Targeting Iron - Respiratory Reciprocity Promotes Bacterial Death

Abstract

Discovering new bacterial signaling pathways offers unique antibiotic strategies. Here, through an unbiased resistance screen of 3,884 gene knockout strains, we uncovered a previously unknown non-lytic bactericidal mechanism that sequentially couples three transporters and downstream transcription to lethally suppress respiration of the highly virulent P. aeruginosa strain PA14 - one of three species on the WHO's 'Priority 1: Critical' list. By targeting outer membrane YaiW, cationic lacritin peptide 'N-104' translocates into the periplasm where it ligates outer loops 4 and 2 of the inner membrane transporters FeoB and PotH, respectively, to suppress both ferrous iron and polyamine uptake. This broadly shuts down transcription of many biofilm-associated genes, including ferrous iron-dependent TauD and ExbB1. The mechanism is innate to the surface of the eye and is enhanced by synergistic coupling with thrombin peptide GKY20. This is the first example of an inhibitor of multiple bacterial transporters.

Figure 1.. Lysis-free ‘N-104’ killing of virulent and multi-drug resistant P. aeruginosa strain ‘PA14’.

(A) Diagram of lacritin with bioactive synthetic peptide ‘N-104’, recombinant ‘N-65’ and inactive ‘N-80/C-25’ and ‘C-95’. (B) N-104 interaction with Gram-negative model membrane as monitored by surface plasmon resonance (SPR). Inserts: diagram of SPR with N-104 on model membrane and SPR response at 600 s. Two-way ANOVA with Tukey’s multiple comparison test (n = 4 experiments). (C) Maximum SPR as a function of N-104 concentration. (D) N-104 amino acids necessary for killing determined by colony forming unit (CFU) and C. elegans pathogenesis assays. Scale bars are 2 mm and 200 μm, respectively. (E, F) Correlation assay of N-104 or N-104 analog bactericidal activity versus membrane disruption. (SPR: n = 4 experiments, CFU: n = 3 experiments). (G) ¹H-NMR of N-104. (H) Assay of N-104 hemolytic activity versus positive control 5% Triton X-100. Nonparametric Friedman test with Dunn’s multiple comparisons test (n = 3 experiments). Data represent the mean ± SD, **p<0.01, *p<0.05, ns, non-significant. See also Figure S1.

Figure 2.. N-104 killing involves both an iron and a polyamine transporter.

(A) Keio E. coli K12 gene knockout library. (B) E. coli K12 viability after treatment with N-104, N-80/C-25 (negative control) or ampicillin (positive control). Two-way ANOVA with Dunnett’s multiple comparisons test (n = 3 experiments). (C) Resistance screen relative slope ratio before (slope 1) and after (slope 2) N-104 treatment (model example). (D) 120 mutant strains display a slope ratio ≥ 0.75 indicative of N-104 resistance, identified from duplicate screens of the entire Keio library. (E) Resistant mutant strains from (D) narrowed by duplicate screens. (F) Combined slope ratios versus alamarBlue fluorescence at five hours of (E; blue) resistant compared to wild type (red) strains. One-way ANOVA (n = 4 experiments). (G) Viability assays of the 10 resistant mutants, inclusive of E. coli K12 lacking ferrous iron transporter FeoB and the PotH subunit of the PotFGHI polyamine transporter. Two-way ANOVA (n = 3 experiments). (H) P. aeruginosa PA14 Transposon Insertion Mutant Library. (I) Analogous feoB and potH PA14 mutants are also N-104 resistant. Two-way ANOVA with Šidák’s multiple comparisons test (n = 3 experiments). Data represent the mean ± SD, ****p<0.0001, ns, non-significant. See also Figure S2 and Table S1.

Figure 3.. N-104 killing requires N-104 transport across the outer membrane into the periplasm.

