Pseudomonas aeruginosa disrupts Caenorhabditis elegans iron homeostasis, causing a hypoxic response and death

Abstract

The opportunistic pathogen Pseudomonas aeruginosa causes serious human infections, but effective treatments and the mechanisms mediating pathogenesis remain elusive. Caenorhabditis elegans shares innate immune pathways with humans, making it invaluable to investigate infection. To determine how P. aeruginosa disrupts host biology, we studied how P. aeruginosa kills C. elegans in a liquid-based pathogenesis model. We found that P. aeruginosa-mediated killing does not require quorum-sensing pathways or host colonization. A chemical genetic screen revealed that iron chelators alleviate P. aeruginosa-mediated killing. Consistent with a role for iron in P. aeruginosa pathogenesis, the bacterial siderophore pyoverdin was required for virulence and was sufficient to induce a hypoxic response and death in the absence of bacteria. Loss of the C. elegans hypoxia-inducing factor HIF-1, which regulates iron homeostasis, exacerbated P. aeruginosa pathogenesis, further linking hypoxia and killing. As pyoverdin is indispensable for virulence in mice, pyoverdin-mediated hypoxia is likely to be relevant in human pathogenesis.

Figure 1. The C. elegans-P. aeruginosa LK assay shows robust, virulence- and time-dependent killing, see also Figure S1

(A–C) C. elegans were fed a normal, non-pathogenic E. coli OP50 food source (A) or P. aeruginosa PA14 (B–C). In (C) gentamicin, an anti-Pseudomonas antimicrobial was added. In (A–C), upper panels are brightfield images, lower panels show fluorescence from Sytox orange absorbed by dead worms. (D) Killing by PA14 is time-dependent, with most death occurring between 36 and 60 hpi (squares). OP50 shows virtually no killing (circles). (E) Time course with PA14 or OP50 over varying initial bacterial concentrations. Error bars (D, E) show standard deviation.

Figure 2. LK involves different virulence mechanisms than plate-based assays, see also Figure S2 and Table S1

(A) PA14Δphz and mutants exhibiting attenuation in SK were statistically indistinguishable from PA14 in liquid. Exposure time was 44 hours, and killing was normalized to PA14. (B–D) DIC (upper) and fluorescence (lower) images show whether worms are colonized by OP50-dsRed (B), PA14-dsRed (C), or PA14ΔlasR-dsRed (D) in SK. (E–G) No colonization was observed in worms fed OP50-dsRed (E), PA14-dsRed (F), or PA14ΔlasR-dsRed (G) in a parallel liquid assay. Error bars (A) are standard error, asterisks (A) represent p<0.01.

Figure 3. A high-throughput screen identified 18 compounds that rescue LK, see also Table S2

(A) Fluorescence micrographs of each well of a 384-well plate with small molecules screened for curing in this assay. Columns outlined in blue and red are gentamicin (positive) and DMSO (negative) controls, respectively. Circled in orange are three compounds that showed significant rescue, as demonstrated by the absence of stained worms. (B) A scatter plot shows the percent survival in each of the two replicates for the example plate. Circles colored red represent wells containing DMSO, circles shaded blue depict gentamicin wells, and test compounds are shown as orange-colored circles. Three compounds showing significant rescue are at the upper right. (C) A list of the compounds, and the classes to which they belong, is also shown. (D) Dose-dependence of rescue by chelating molecules was tested. Ciclopirox olamine (CPX), 1,10-phenanthroline, or 2,2′-bipyridine were added to LK assay in serial two-fold dilutions and death was measured after 48 hours of infection. Error bars represent the standard error of three biological replicates.

Figure 4. Pyoverdin biosynthesis is important for virulence in LK,see also Table S3

Pyoverdin, but not pyochelin, biosynthesis mutants in PA14 (A) and PAO1 (B) showed attenuated virulence in liquid. Exposure length was 44 hours and killing was normalized to PA14 (A) or PAO1 (B). (C, D) Killing assays were performed for a panel of PA14 pyoverdin biosynthesis mutants and PA14ΔgacA in LK (C) and SK (D). All pvd mutants were significantly attenuated in liquid (p<0.01) but not in SK conditions. Error bars (A, B) represent standard error, and asterisks (A, B) are p<0.01.

Figure 5. Exposure to PA14 or pyoverdin triggers a hypoxic response, see also Figure S3

(A, B) A panel of twenty hypoxia-upregulated genes was assayed by qRT-PCR under various conditions, including exposure to OP50 in liquid, OP50 with 1.2 mM Phe, PA14 or PA14pvdF in liquid, and PA14 infection in SK (A) or filtrates from OP50 or PA14 with or without iron supplementation during growth (B). Values were normalized to pre-infected conditions. Fold-changes were color-coded as follows: blue, < −5; white, between −5 and 2; pale yellow, between 2 and 5; yellow, between 5 and 10; orange, between 10 and 100; red, > 100. HIF-1-dependent gene names are shaded gray. (C, D) cdc-25.1(RNAi) (called N2) and cdc-25.1(RNAi);hif-1(ia4) (called hif-1) survival was compared in LK (C) and SK (D) conditions. cdc-25.1(RNAi);hif-1(ia4) exhibited enhanced susceptibility in liquid, and enhanced resistance on agar (p < 0.001).

Figure 6. Pyoverdin exposure causes host death, see also Figure S4

(A) Relative pyoverdin concentration in filtrates was measured by fluorescence emission at 460nm. (B) PA14 filtrate was incubated with (PA14/Fe) and without (PA14) iron and fluorescence emission was assayed during an LK assay. (C) Survival of OP50-fed worms exposed to various filtrates in LK. PA14 filtrate shows significant killing (PA14, p=0) that is ameliorated by growing the pathogen with 100 μM iron supplementation (PA14/Fe, p=0, compared to PA14 filtrate) or by overnight incubation of PA14 filtrate with 100 μM iron (PA14/Fe*, p < 10⁻⁶). Error bars (A, B) represent standard error.

Figure 7. A schematic of iron-scavenger toxicity in liquid

Exposure to a high concentration of chelator, PA14, or pyoverdin present in filtrate triggers a hypoxic response and subsequent worm death. A low concentration of chelator precludes bacterial growth, but does not induce a hypoxic response, thereby preventing death.