Arginine as an environmental and metabolic cue for cyclic diguanylate signalling and biofilm formation in Pseudomonas putida

Abstract

Cyclic diguanylate (c-di-GMP) is a broadly conserved intracellular second messenger that influences different bacterial processes, including virulence, stress tolerance or social behaviours and biofilm development. Although in most cases the environmental cue that initiates the signal transduction cascade leading to changes in cellular c-di-GMP levels remains unknown, certain L- and D-amino acids have been described to modulate c-di-GMP turnover in some bacteria. In this work, we have analysed the influence of L-amino acids on c-di-GMP levels in the plant-beneficial bacterium Pseudomonas putida KT2440, identifying L-arginine as the main one causing a significant increase in c-di-GMP. Both exogenous (environmental) and endogenous (biosynthetic) L-arginine influence biofilm formation by P. putida through changes in c-di-GMP content and altered expression of structural elements of the biofilm extracellular matrix. The contribution of periplasmic binding proteins forming part of amino acid transport systems to the response to environmental L-arginine was also studied. Contrary to what has been described in other bacteria, in P. putida these proteins seem not to be directly responsible for signal transduction. Rather, their contribution to global L-arginine pools appears to determine changes in c-di-GMP turnover. We propose that arginine plays a connecting role between cellular metabolism and c-di-GMP signalling in P. putida.

Figure 1

Modulation of c-di-GMP cell content by l-arginine in P. putida KT2440, the arginine biosynthesis mutants ΔargG and ΔargH, and the ΔcfcR mutant. Strains harbouring pCdrA::gfp^C were grown in diluted LB (1:3) supplied with different final concentrations of l-arginine (0, 5, 15 and 25 mM). Data correspond to the fluorescence values corrected by culture growth (OD₆₆₀) over time. Averages and standard deviations of two biological replicates with three experimental replicates each are plotted. A Synergy Neo2 Biotek fluorimeter was used in these experiments.

Figure 2

Modulation of c-di-GMP cell content by l-amino acids in P. putida KT2440. Cultures harbouring pCdrA::gfp^C were grown in 96-well plates during 24 h in 1:3 diluted LB (a) or M9 minimal medium with glucose (b) in the presence of each l-amino acid at 5 mM (light bars) or 15 mM (dark bars). Fluorescence and turbidity were quantified every 30 min for 24 h using a Tecan Infinite 200 fluorimeter. Values corresponding to the area under the curve derived from fluorescence measurements normalized by culture growth (OD₆₀₀) were calculated, to obtain a global overview of fluorescence along the whole growth curve. Data are given as percentage relative to the value obtained for KT2440 (pCdrA::gfp^C) without any added amino acid (established as 100%, dotted line). Averages and standard deviations of three independent experiments with three replicates each are presented. Values at least 10% higher or lower than the control and showing statistically significant differences with it are indicated by asterisks (Student´s t test; p ≤ 0.05).

Figure 3

Effect of l-arginine on planktonic growth (a) and biofilm formation (b) by P. putida KT2440. Cultures were grown in 96-well plates in FAB medium with glucose and different concentrations of l-arginine (0, 5 and 15 mM). Growth was measured at 660 nm, and attached biomass was quantified as absorbance at 595 nm after staining with crystal violet (CV) and subsequent solubilisation of the dye. Results are averages and standard errors of two independent experiments with four technical replicates each. Asterisks indicate statistically significant differences with respect to the control without l-arginine (Student’s t test; p ≤ 0.05).

Figure 4

Influence of ΔargG and ΔargH deletions on expression of adhesin- and EPS-encoding genes. KT2440 (circles), ΔargG (triangles) and ΔargH (squares) strains carrying reporter fusions corresponding to lapA::lacZ (a), lapF::lacZ (b), pea (PP_3132::lacZ) (c), peb (PP_1795::lacZ) (d), bcs (PP_2629::lacZ) (e), and alg (algD::lacZ) (f) were grown in LB. Turbidity (OD₆₀₀, hollow symbols) and β-galactosidase activity (Miller units, solid symbols) at the indicated time points are shown. d-cycloserine (75 μg/ml) was added in f after 1 h of growth, since the algD promoter is inactive in the absence of cell wall stress in P. putida. The data are averages and standard deviations of at least two biological replicates with two technical repetitions each.

