Chemical Optimization of Selective Pseudomonas aeruginosa LasB Elastase Inhibitors and Their Impact on LasB-Mediated Activation of IL-1β in Cellular and Animal Infection Models

Abstract

LasB elastase is a broad-spectrum exoprotease and a key virulence factor of Pseudomonas aeruginosa, a major pathogen causing lung damage and inflammation in acute and chronic respiratory infections. Here, we describe the chemical optimization of specific LasB inhibitors with druglike properties and investigate their impact in cellular and animal models of P. aeruginosa infection. Competitive inhibition of LasB was demonstrated through structural and kinetic studies. In vitro LasB inhibition was confirmed with respect to several host target proteins, namely, elastin, IgG, and pro-IL-1β. Furthermore, inhibition of LasB-mediated IL-1β activation was demonstrated in macrophage and mouse lung infection models. In mice, intravenous administration of inhibitors also resulted in reduced bacterial numbers at 24 h. These highly potent, selective, and soluble LasB inhibitors constitute valuable tools to study the proinflammatory impact of LasB in P. aeruginosa infections and, most importantly, show clear potential for the clinical development of a novel therapy for life-threatening respiratory infections caused by this opportunistic pathogen.

Keywords: IL-1β; LasB; Pseudomonas aeruginosa; antivirulence; elastase; pseudolysin.

Figure 1

Crystal structure of P. aeruginosa LasB inhibited by compound 16 (PDB code, 7QH1). (A) Content of the asymmetric unit showing the presence of four subunits, all containing an inhibitor molecule in their active site (all four subunits were nearly identical, r.m.s.d. 0.10–0.20 Å); (B) superimposition of native LasB (PDB code, 1EZM; orange) with subunit A of the LasB:16 complex (green; r.m.s.d., 1.16 Å); (C) close-up view of LasB (green) showing the interaction of compound 16 (cyan) with several key residues and the catalytic zinc cofactor (gray sphere) (see text for details); and (D) surface representation of the LasB active site showing the close contacts with the difluoroindanyl and benzothiazole moieties of the inhibitor and the protruding, poorly interacting, quaternary ammonium substituent.

Figure 2

Inhibition of LasB-mediated proteolysis of host substrates by compounds 12 and 16. (A) Proteolysis of IgG by LasB results in the generation of a 36 kDa degradation product, which is inhibited by increasing concentrations of LasB inhibitors 12 and 16. (B, C). Hydrolysis of different concentrations of the pro-IL-1β fluorescent peptide Sub115 by LasB was inhibited by compounds 12 (B) and 16 (C). Experiments were performed twice on two separate occasions (results from a single set of experiments are shown).

Figure 3

Compounds 12 and 16 inhibit secreted LasB and activation of IL-1β in P. aeruginosa-infected THP-1 macrophages. Compounds 12 and 16 at concentrations from 0.5 to 8 μM were added to THP-1 macrophages immediately prior to infection (MOI 10:1) with P. aeruginosa PAO1 (black), PAO1ΔlasB (white), or PAO1 in the presence of the caspase inhibitor VX-765 (5 μM) (gray). Controls with no compounds (0 μM) were included for each condition. Culture supernatants were collected after 2 h for analysis. Top panel shows the effect of compounds 12 (A) and 16 (B) on LasB activity as measured by hydrolysis of the N-terminal pro-IL-1β fluorescent peptide (Sub115). Bottom panel shows the effect of compounds 12 (C) and 16 (D) on activation of cytokine IL-1β as measured by induction of a signal in an IL-1 luciferase cell reporter assay. Statistical significance was determined by ANOVA with a Dunnett post-test: * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001; ns, not significant.

Figure 4

Plasma and ELF PK in mouse after single 10 mg/kg IV dose of compounds 12 and 16. Epithelial lining fluid (ELF) concentrations were determined by normalizing bronchoalveolar lavage fluid (BALF) concentrations to plasma urea concentrations to adjust for dilution during the sampling procedure.

Figure 5

Compounds 12 and 16 inhibit LasB-mediated IL-1 β activation and reduce P. aeruginosa numbers in a mouse lung infection model. (A) Schematic representation of mouse infection, drug treatment, and endpoint measurements: doses of either 10 or 30 mg/kg compound were administered to the tail vein of immunocompetent C57Bl/6 mice at 1 and 4 h post-intranasal inoculation. (B) Measurement of activated IL-1β in lung homogenates at 24 h post-infection via luminescence (RU) in cell reporter. (C) Measurement of bacterial cell numbers (CFUs) in lung homogenates at 24 h post-infection. The horizontal line represents the mean value for each dataset. Statistical significance was determined by ANOVA with a Dunnett post-test: ** p < 0.01; *** p < 0.001; **** p < 0.0001; ns, not significant.