A zebrafish embryo infection model can be used for medium-throughput compound screening and can predict the oral bioavailability of test compounds. (A) Schematic representation of the in vivo drug-screening setup in the zebrafish-M. marinum infection model. (B) Representative images of different readout groups of M. marinum-infected zebrafish embryos. (C) M. marinum-tdTomato yolk-infected zebrafish embryos treated with antibiotics at 10 µM, or as specified. Each data point represents the integrated red fluorescence intensity of a single zebrafish embryo, and the signal of each group (10-12 embryos) is expressed as mean±s.e.m. Statistical significance was determined by one-way ANOVA, following Dunnett's multiple comparison test by comparing the signal from the DMSO-treated control sample with each treatment group (*P≤0.05, **P≤0.01, ***P≤0.001, ****P≤0.0001). (D) Zebrafish embryos were yolk infected with M. marinum-tdTomato and at 24 hpi treated by adding the antibiotics into the fish water. Survival was scored 4 dpi. Each group consisted of ten embryos. A non-treated group of embryos (0×MIC) served as control. (E,F) Zebrafish were infected via the caudal vein route with E. coli GK1161434 (E) or S. pneumonia D39V (F) and treated by the addition of the antibiotics to the fish water at 1 hpi. Survival was scored 24 hpt. Each group consisted of 20-40 embryos. Concentrations of all antibiotics were based on the MIC value of the antibiotic for each strain (see Table S1). A non-treated group of embryos (0×MIC) served as control.

Screening a library of anti-mycobacterial compounds in zebrafish-infection model identifies 14 hit-compounds. (A) Schematic representation of the screen design. Compounds active against Mtb and M. marinum in vitro were tested in the zebrafish embryo-M. marinum yolk-infection model. (B) Hit compounds were tested in a dose-response assay. Each data point represents the integrated red fluorescence intensity of a single zebrafish embryo, and the signal of each group (10-20 embryos) is expressed as mean±s.e.m. Statistical significance was determined by one-way ANOVA, following Dunnett's multiple comparison test by comparing the signal from the DMSO-treated control sample with each treatment group (*P≤0.05, **P≤0.01, ***P≤0.001, ****P≤0.0001).

Activity of TBA161 variants in macrophage and zebrafish infection models. (A) THP-1 macrophages were infected with Mtb carrying pTetDuo, expressing GFP with the tetracycline-inducible promoter and tdTomato under the constitutive promoter p_smyc. Infected macrophages were treated with various doses of each test compound for 6 days. GFP expression was induced by the addition of ATc, and macrophage nuclei were stained with Hoechst dye to detect macrophages (grey bars). The GFP signal within each macrophage was quantified, representing the amount of viable bacteria (green bars). DMSO- and rifampicin (RIF, 3 µM)-treated samples served as a negative and positive control, respectively. Data points represent the average of duplicates with s.d. (B) Representative images of Mtb-pTetDuo infected THP-1 macrophages treated with DMSO or compound TBA161-C at 6 dpt. Blue, macrophage nuclei (Hoechst); yellow, merged signal of Mtb expressing tdTomato (red) and gfp (green). Scale bars: 50 µM. (C) Dose-dependent activity of TBA161 variants in the zebrafish-M.marinum infection model. Each data point represents the integrated red fluorescence intensity of a single zebrafish embryo, and the signal of each group (10-20 embryos) is expressed as mean±s.e.m. Statistical significance was determined by one-way ANOVA, following Dunnett's multiple comparison test by comparing the signal from the DMSO-treated control sample with each treatment group (**P≤0.01, ***P≤0.001, ****P≤0.0001). (D) Representative images of M. marinum-tdTomato yolk-infected zebrafish embryos treated with DMSO (left) or TBA161-C (right) at 3 dpt. (E) Survival curves of M. marinum yolk-infected zebrafish embryos after dose-dependent drug treatment. The treatment started at 1 dpi by adding compounds to the fish water. Each treatment group consisted of 25-30 embryos. Scale bars: 1 mm.

Mutations in aspS are associated with TBA161-C resistance. (A) Susceptibility of M. marinum wild type (Mmar) and TBA161-C-resistant isolates (Mmar-R-TBA161-C) towards TBA161-C after 4 days of incubation. (B) Susceptibility of Mtb wild type and TBA 161-resistant isolates (Mtb-R-TBA161-C) towards TBA161-C was measured after 7 days. (C) M. marinum wild type transformed with pMS2-aspS_Mtb (Mmar+aspS_Mtb), and pMS2-aspS_MtbF526L (Mmar+aspS_MtbF526L) were incubated with compound TBA161-C for 4 days at the indicated concentrations. (D) Mtb carrying plasmids pMS2-aspS_Mmar (Mtb+aspS_Mmar) and pMS2-aspS_MmarR168G (Mtb+aspS_MmarR168G) were incubated with twofold dilutions of compound TBA161-C for 7 days. (E) TBA161-C (orange) docked into the catalytic subdomain of chain A of Mtb AspS (PDB ID: 5W25). The zoom-in shows TBA161-C in stick representation, together with AspS residues aligning the binding pocket. These include R171 (blue), the three residues of which side chains were treated flexibly during docking (grey), and T570 of chain B (dark brown). The distant F526 residue is shown in blue. For clarity, the L200 label and all hydrogen atoms are omitted. Data are mean of duplicates±s.d.

TBA161-C has potent activity in the zebrafish infection model compared to other AspS inhibitors. (A) The chemical structures of the test compounds, TBA161-C, C1 and GSK93A. (B) Zebrafish embryos were yolk infected with M. marinum-tdTomato and treated with compounds at the indicated concentrations. Each data point represents the integrated red fluorescence intensity of a single zebrafish embryo, and the signal of each group is expressed as mean±s.e.m. Statistical significance was determined by one-way ANOVA, following Dunnett's multiple comparison test by comparing the signal from the DMSO-treated control sample with each treatment group (***P≤0.001, ****P≤0.0001).