/em\u003e\u003cstrong\u003e22\u003c/strong\u003e, 3858\u0026ndash;3864 (2004).\u003c/li\u003e\n\u003cli\u003eChen, A. K.-L., Reuveny, S. \u0026amp; Oh, S. K. W. Application of human mesenchymal and pluripotent stem cell microcarrier cultures in cellular therapy: Achievements and future direction. \u003cem\u003eBiotechnol. Adv.\u003c/em\u003e\u003cstrong\u003e31\u003c/strong\u003e, 1032\u0026ndash;1046 (2013).\u003c/li\u003e\n\u003cli\u003eDrummond, C. G., Nickerson, C. A. \u0026amp; Coyne, C. B. A Three-Dimensional Cell Culture Model To Study Enterovirus Infection of Polarized Intestinal Epithelial Cells. \u003cem\u003emSphere\u003c/em\u003e\u003cstrong\u003e1\u003c/strong\u003e, 1\u0026ndash;17 (2016).\u003c/li\u003e\n\u003cli\u003eStraub, T. M. \u003cem\u003eet al.\u003c/em\u003e In vitro cell culture infectivity assay for human noroviruses. \u003cem\u003eEmerg. Infect. Dis.\u003c/em\u003e\u003cstrong\u003e13\u003c/strong\u003e, 396\u0026ndash;403 (2007).\u003c/li\u003e\n\u003cli\u003eStraub, T. M. \u003cem\u003eet al.\u003c/em\u003e Human norovirus infection of Caco-2 cells grown as a three-dimensional tissue structure. \u003cem\u003eJ. Water Health\u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 225\u0026ndash;240 (2011).\u003c/li\u003e\n\u003cli\u003eKhoshnood, N. \u0026amp; Zamanian, A. A comprehensive review on scaffold-free bioinks for bioprinting. \u003cem\u003eBioprinting\u003c/em\u003e\u003cstrong\u003e19\u003c/strong\u003e, e00088 (2020).\u003c/li\u003e\n\u003cli\u003eFerruzza, S., Rossi, C., Scarino, M. L. \u0026amp; Sambuy, Y. A protocol for in situ enzyme assays to assess the differentiation of human intestinal Caco-2 cells. \u003cem\u003eToxicol. Vitr.\u003c/em\u003e\u003cstrong\u003e26\u003c/strong\u003e, 1247\u0026ndash;1251 (2012).\u003c/li\u003e\n\u003cli\u003eChen, L. \u003cem\u003eet al.\u003c/em\u003e Mechanistic studies of the transport of peimine in the Caco-2 cell model. \u003cem\u003eActa Pharm. Sin. B\u003c/em\u003e\u003cstrong\u003e6\u003c/strong\u003e, 125\u0026ndash;131 (2016).\u003c/li\u003e\n\u003cli\u003eValenzano, M. C. \u003cem\u003eet al.\u003c/em\u003e Remodeling of Tight Junctions and Enhancement of Barrier Integrity of the CACO-2 Intestinal Epithelial Cell Layer by Micronutrients. \u003cem\u003ePLoS One\u003c/em\u003e\u003cstrong\u003e10\u003c/strong\u003e, e0133926 (2015).\u003c/li\u003e\n\u003cli\u003eLiu, W. \u003cem\u003eet al.\u003c/em\u003e Intestinal Alkaline Phosphatase Regulates Tight Junction Protein Levels. \u003cem\u003eJ. Am. Coll. Surg.\u003c/em\u003e\u003cstrong\u003e222\u003c/strong\u003e, 1009\u0026ndash;1017 (2016).\u003c/li\u003e\n\u003cli\u003eInglese, J. \u003cem\u003eet al.\u003c/em\u003e High-throughput screening assays for the identification of chemical probes. \u003cem\u003eNat. Chem. Biol.\u003c/em\u003e\u003cstrong\u003e3\u003c/strong\u003e, 466\u0026ndash;479 (2007).\u003c/li\u003e\n\u003cli\u003eBar, H. \u0026amp; Zweifach, A. Z\u0026rsquo; Does Not Need to Be \u0026gt;\u0026thinsp;0.5. \u003cem\u003eSLAS Discov. Adv. Sci. Drug Discov.\u003c/em\u003e 247255522094276 (2020) doi:10.1177/2472555220942764.\u003c/li\u003e\n\u003cli\u003eHonma, Y. Effect of erythromycin on Shigella infection of Caco-2 cells. \u003cem\u003eFEMS Immunol. Med. Microbiol.\u003c/em\u003e\u003cstrong\u003e27\u003c/strong\u003e, 139\u0026ndash;145 (2000).\u003c/li\u003e\n\u003cli\u003eMounier, J., Vasselon, T., Hellio, R., Lesourd, M. \u0026amp; Sansonetti, P. J. Shigella flexneri enters human colonic Caco-2 epithelial cells through the basolateral pole. \u003cem\u003eInfect. Immun.\u003c/em\u003e\u003cstrong\u003e60\u003c/strong\u003e, 237\u0026ndash;248 (1992).