(A) Imaging flow cytometry of FITC-N-104 and FITC-C-95 (negative control) treated PA14 cells versus positive control ethanol permeabilized cells. (B) Dose dependence of (A). Two-way ANOVA with Šidák’s multiple comparisons test (n = 3 experiments). (C) Selective affinity of N-104 with tandem mass spectometry identified outer membrane lipoprotein ‘YaiW’, previously known to transport Bac7(1–35) peptide . (D) Hypothetical YaiW-mediated translocation of N-104 through the outer membrane to reach inner membrane FeoB and PotH. (E) PA14 YaiW transposon mutant is N-104 resistant. Two-way ANOVA with Šidák’s multiple comparisons test (n = 3 experiments). (F) Wide-field fluorescence imaging of Janelia Fluor 549-N-104 and -C-95 (negative control) treated PA14 or PA14 YaiW transposon mutant cells. SYTOX green controls for Janelia Fluor 549 permeability. (G) Single cell axial pixel intensity tracings of cells in (F). (H) Schematic of Bacterial ChloroAlkane Penetration Assay in E. coli in which chloroalkane tagged N-104 and C-95 (negative control) compete with chloroalkane modified fluorescent dye for cells expressing either HaloTag periplasmic or cytoplasmic marker. (I) Results of with HaloTag periplasmic marker. Two-way ANOVA with Šidák’s multiple comparisons test (n = 3 experiments). (J) Results with HaloTag cytoplasmic marker. Two-way ANOVA with uncorrected Fisher’s least significant difference test (n = 3 experiments). Data represent the mean ± SD, ****p<0.0001, ***p<0.001. See also Figure S3 and Table S3.

Figure 4.. N-104 targets periplasmic loops 4 and 2 respectively of FeoB and PotH.

(A) FeoB expression construct and TMHMM (v 2) predicted membrane structure. (B) Western blot of IPTG-induced FeoB and passage over N-104 vs C-95 (negative control) columns, the former in the absence or presence of FeoB synthetic peptides corresponding to periplasmic loops. Unpaired t-test with Welch’s correction (n = 3 experiments). (C) PotH expression construct and TMHMM (v 2)/Deep TMHMM (1.0.24) predicted membrane structure. (D) Western blot of IPTG-induced PotH and passage over N-104 vs C-95 (negative control) columns, the former in the absence or presence of PotH synthetic peptides corresponding to periplasmic loops. Unpaired t-test with Welch’s correction (n = 3 experiments). (E) Western blots of flowthrough, final wash fraction or KCl eluants from N-104 or C-95 columns of (i) FeoB or FeoB deletion/substitution mutants, or (ii) FeoB with periplasmic (outer) loop synthetic peptides or SDC1 peptide. (F) Analysis of (E). Friedman ANOVA (n = 3 experiments). (G) Western blots of flowthrough, final wash fraction or KCl eluants from N-104 or C-95 columns of (i) PotH or PotH deletion/substitution mutants, or (ii) PotH with periplasmic (outer) loop synthetic peptides. (H) Analysis of (G). Friedman ANOVA (C-95; 2,3; 1,3; 1,2; 1,3/2,3), Kruskal-Wallis ANOVA (OL2; OL3) (n = 3 experiments). Data represent the mean ± SD, **p<0.01, *p<0.05, ns, non-significant. See also Figure S4 and Table S4.

Figure 5.. By targeting FeoB and PotH, N-104 suppresses ferrous iron and spermidine uptake.

