Transposon Mutagenesis Screen of Klebsiella pneumoniae Identifies Multiple Genes Important for Resisting Antimicrobial Activities of Neutrophils in Mice

Abstract

Klebsiella pneumoniae is a Gram-negative bacterial pathogen that causes a range of infections, including pneumonias, urinary tract infections, and septicemia, in otherwise healthy and immunocompromised patients. K. pneumoniae has become an increasing concern due to the rise and spread of antibiotic-resistant and hypervirulent strains. However, its virulence determinants remain understudied. To identify novel K. pneumoniae virulence factors needed to cause pneumonia, a high-throughput screen was performed with an arrayed library of over 13,000 K. pneumoniae transposon insertion mutants in the lungs of wild-type (WT) and neutropenic mice using transposon sequencing (Tn-seq). Insertions in 166 genes resulted in K. pneumoniae mutants that were significantly less fit in the lungs of WT mice than in those of neutropenic mice. Of these, mutants with insertions in 51 genes still had significant defects in neutropenic mice, while mutants with insertions in 52 genes recovered significantly. In vitro screens using a minilibrary of K. pneumoniae transposon mutants identified putative functions for a subset of these genes, including in capsule content and resistance to reactive oxygen and nitrogen species. Lung infections in mice confirmed roles in K. pneumoniae virulence for the ΔdedA, ΔdsbC, ΔgntR, Δwzm-wzt, ΔyaaA, and ΔycgE mutants, all of which were defective in either capsule content or growth in reactive oxygen or nitrogen species. The fitness of the ΔdedA, ΔdsbC, ΔgntR, ΔyaaA, and ΔycgE mutants was higher in neutropenic mouse lungs, indicating that these genes encode proteins that protect K. pneumoniae against neutrophil-related effector functions.

Keywords: Klebsiella; Klebsiella pneumoniae; Tn-seq; dedA; gntR; lung infection; neutropenia; wzm; yaaA; ycgE.

FIG 1

High-throughput screen to identify Klebsiella pneumoniae genes needed for lung infection of healthy and neutropenic hosts. (A) Reproducibility between inputs was assessed by calculating Spearman’s correlation efficient, which averaged 0.953 and which ranged from 0.9374 to 0.9657, with P being <0.001. (B and C) Number of individual insertion mutants (B) and genes (C) represented in the inputs and outputs of WT and anti-Ly6G-treated (α-Ly6G) mice. (D and E) WT and α-Ly6G mice were intranasally infected with 2 × 10⁴ CFU of a library containing ∼13,000 Tn insertion mutants and sacrificed at 33 hpi, and their lungs were collected for analysis. The number of CFU per gram of lung tissue (D) and the polymorphonuclear cell (PMN) population (the percentage of Gr1^hi CD11b⁺ cells within the live cell population) (E) were quantified by flow cytometry. Each dot represents a mouse, and the lines indicate the geometric mean (D) or the mean (E). The dotted lines indicate the cutoffs for mice included in further analyses. (F and G) Histograms of the log₂-transformed nCI values of insertion mutants by gene in WT (F) and α-Ly6G (G) mice. (H and I) Genes were considered to have statistically significant contributions in WT (H) or α-Ly6G (I) mice if the Tn mutants of that gene had a WT-MAX99% of <0.27 (H) or a PMN-MAX99% of <0.43 (I), respectively, as indicated by dotted lines. Data either are representative of those from 4 to 6 independent experiments (A) or were pooled from 4 to 6 independent experiments (B to I). Statistical significance was determined by one-way analysis of variance on log-transformed values with Tukey’s posttest (B and C) or Student’s t test (D and E) on either log-transformed (D) or nontransformed (E) values. ns, not significant; ****, P < 0.0001.

FIG 2

Multiple gene hits in Tn-seq are utilized by K. pneumoniae to counteract neutrophil effector functions. (A) Mock-treated (circles) and DFO-injected (triangles) mice were intranasally infected with 2 × 10⁴ CFU of WT K. pneumoniae and sacrificed at 33 hpi. Their lungs were collected for analysis, and the number of CFU per gram of lung tissue is presented. Each symbol represents a mouse, and lines indicate the geometric mean. (B) The K. pneumoniae Tn::appC mutant and neutral insertion mutants (MKP330, MKP332, and MKP354) were grown in L until they were in mid-log phase. The Tn::appC mutant or the neutral insertion mutants were pooled 1:1 with WT K. pneumoniae grown under the same conditions and used to inoculate L with 0.125 mM DIP. The cultures were grown for 5 h and then plated on L and L-KAN plates. The competitive index (CI) was calculated using (number of CFU of mutant-treated cultures/number of CFU WT-treated cultures)/(number of CFU of mutant-untreated cultures/number of CFU WT-untreated cultures) and normalized (norm) to the neutral WT growth. (C) WT (circles) and gp91phox^−/− (squares) C57BL/6 mice were retropharyngeally infected with 1.5 × 10³ CFU of WT K. pneumoniae and sacrificed at 45 hpi. The bacteria in the lungs were quantified as described in the legend to panel A. (D and E) The indicated K. pneumoniae strains were grown, pooled as described in the legend to panel B, incubated in M9-glucose with 1 mM H₂O₂ (D) and M9-glucose with 1 mM DETA-NONOate (E) for 1 h, and then plated. CI was calculated as described in the legend to panel B. (A to E) Data were pooled from 3 independent experiments with at least 2 mice per cohort (A, C) or in technical duplicate (B, D, and E). Statistical significance was determined by Student’s t test on log-transformed values (A, C) or one-way analysis of variance on nontransformed values with Dunnett’s posttest (B, D, and E). ns, not significant (P > 0.05); *, P < 0.05; ****, P < 0.0001.