Figure 5

l-arginine increases expression of pea. KT2440 harbouring the PP_3132::lacZ fusion was grown in M9 minimal medium with glucose as carbon source, supplied with 0, 5, and 15 mM l-arginine (shown as increasing intensity colour bars), and β-galactosidase activity was measured at different time points. The experiment was done in duplicate with three technical repetitions each. Statistically significant differences were observed at 8, 9, 10 and 11 h between the absence and presence of l-arginine (Student´s t test; p ≤ 0.05), but quantitatively relevant differences were obvious only at 10 and 11 h (early stationary phase).

Figure 6

Influence of l-arginine biosynthesis on expression of rpoS and cfcR. (a) KT2440 (circles), and the ΔargG (triangles) and ΔargH (squares) strains harbouring pMAMV21 (rpoS’–‘lacZ) were grown in LB and β-galactosidase activity was measured at the indicated times. Data correspond to averages and standard errors of two biological replicas with three technical repetitions each. Statistically significant differences between the wild type and mutants were detected from 10 h onwards (Student´s t test: p < 0.05). (b) Influence of increasing concentrations of l-arginine on expression of rpoS’–‘lacZ in KT2440 and the ΔargG and ΔargH strains harbouring pMAMV21. Cultures were grown for 24 h in LB or LB with increasing concentrations of l-arginine (shown as increasing intensity colour bars) and β-galactosidase activity was analysed. Graph corresponds to averages and standard deviations of two biological replicas with three technical repetitions each. Statistically significant differences with respect to each control without amino acid supplementation are indicated by asterisks (Student´s t test: p ≤ 0.01). (c) KT2440 (circles), and the ΔargG (triangles) and ΔargH (squares) strains harbouring pMIR200 (cfcR::lacZ) were grown in LB and β-galactosidase activity was measured at the indicated times. Data correspond to averages and standard errors of two biological replicas with three technical repetitions each.

Figure 7

Growth of P. putida KT2440 and mutant derivatives with l-arginine as nitrogen (a) or carbon and energy source (b). Strains KT2440 (blue lines), ΔargT (crimson lines), ΔartJ (green lines), ΔoccT (orange lines), and ΔargTΔartJ (purple lines) were grown in a Bioscreen C MBR apparatus at 30 °C with shaking during 24 h in 100-well plates in M8-glucose or M9 minimal medium with 10 mM l-arginine. Absorbance in the 420–580 nm range was measured every 30 min. Three independent assays were done with three technical replicas each. Averages and standard deviations of one representative experiment are shown.

Figure 8

Role of substrate binding proteins in modulation of c-di-GMP cell content by environmental l-arginine. P. putida KT2440 (blue lines), ΔargT (crimson lines), ΔartJ (green lines), ΔoccT (orange lines), and ΔargTΔartJ (purple lines) strains harbouring pCdrA::gfp^C were grown in M9 minimal medium with glucose (a) and 5 mM (b) or 15 mM (c) l-arginine. Data correspond to fluorescence values corrected by culture growth (OD₆₆₀). Measurements were done every 30 min on a Varioskan Lux fluorimeter. Averages and standard deviations are plotted.

Figure 9

Influence of periplasmic substrate binding proteins on biofilm formation by P. putida. KT2440 and the ΔargT, ΔartJ, ΔargTΔartJ, and ΔoccT mutants were grown in FAB minimal medium with glucose and different l-arginine concentrations (shown as increasing intensity colour bars). Attached biomass was analysed after 10 h of growth. Values correspond to absorbance (A₅₉₅) after staining with crystal violet and subsequent solubilisation of the dye, normalized with respect to culture growth (OD₆₆₀). Data are averages and standard errors from two independent experiments with four technical replicates each. Asterisks indicate statistically significant differences between the wild type and the corresponding mutant in each condition (Student’s t test; *p ≤ 0.05; **p ≤ 0.01).