\u003c/li\u003e\n\u003cli\u003eAmidon, G. L., Lennern\u0026auml;s, H., Shah, V. P. \u0026amp; Crison, J. R. A theoretical basis for a biopharmaceutic drug classification: the correlation of in vitro drug product dissolution and in vivo bioavailability. \u003cem\u003ePharm. Res.\u003c/em\u003e\u003cstrong\u003e12\u003c/strong\u003e, 413\u0026ndash;20 (1995).\u003c/li\u003e\n\u003cli\u003eVargas, H. M. \u003cem\u003eet al.\u003c/em\u003e Time for a Fully Integrated Nonclinical\u0026ndash;Clinical Risk Assessment to Streamline QT Prolongation Liability Determinations: A Pharma Industry Perspective. \u003cem\u003eClin. Pharmacol. Ther.\u003c/em\u003e\u003cstrong\u003e109\u003c/strong\u003e, 310\u0026ndash;318 (2021).\u003c/li\u003e\n\u003cli\u003eLuck, S. N., Bennett-Wood, V., Poon, R., Robins-Browne, R. M. \u0026amp; Hartland, E. L. Invasion of Epithelial Cells by Locus of Enterocyte Effacement-Negative Enterohemorrhagic Escherichia coli. \u003cem\u003eInfect. Immun.\u003c/em\u003e\u003cstrong\u003e73\u003c/strong\u003e, 3063\u0026ndash;3071 (2005).\u003c/li\u003e\n\u003cli\u003eGaland, G. Brush border membrane sucrase-isomaltase, maltase-glucoamylase and trehalase in mammals. Comparative development, effects of glucocorticoids, molecular mechanisms, and phylogenetic implications. \u003cem\u003eComp. Biochem. Physiol. Part B Comp. Biochem.\u003c/em\u003e\u003cstrong\u003e94\u003c/strong\u003e, 1\u0026ndash;11 (1989).\u003c/li\u003e\n\u003cli\u003eVan Beers, E. H. \u003cem\u003eet al.\u003c/em\u003e Lactase and sucrase-isomaltase gene expression during Caco-2 cell differentiation. \u003cem\u003eBiochem. J.\u003c/em\u003e\u003cstrong\u003e308\u003c/strong\u003e, 769\u0026ndash;775 (1995).\u003c/li\u003e\n\u003cli\u003eOlsen, L., Bressendorff, S., Troelsen, J. T. \u0026amp; Olsen, J. Differentiation-dependent activation of the human intestinal alkaline phosphatase promoter by HNF-4 in intestinal cells. \u003cem\u003eAm. J. Physiol. Liver Physiol.\u003c/em\u003e\u003cstrong\u003e289\u003c/strong\u003e, G220\u0026ndash;G226 (2005).\u003c/li\u003e\n\u003cli\u003eDunagan, M., Chaudhry, K., Samak, G. \u0026amp; Rao, R. K. Acetaldehyde disrupts tight junctions in Caco-2 cell monolayers by a protein phosphatase 2A-dependent mechanism. \u003cem\u003eAm. J. Physiol. Liver Physiol.\u003c/em\u003e\u003cstrong\u003e303\u003c/strong\u003e, G1356\u0026ndash;G1364 (2012).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTable 1 is available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-antimicrobials-and-resistance","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjamar","sideBox":"Learn more about [npj Antimicrobials and Resistance](http://www.nature.com/npjamar/)","snPcode":"44259","submissionUrl":"https://submission.springernature.com/new-submission/44259/3","title":"npj Antimicrobials and Resistance","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"drug discovery, high-throughput screening, Shigellosis, caco-2 cell","lastPublishedDoi":"10.21203/rs.3.rs-5874972/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5874972/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe Gram-negative bacterial species \u003cem\u003eShigella\u003c/em\u003e is the second leading cause of diarrhea among children in low and middle-income countries (LMICs) and is a World Health Organization (WHO) priority pathogen. \u003cem\u003eShigella\u003c/em\u003e infections are becoming increasing difficult to treat due to antimicrobial resistance (AMR), leading to an urgent for new antimicrobial agents with novel modes of action. \u003cem\u003eShigella\u003c/em\u003e pathogenesis is largely intracellular and antibacterial chemicals that preferentially work inside cells