Bypasses in intracellular glucose metabolism in iron-limited Pseudomonas putida

Abstract

Decreased biomass growth in iron (Fe)-limited Pseudomonas is generally attributed to downregulated expression of Fe-requiring proteins accompanied by an increase in siderophore biosynthesis. Here, we applied a stable isotope-assisted metabolomics approach to explore the underlying carbon metabolism in glucose-grown Pseudomonas putida KT2440. Compared to Fe-replete cells, Fe-limited cells exhibited a sixfold reduction in growth rate but the glucose uptake rate was only halved, implying an imbalance between glucose uptake and biomass growth. This imbalance could not be explained by carbon loss via siderophore production, which accounted for only 10% of the carbon-equivalent glucose uptake. In lieu of the classic glycolytic pathway, the Entner-Doudoroff (ED) pathway in Pseudomonas is the principal route for glucose catabolism following glucose oxidation to gluconate. Remarkably, gluconate secretion represented 44% of the glucose uptake in Fe-limited cells but only 2% in Fe-replete cells. Metabolic (13) C flux analysis and intracellular metabolite levels under Fe limitation indicated a decrease in carbon fluxes through the ED pathway and through Fe-containing metabolic enzymes. The secreted siderophore was found to promote dissolution of Fe-bearing minerals to a greater extent than the high extracellular gluconate. In sum, bypasses in the Fe-limited glucose metabolism were achieved to promote Fe availability via siderophore secretion and to reroute excess carbon influx via enhanced gluconate secretion.

Keywords: Pseudomonas; gluconate; glucose metabolism; iron limitation; pyoverdine; siderophore.

Figure 1

Schematic of metabolic pathways in Pseudomonas putida, growth phenotype, sugar consumption, and carbon secretion. (A) The metabolic routes through the Entner–Doudoroff pathway, the pentose phosphate pathway (PPP), and the tricarboxylic acid cycle are shown in black, the metabolites or pathways that are involved in siderophore biosynthesis are shown in blue, the anabolic pathways toward biomass production are shown in red, and location of Fe‐containing enzymes are shown in yellow. Gluconate, Glucn; 6‐phospho‐gluconate, 6P‐Glucn; glucose‐6‐phosphate, G6P; fructose‐6‐phosphate, F6P; fructose‐1,6‐bisphosphate, FBP; dihydroxyacetone‐phosphate, DHAP; glyceraldehyde‐3‐phosphate, GAP; 3‐PG; phosphoenolpyruvate, PEP; acetyl coenzyme A, acetyl‐CoA; adenosine triphosphate, ATP; reduced ubiquinone, UQH ₂; reduced nicotinamide adenosine, NADH; carbon dioxide, CO ₂; amino acids, AAs. (B) Growth rate of exponentially growing P. putida cells under Fe‐replete [Fe(+)] and Fe‐limiting [Fe(−)] conditions. (C) Glucose uptake rate. (D) Major sources of excreted carbons. See the text and Table S2 for more details. Values were calculated out of total carbon‐equivalent glucose uptake (from Fig. 2B): 28.26 ± 1.78 mmol C g_CDW ⁻¹ h⁻¹ and 14.77 ± 1.95 mmol C g_CDW ⁻¹ h⁻¹, respectively, in (+)Fe and (−)Fe growth media. The measured data (average ± standard deviation) were from four biological replicates. Two‐tailed unpaired t‐test analysis comparing the measured values for the two Fe conditions: P < 0.05.

Figure 2

Elucidating the metabolic network structure through the assimilation of [1,2‐¹³C₂]‐glucose into intracellular metabolites. (A) Carbon mapping (left) and metabolite labeling (right) in the periplasm, the Embden–Meyerhof–Parnas (EMP), and the Entner–Doudoroff (ED) pathways. The red‐colored arrows and carbon skeletons are to illustrate the formation of nonlabeled metabolites that are directly a result of the ED pathway into the reverse route of the EMP pathway. (B) Carbon mapping (left) and metabolite labeling (right) in the pentose‐phosphate pathway (PPP). (C) Carbon mapping (left) and metabolite labeling (right) in the tricarboxylic acid cycle (C). The broken‐lined arrows indicate the minor formation routes of the metabolites. Labeling patterns: nonlabeled (blue), singly labeled (red), doubly labeled (yellow), and triply labeled (green). Legend for metabolite names are the same as reported in Figure 1. Isotopologue data (average ± standard deviation) were obtained from four biological replicates. [Correction added on 30 September 2015 after first online publication: Figure 2 had inconsistent labelling and these have now been corrected in this version].

Figure 3

Quantitative metabolic flux analysis. (A) Metabolic flux modeling constrained by ¹³C metabolite labeling, metabolite secretion, siderophore secretion, and biomass composition. The top and bottom numbers represent, respectively, the metabolic reaction rates normalized to glucose uptake in Fe‐replete [(+)Fe] and the Fe‐limited [(−)Fe] Pseudomonas putida cells. Metabolites outlined with broken‐lined squares contribute to biomass and siderophore biosynthesis. The different colors reflect the log₂ fold‐change of the reaction rates in the (−)Fe cells compared to those in (+)Fe cells. Statistics on the estimated reaction rates are presented in Table S3 and S4. (B) Percentage investment of the total glucose consumption rate toward gluconate secretion (yellow), siderophore secretion as pyoverdine (orange), net carbon dioxide efflux generated from metabolic reactions (blue), and biomass production (biosynthesis of amino acids, ribonucleotides, and cell membrane components) (dark red). Others denote the sum of other metabolite excretions. The values were estimated from the experimentally constrained metabolic flux analysis (illustrated in part A) on P. putida cells exponentially growing under (+)Fe and (−)Fe conditions. [Correction added on 2 December 2015, after first online publication: Some of the values in Figure 3 were incorrect and have now been amended in this current version and in Table S4 in the Supporting Information.]

Figure 4

Kinetic ¹³C metabolic flux profiling and intracellular metabolite pools. (A) Kinetic incorporation of fully labeled glucose into intracellular metabolites of Fe‐limited [(−)Fe, light brown] and Fe‐replete [(+)Fe, black] cells. For clarity, only the averaged values (from four biological replicates) are shown; standard deviation values are less than 5%. (B) Changes in the intracellular pool of selected metabolites in (−)Fe cells relative to (+)Fe cells during exponential growth. Data (average ± standard deviation) were obtained from three biological replicates. Metabolite names in (A) and (B) are the same as in Figure 1.

Figure 5

Dissolution of iron (Fe)‐bearing minerals and schematic of rerouted glucose metabolism under Fe limitation. (A) Total dissolved Fe (μmol L⁻¹) following reactions of (1 g L⁻¹) hematite (black), goethite (white), magnetite (gray) with bacterial secretions obtained from glucose‐grown Fe‐replete [(+)Fe] and Fe‐limited [(−)Fe] Pseudomonas putida (A), and prepared solutions containing either 100 mmol/L glucose (Gluc) (B), 10 mmol/L gluconate (Glucn) (C), or 100 μmol/L pyoverdine (PVD) (D). The error bars represent one standard error (n = 2–6). (E) Schematic illustration of rerouted glucose metabolism in P. putida KT2440 in response to Fe limitation. The catabolic route in Fe‐replete and Fe‐limited cells is shown by the blue and red lines, respectively.