FIG 3

K. pneumoniae (Kp) requires dedA, dsbC, gntR, wzm, yaaA, and ycgE in the lungs for virulence during single-strain infections. (A to D) WT (circles) and α-Ly6G (triangles) mice were intranasally infected with 1 × 10³ CFU of the indicated Tn mutant (A) or 2 × 10³ CFU of the indicated strains (B to D) and sacrificed at 45 hpi (A and B) or 42 hpi (C and D). The lungs were collected for analysis, and the number of CFU per gram of lung tissue (A to C) and the PMN population (the percentage of Gr1^hi CD11b⁺ cells within the live cell population), as quantified by flow cytometry (D), are presented. Each dot represents one mouse, and open dots indicate the limit of detection where no CFU was recovered. Bars indicate the geometric mean (A to C) or the mean (D). Each strain was tested in cohorts of at least 2 mice in at least 2 independent experiments, and the data were pooled. Statistical significance was determined by one-way analysis of variance on log-transformed values (A to C) with Dunnett’s (A and B) or Fisher’s (C) least-significant-difference posttests. ns, not significant (P > 0.05); *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

FIG 4

K. pneumoniae requires dsbC, wzm, and ycgE for mucoviscosity and wzm for capsule production. (A to E) WT K. pneumoniae and the indicated mutants were grown overnight in L. (A) L (circles) or L with 0.25 mM DIP (triangles) was inoculated with overnight cultures, and the OD₆₃₀ was measured for 16 h. (B) The total bacterial density (OD₆₀₀) of each strain was measured, the strains were centrifuged at 1,000 × g for 5 min, and the OD₆₀₀ of the supernatant was measured. Mucoviscosity was calculated as (OD₆₀₀ supernatant)/(OD₆₀₀ total), and the value was normalized to the value for the WT. (C and D) The glucuronic acid content following capsule precipitation was determined by a colorimetric reaction (OD₅₂₀) and interpolated from a standard curve for glucuronic acid. (E) RNA was isolated, DNase treated, and reverse transcribed into cDNA, and wzt mRNA was quantified and normalized relative to the level of 16S mRNA expression for each strain. Data are representative of those from 3 independent experiments (A) or were pooled from at least 2 (B) or 3 (C to E) independent experiments and are presented as the mean ± SEM (B to E). Significance was determined by one-way analysis of variance on the OD₆₃₀ at 16 h with Dunnett’s posttest (A), with nontransformed values with Dunnett’s posttest (B to D), or with nontransformed values with Welch’s t test (E). ns, not significant (P > 0.05); *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

FIG 5

K. pneumoniae requires gntR and yaaA for ROS resistance in vitro. (A to D) WT K. pneumoniae and the indicated mutants were grown overnight in L. (A and B) The strains were then recultured until log phase, incubated in M9-glucose with or without 1 mM ROS for 1 h, and plated to determine survival, and the results were normalized to those for the M9-glucose-only condition for each strain. (C and D) RNA was isolated from log-phase cultures of the indicated strains, DNase treated, and reverse transcribed into cDNA, and gntK (C) or edd (D) mRNA was quantified and normalized relative to the level of 16S mRNA expression for each strain. (A to D) Data were pooled from at least 3 independent experiments and are presented as the mean + SEM. Statistical significance was determined by one-way analysis of variance on log-transformed values with Dunnett’s posttest (A and B) or Welch’s t test on nontransformed values (C and D). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

FIG 6

K. pneumoniae requires dedA and ycgE for RNS resistance in vitro. (A to E) WT K. pneumoniae and the indicated mutants were grown overnight in L. (A and B) Strains were cultured until log phase and then incubated in M9-glucose with or without 1 mM DETA-NONOate for 1 h and plated for survival, and the survival rate was normalized to the rate for the M9-glucose-only condition for each strain. (C to E) RNA was isolated from log-phase cultures of the indicated strains, DNase treated, and reverse transcribed into cDNA, and the relative copy number of each target mRNA was quantified and normalized relative to the level of 16S mRNA expression for each strain. Data are representative of those from at least 3 independent experiments (A and B) or were pooled from at least 3 independent experiments (C to E) and are presented as the mean ± SD (A and B) or the mean + SEM (C to E). Statistical significance was determined by one-way analysis of variance with Dunnett’s posttest (A and B), Welch’s t test (C and D), or Student's t test (E). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.