Table 1.
New stresses in the chemical genomic screen.
Fig 1.
The chemical genomic screen expands our view of gene function and intrinsic resistance in E. coli K-12.
(A) A scatter plot of individual fitness-scores for conditions present in both screens (n = 17). Measurements between screens are reproducible, with a Pearson’s correlation of 0.61. (B) Conditional-phenotypes were assigned using a stress-specific cutoff for fitness-scores that allowed a false discovery rate (FDR) of 5%. A responsive gene is defined as a gene with at least one phenotype in the dataset. (C) Significant correlations between genes were determined using a cutoff for Pearson’s correlation that allowed an FDR of 5%.
Fig 2.
A diverse set of pathways contribute to antibiotic resistance.
Fitness-scores from the integrated dataset connected antibiotic resistance to multiple biological pathways. (A) Genes with different resistance mechanisms (target pathway, stress response, drug efflux, detoxification, and drug import) that protect against stresses from the current screen are organized according to mechansim. (B) A heatmap of fitness-scores for translation related genes. Sensitivities of deletions of various translation associated factors distinguish between drug families targeting translation. Sensitivities to aminoglycosides (gentamicin), macrolides (clarithromycin), tetracyclines (tetracycline), chloramphenicol, and tRNA synthetase inhibitors (pseudomonic acid A and serine hydroxamate) are shown. Fitness-scores from the integrated dataset that were determined in Nichols et al. [8] are marked with an asterisk (*). (C) A heatmap of fitness-scores for genes related to drug efflux and detoxification. Multiple drugs from the new screen were connected with the major efflux pump of E. coli, AcrAB-TolC.
Fig 3.
Peptide ABC-importers determine susceptibility to kasugamycin and blasticidin S in E. coli K-12.
(A) Structures of kasugamycin (Ksg) and blasticidin S (BcS). (B) Deletions of peptide importer genes are resistant to kasugamycin and blasticidin S. The heat-map of fitness-scores for dipeptide permease (ΔdppA, ΔdppB, ΔdppC, ΔdppD, and ΔdppF), oligopeptide permease (ΔoppA, ΔoppB, ΔoppC, ΔoppD, and ΔoppF), and their negative regulators (Δhfq, ΔgcvA, and ΔgcvB) for the entire set of new stresses is shown. Ksg and BcS are highlighted within the heatmap. (C) Deletions of each peptide permease operon show an increase in resistance to Ksg and BcS. 10-fold spot dilutions are shown for operon deletions Δopp, Δdpp and the double mutant Δopp Δdpp (D) Overexpression of opp results in a decrease in resistance to Ksg and BcS. 10-fold spot dilutions of cells with the high copy vector pDSW204 containing the opp operon (pOpp) grown without induction indicate decreased resistance to both Ksg and BcS relative to the empty vector control (vector).
Fig 4.
Peptide ABC-importers determine the rate of translation inhibition by kasugamycin and blasticidin S.
Altered kinetics of translation inhibition, as measured by 35S-methionine incorporation, serve as a proxy for changes in antibiotic uptake rates. (A) Deletion of opp and dpp slows the rate of translation inhibition following addition of 25mM kasugamycin (B) Deletion of opp slows the rate of translation inhibition by 1mM blasticidin S. Error bars represent standard deviation of technical replicates (n = 3) Each experiment was repeated with at least one biological replicate with similar results. (C) Overexpression of opp from a high copy vector increases the rate of translation inhibition by kasugamycin. Error bars represent measurements from two biological replicates (left). Significance was tested using a paired two-tailed student’s t-test (n = 2) at one minute after addition of kasugamycin (right). The kinetics of translation inhibition by kasugamycin was best fit with a double exponential decay function, whereas inhibition by blasticidin S was best fit using a single exponential decay function.
Fig 5.
The peptide ABC-importers directly import kasugamycin.
(A) The effect of peptide competitors for Dpp (Ala-Ala) and Opp (Pro-Phe-Lys) on the kinetics of inhibition of 35S-methionine incorporation. Addition of peptide competitors slowed the rate of translation inhibition by kasugamycin for wild-type E. coli but had a negligible effect for Δopp Δdpp. Error bars represent standard deviation of technical replicates (n = 3). This experiment was repeated with one biological replicate with similar results. (B) PFK binding to purified OppA induced an increase in intrinsic tryptophan fluorescence, as measured at 343 nm. Addition of 1 mM kasugamycin increased the effective concentration of PFK required to reach half-maximal fluorescence shift. Independent experiments with unique protein purifications had similar results. Equilibrium binding of PFK with and without kasugamycin was fit with a quadratic model that incorporates ligand-depletion.
Fig 6.
A common import pathway for kasugamycin and blasticidin S is likely to be conserved across multiple species.
Genetic context is shown for importer genes likely to be involved in kasugamycin and blasticidin S uptake for multiple species. Sequence homologues of opp are shown in white. Sequence homologues of dpp are shown in black. The importer npp is shown in red. The proposed eukaryotic uptake gene from Magnaporthe oryzae is shown in blue. Genes that encode solute-binding proteins are named in each species. Periplasm (P), Inner Membrane (IM), and Cytoplasm (C) are labeled. (A) The opp and dpp operons in E. coli contribute to uptake of both kasugamycin and blasticidin S. The reference genome used was MG1655 (U00096.3). (B) Erwinia amylovora has high sequence homology to E. coli for both opp and dpp, and a duplication of the oppA gene. Uptake pathways in this species are unknown but likely to be similar to E. coli. The reference genome used was CFBP1430 (FN434113.1) (C) Pseudomonas aeruginosa has no clear sequence homologues of opp and multiple copies of dppA. Both the npp and dpp importers appear to uptake blasticidin S, but the responsible solute binding proteins have not been identified. The reference genome used was UCBPP-PA14 (CP000438.1) (D) Magnaporthe oryzae has no sequence homologues of oppA or dppA, but cross-resistance between kasugamycin and blasticidin S indicate that an analogous importer may be operating in this eukaryote. The gene responsible for cross-resistance was tentatively named kas-3 but this gene has not been mapped [48]. The reference genome used was 70–15 (AACU00000000.3).
Table 2.
Genetic interactions of the peptide ABC-importer subunits.