Hydroxybiphenylamide GroEL/ES Inhibitors Are Potent Antibacterials against Planktonic and Biofilm Forms of Staphylococcus aureus

Abstract

We recently reported the identification of a GroEL/ES inhibitor (1, N-(4-(benzo[ d]thiazol-2-ylthio)-3-chlorophenyl)-3,5-dibromo-2-hydroxybenzamide) that exhibited in vitro antibacterial effects against Staphylococcus aureus comparable to vancomycin, an antibiotic of last resort. To follow up, we have synthesized 43 compound 1 analogs to determine the most effective functional groups of the scaffold for inhibiting GroEL/ES and killing bacteria. Our results identified that the benzothiazole and hydroxyl groups are important for inhibiting GroEL/ES-mediated folding functions, with the hydroxyl essential for antibacterial effects. Several analogs exhibited >50-fold selectivity indices between antibacterial efficacy and cytotoxicity to human liver and kidney cells in cell culture. We found that MRSA was not able to easily generate acute resistance to lead inhibitors in a gain-of-resistance assay and that lead inhibitors were able to permeate through established S. aureus biofilms and maintain their bactericidal effects.

Figure 1.

A. Chemical structure of the initial GroEL/ES hit inhibitor, 1, which was previously reported to potently inhibit the proliferation of E. faecium (EC₅₀ = 0.15 μM) and S. aureus (EC₅₀ = 0.20 μM). Analogs of inhibitor 1 have been synthesized and evaluated in this study, where the R¹ through R⁵ substituents and substructures have been systematically removed to probe for the contributions that each make to inhibiting chaperonin system biochemical functions and bacterial and human cell viabilities. B. Chemical structures of related compounds used as anthelmintics in veterinary medicine.

Figure 2.

Correlation plots of IC₅₀ values for compounds evaluated in the respective biochemical assays. A. Compounds inhibit nearly equipotently in the GroEL/ES-dMDH and the GroEL/ES-dRho refolding assays, supporting on-target effects (Spearman correlation coefficient comparing log(IC₅₀) values in each assay is 0.9663, p < 0.0001). B. While some compounds inhibit in the native MDH enzymatic reporter counter screen, none inhibit native Rho enzymatic activity, further supporting on-target effects for inhibiting the chaperonin-mediated refolding cycle. Results plotted in the grey zones represent IC₅₀ values higher than the maximum concentrations listed.

Figure 3.

Correlation plots comparing IC₅₀ values for compounds tested in the GroEL/ES-dMDH refolding assay with EC₅₀ values for inhibiting E. faecium (A) and MRSA (B) proliferation. While a general trend is observed between inhibiting the GroEL/ES chaperonin system and E. faecium proliferation (Spearman correlation coefficient comparing log(I/EC₅₀) values in each assay is 0.9628, p < 0.0001), supporting on-target effects in bacteria, inhibitors are more potent against MRSA (Spearman correlation coefficient comparing log(I/EC₅₀) values in each assay is 0.8042, p < 0.0001), suggesting potential off-target effects and/or greater GroEL/ES sensitivity in S. aureus bacteria. Results plotted in the grey zones represent IC₅₀ and EC₅₀ values higher than the maximum concentrations listed.

Figure 4.

Correlation plots comparing human HSP60/10-dMDH and GroEL/ES-dMDH refolding assay IC₅₀, human cell viability CC₅₀, and MRSA proliferation EC₅₀ results. A. Compounds inhibit the HSP60/10 and GroEL/ES chaperonin systems nearly equipotently, suggesting binding sites may be highly conserved between the two (Spearman correlation coefficient comparing log(IC₅₀) values in each assay is 0.8351, p < 0.0001). B. Despite compounds inhibiting human HSP60/10 in vitro, many exhibit low to no cytotoxic effects against human liver (THLE-3) and kidney (HEK 293) in cell viability assays (Spearman correlation coefficient values are 0.4791 (p < 0.0008) and 0.3286 (p < 0.0258) when comparing HSP60/10-dMDH refolding assay log(IC₅₀) values with liver and kidney cell viability log(CC₅₀) values, respectively). C. Lead analogs inhibit MRSA proliferation with high selectivity compared to cytotoxicity to human liver (THLE-3) and kidney (HEK 293) cells. Results plotted in the grey zones represent IC₅₀, CC₅₀, and EC₅₀ values higher than the maximum concentrations listed.

Figure 5.

Exploring adaptive tolerance by MRSA bacteria to analogs 1, 11, vancomycin, and a previously-reported GroEL/ES inhibitor, “28R” (structure shown and numbering as previously reported). Average EC₅₀ values of compounds tested after each 24 h passage are plotted from triplicate analyses. MRSA rapidly evolved resistance to 28R, but retained sensitivity to 1, 11, and vancomycin throughout 12 day experiment. Data plotted in the gray zones represent EC₅₀ results beyond the assay detection limits (i.e., >100 μM).

Figure 6.

Representative dose-response plots for compound 1 (upper panel) and vancomycin (lower panel) evaluated in the S. aureus planktonic growth, biofilm formation, and biofilm penetration/bactericidal activity assays. Compound 1 is effective in all three assays, while vancomycin is ineffective at killing S. aureus bacteria in established biofilms. EC₅₀ results for 1, vancomycin, and additional lead inhibitors tested in these assay are presented in Table 4.

Scheme 1^a