(A) Pretreatment with N-104 (or N-80/C-25 or C-95 as negative controls) prior to Fe²⁺ supplementation. (B) PA14 intracellular iron in the absence and presence of supplemented Fe²⁺. Unpaired t-test with Welch’s correction (n = 3 experiments). (C) Comparative iron uptake by untreated, N-104, or negative control N-80/C-25 or C-95 treated PA14 cells. Two-way ANOVA with Dunnett’s multiple comparisons test (n = 3 experiments). (D) Pretreatment with N-104 (or N-80/C-25 or C-95 as negative controls) prior to ¹⁴C-spermidine (¹⁴C-Spd) supplementation. (E) PA14 uptake of supplemented ¹⁴C-spermidine (¹⁴C-Spd) and ¹⁴C-putrescine (¹⁴C-Put). Unpaired t-test (n = 3 experiments). (F) Comparative ¹⁴C-spermidine uptake by untreated, N-104, or negative control N-80/C-25 or C-95 treated PA14 cells. Two-way ANOVA with Dunnett’s multiple comparisons test (n = 3 experiments). (G) N-104, but not N-80/C-25 nor C-95, killing of PA14. (H) N-104 killing of PA14 in the presence of excess putrescine, but not spermidine. Two-way ANOVA with Dunnett’s multiple comparisons test (n = 3 experiments). (I) N-104 killing of PA14 in the presence of excess putrescine, but not spermidine. Data represent the mean ± SD. ****p<0.0001, ***p<0.001, **p<0.01, ns, non-significant. See also Figure S5.

Figure 6.. N-104 killing synergizes with thrombin C-terminal ‘GKY20’ peptide.

(A) N-104 killing of PA14 at two different osmolalities. (B) N-104 affinity screen of antimicrobial-rich human tears by: (i) tandem mass spectrometry (MS/MS) and (ii) western blotting (WB). (C) Thrombin identified as a N-104 ligand, among others. (n = 3 experiments). (D) Thrombin western blots of human tear flowthrough, final wash fraction or KCl eluants off N-104 or negative control C-95 columns. (E) N-104 plus thrombin peptide GKY20 checkerboard killing assay. Both share an amphipathic α-helix and coiled-coiled secondary structure. (F) N-104 (but not C-95) ligation of immobilized GKY20 as respectively detected by ab-C-term and ab-N-term lacritin antibodies (n = 5 experiments). (G) Affinity of GKY20 for N-104 is similar to that for (Figure 1) N-104 analogs. Two-way ANOVA with Dunnett’s multiple comparisons test (n = 3 experiments). (H) GKY20 ligation of N-104 L108S/L109S/F112S/L114S/L115S/W118S (‘LFW:S’) is less. Friedman ANOVA with Dunn’s multiple comparisons test (n = 3 experiments). Data represent the mean ± SD, ****p<0.0001, **p<0.01. See also Figure S6 and Table S5.

Figure 7.. N-104 plus GKY20 kill multidrug resistant clinical isolates as well as PA14 in infected mice.

(A) N-104 plus GKY20 killing of clinical isolates. (B) Test of N-104 plus GKY20 and GKY20 alone for hemolysis activity (see Figure 1H for N-104 hemolysis assay ). Friedman ANOVA with Dunn’s multiple comparisons test (n = 3 experiments). (C) PA14 infection of mouse cornea after scratching without or with N-104 plus GKY20. (D) Corneal surface pathology scores 24 hours after infection without or with N-104 plus GKY20 treatment (0, eye macroscopically identical to the uninfected contralateral control eye; 1, faint opacity partially covering the pupil; 2, dense opacity covering the pupil; 3, dense opacity covering the entire anterior segment; and 4, perforation of the cornea and/or phthisis bulbi). Mann-Whitney test and effect size estimate for non-parametric tests from Psychometrica ‘Computation of Effect Sizes’ (n = 3 experiments). (E) CFU infectivity remaining in the cornea without or with N-104 plus GKY20 treatment. Mann-Whitney test and effect size estimate for non-parametric tests from Psychometrica ‘Computation of Effect Sizes’ (n = 3 experiments). (F) Simulation of reduced CFU infectivity as a consequence of increased PA14 doubling time. (G) Simulated doubling time and CFU infectivity when the density of PA14 is 3.9, 5 or 6.2% (arrows in (F)) of all corneal cells. Unpaired t test (one simulation over 24 time points). All bar graphs represent the mean ± SD. (D) and (E) are represented as the median ± 95% confidence interval. ***p<0.001, **p<0.01, *p<0.05, ns, nonsignificant. See also Figure S7.