Iron Limitation in Klebsiella pneumoniae Defines New Roles for Lon Protease in Homeostasis and Degradation by Quantitative Proteomics

Abstract

Nutrient adaptation is key in limiting environments for the promotion of microbial growth and survival. In microbial systems, iron is an essential component for many cellular processes, and bioavailability varies greatly among different conditions. In the bacterium, Klebsiella pneumoniae, the impact of iron limitation is known to alter transcriptional expression of iron-acquisition pathways and influence secretion of iron-binding siderophores, however, a comprehensive view of iron limitation at the protein level remains to be defined. Here, we apply a mass-spectrometry-based quantitative proteomics strategy to profile the global impact of iron limitation on the cellular proteome and extracellular environment (secretome) of K. pneumoniae. Our data define the impact of iron on proteins involved in transcriptional regulation and emphasize the modulation of a vast array of proteins associated with iron acquisition, transport, and binding. We also identify proteins in the extracellular environment associated with conventional and non-conventional modes of secretion, as well as vesicle release. In particular, we demonstrate a new role for Lon protease in promoting iron homeostasis outside of the cell. Characterization of a Lon protease mutant in K. pneumoniae validates roles in bacterial growth, cell division, and virulence, and uncovers novel degradation candidates of Lon protease associated with improved iron utilization strategies in the absence of the enzyme. Overall, we provide evidence of unique connections between Lon and iron in a bacterial system and suggest a new role for Lon protease in the extracellular environment during nutrient limitation.

Keywords: Klebsiella pneumoniae; Lon protease; iron limitation; quantitative proteomics; secretome.

FIGURE 1

Mass-spectrometry-based proteomics of K. pneumoniae during iron limitation. (A) Bottom–up proteomics workflow for profiling the cellular proteome and secretome of K. pneumoniae under iron-limited (limited) and iron-replete (10 and 100 μM iron) conditions. Proteins were extracted from cells and supernatant followed by enzymatic digestion and purification on C18 resin tips (Rappsilber et al., 2007; Geddes-McAlister and Gadjeva, 2019). The data are processed and analyzed using the publicly available MaxQuant and Perseus platforms (Cox and Mann, 2008; Tyanova et al., 2016). Figure generated with Biorender.com. (B) Venn diagram comparing total number of proteins identified in cellular proteome (black) and secretome (gray). (C) Venn diagram comparing proteins identified in the cellular proteome under limited (red), 10 μM iron (blue), and 100 μM iron (green) conditions. (D) Venn diagram comparing proteins identified in the secretome under limited (red), 10 μM iron (blue), and 100 μM iron (green) conditions. (E) Stacked bar plot quantifying the number of iron-associated proteins identified across the conditions in both the cellular proteome and secretome among significantly different proteins [Student’s t-test p-value < 0.05; false discovery rate (FDR) = 0.05; S0 = 1].

FIGURE 2

Profiling the impact of iron availability on the cellular proteome of K. pneumoniae. (A) Principal component analysis; clustering based on treatments. (B) Volcano plot depicting all proteins identified in limited and replete (10 μM iron) conditions, highlighting proteins with significant increases or decreases in abundance during 10 μM iron-replete (blue) vs. iron-limited (red) conditions. (C) Volcano plot depicting all proteins identified in limited and replete (100 μM iron) conditions, highlighting proteins with significant increases or decreases in abundance during 100 μM iron replete (green) vs. limited (red) conditions. (D) Volcano plot depicting all proteins identified in replete (10 and 100 μM iron) conditions, highlighting proteins with a significant increase in abundance during 100 μM iron-replete (green) conditions. (E) 1D annotation enrichment based on Uniprot Keywords [p < 0.05; false discovery rate (FDR) = 0.05; score > −0.5 < 0.5]. (F) STRING network analysis of significantly different proteins with an increase in abundance during replete vs. limited conditions. Protein clusters highlighted. (G) STRING network analysis of significantly different proteins with a decrease in abundance during replete vs. limited conditions. Protein clusters highlighted. For volcano plots: Student’s t-test, p-value ≤ 0.05; FDR = 0.05, S0 = 1.

FIGURE 3

Secretome profiling of K. pneumoniae under iron limitation. (A) Principal component analysis; clustering based on treatments (component 1) and biological reproducibility (component 2). (B) Volcano plot depicting all proteins identified in limited and replete (10 μM iron) conditions, highlighting proteins with significant increases or decreases in abundance during 10 μM iron-replete (blue) vs. iron-limited (red) conditions. Lon protease demonstrated the largest decrease in abundance relative to the replete conditions (>8-fold) (highlighted). (C) Volcano plot depicting all proteins identified in limited and replete (100 μM iron) conditions, highlighting proteins with significant increases or decreases in abundance during 100 μM iron-replete (green) vs. iron-limited (red) conditions. Lon protease is highlighted. (D) Volcano plot depicting all proteins identified in replete (10 and 100 μM iron) conditions, highlighting proteins with significant increases or decreases in abundance during 100 μM (green) vs. 10 μM (blue) iron conditions. (E) Pie chart illustrating cellular compartments and signal peptides (undesignated by cellular compartment, but contain signal peptide) of secreted proteins. (F) Venn diagram of K. pneumoniae proteins identified in this secretome study (orange) vs. proteins identified in previous profiling of outer membrane vesicles (yellow). (G) STRING network analysis of significantly different proteins with an increase in abundance during replete vs. limited conditions. Protein clusters highlighted. (H) STRING network analysis of significantly different proteins with a decrease in abundance during replete vs. limited conditions. Protein clusters highlighted. For volcano plots: Student’s t-test, p-value ≤ 0.05; FDR = 0.05, S0 = 1.

FIGURE 4

Phenotypic profiling of Δlon highlights the diverse impact of Lon protease in K. pneumoniae. (A) Growth curve of WT, Δlon:LON, and Δlon strains under limited (LIM), 10 μM iron-replete, and 100 μM iron-replete conditions. OD₆₀₀ measured; error bars represent standard deviation. Experiment performed in biological triplicate, technical duplicate with two independent mutant strains. (B) Iron utilization plates of WT, Δlon:LON, and Δlon strains under limited (LIM), 10 μM iron-replete, and 100 μM iron-replete conditions. Images captured at 24 h; dilution series from 10⁷ to 10². (C) Differential interference contrast microscopy at 100 × for WT, Δlon:LON, and Δlon strains under 100 μM iron-replete conditions. Scale bar = 10 μm. (D) Lactate dehydrogenase (LDH) assay of immortalized BALB/c macrophages infected (MOI 50:1) with K. pneumoniae WT, Δlon:LON, and Δlon strains during the time course of infection. Error bars represent standard deviation. (E) Association, invasion, and replication assay highlighting differences among the WT, Δlon:LON, and Δlon strains during infection. *Student’s t-test p-value ≤ 0.05.

FIGURE 5

Defining candidate degradation products of Lon protease in K. pneumoniae. (A) Principal component analysis; clustering based on strain. (B) Volcano plot depicting all proteins identified during cellular proteome profiling of K. pneumoniae wild-type (WT) and Δlon strains. Proteins showing an increase in abundance in the absence of Lon suggest targets of degradation (purple) compared to proteins unchanged (gray) or with a higher abundance in the WT (black). Known interactors (i.e., RcsA, IbpB, and LuxR are highlighted), as well as proteins with highest fold-change differences (i.e., BamA and CmlA2). Student’s t-test, p-value ≤ 0.05; false discovery rate (FDR) = 0.05, S0 = 1.