may be desirable to limit collateral AMR and block key components of the \u003cem\u003eShigella\u003c/em\u003e infection cycle. Aiming to facilitate the process of identifying antibacterial chemicals that kill intracellular \u003cem\u003eShigella\u003c/em\u003e, we developed a high-throughput screening (HTS) cell-based chemical screening assay. The three-dimensional (3-D) assay, incorporating \u003cem\u003eShigella\u003c/em\u003e invasion into Caco-2 cells on Cytodex\u003csup\u003e3\u003c/sup\u003e beads, was scaled into a 384 well platform for screening chemical compound libraries. Using this assay, we evaluated\u0026thinsp;\u0026gt;\u0026thinsp;500,000 compounds, identifying 12 chemical hits that inhibit \u003cem\u003eShigella\u003c/em\u003e replication inside cells. This simple, efficient and HTS-compatible assays circumvents many of the limitations of traditional screening methods with cell monolayers and may be deployed for antibacterial compound screening for other intracellular pathogens.\u003c/p\u003e","manuscriptTitle":"A three-dimensional high throughput assay identifies novel antibacterial molecules with activity against intracellular Shigella","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-02-03 09:04:17","doi":"10.21203/rs.3.rs-5874972/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-02-25T20:10:44+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-02-20T18:57:07+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-02-20T17:47:10+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-02-10T06:46:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"293717332613051689470171800708192550343","date":"2025-02-06T17:47:47+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"269635808866373091121045363212482320774","date":"2025-02-04T18:46:18+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"112855693187665340822775584194514682850","date":"2025-01-29T00:01:41+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"313578461653351213736192466001100859619","date":"2025-01-28T11:35:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"75895728971942893109364082309420726673","date":"2025-01-28T07:46:24+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-01-27T08:05:50+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-01-24T08:55:33+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-01-23T12:16:10+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Antimicrobials and Resistance","date":"2025-01-21T16:50:36+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-antimicrobials-and-resistance","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjamar","sideBox":"Learn more about [npj Antimicrobials and Resistance](http://www.nature.com/npjamar/)","snPcode":"44259","submissionUrl":"https://submission.springernature.com/new-submission/44259/3","title":"npj Antimicrobials and Resistance","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7a5b2af9-e5e9-4aec-b020-087582e45368","owner":[],"postedDate":"February 3rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":43490217,"name":"Biological sciences/Drug discovery"},{"id":43490218,"name":"Health sciences/Diseases/Gastrointestinal diseases"},{"id":43490219,"name":"Health sciences/Diseases/Infectious diseases"},{"id":43490220,"name":"Biological sciences/Microbiology"},{"id":43490221,"name":"Biological sciences/Microbiology/Antimicrobials"}],"tags":[],"updatedAt":"2025-04-27T07:38:05+00:00","versionOfRecord":[],"versionCreatedAt":"2025-02-03 09:04:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5874972","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5874972","identity":"rs-5874972","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}