Bis-(3′-5′)-Cyclic Dimeric GMP Regulates Antimicrobial Peptide Resistance in Pseudomonas aeruginosa

Abstract

Bis-(3′-5′)-cyclic dimeric GMP (c-di-GMP) is an intracellular second messenger that controls the lifestyles of many bacteria. A high intracellular level of c-di-GMP induces a biofilm lifestyle, whereas a low intracellular level of c-di-GMP stimulates dispersal of biofilms and promotes a planktonic lifestyle. Here, we used the expression of different reporters to show that planktonic cells, biofilm cells, and cells dispersed from biofilms (DCells) had distinct intracellular c-di-GMP levels. Proteomics analysis showed that the low intracellular c-di-GMP level of DCells induced the expression of proteins required for the virulence and development of antimicrobial peptide resistance in Pseudomonas aeruginosa. In accordance with this, P. aeruginosa cells with low c-di-GMP levels were found to be more resistant to colistin than P. aeruginosa cells with high c-di-GMP levels. This finding contradicts the current dogma stating that dispersed cells are inevitably more susceptible to antibiotics than their sessile counterparts.

INTRODUCTION

It is now widely accepted that microbes are able to form surfaced-attached biofilm communities in the environment and during infection as an alternative to the planktonic or free-living style. Biofilm formation proceeds through several distinct steps, including initial attachment, with subsequent development of dense microcolonies embedded in self-generated extracellular matrix materials (1) and finally dispersal to seed new areas of biofilm formation (2).

Bis-(3′-5′)-cyclic dimeric GMP (c-di-GMP) is a global, intracellular second messenger that controls the lifestyles of many bacteria (3). The intracellular c-di-GMP concentration is controlled by diguanylate cyclases (DGCs) which catalyze the formation of c-di-GMP and phosphodiesterases (PDEs) which degrade c-di-GMP (4). Many bacteria contain multiple copies of DGCs and PDEs, which allow bacterial cells to sense and respond to diverse sets of environmental signals by adjusting the intracellular c-di-GMP content accordingly.

As a secondary messenger that binds to specific domains of regulatory proteins, high level of c-di-GMP stimulates bacteria to form biofilm by enhancing the synthesis of adhesive structures and biofilm matrix components and by reducing motility and chemotaxis (5, 6). In the aggregated biofilm mode, quorum sensing contributes to the production of matrix components that facilitate protection of the biofilm cells against cellular immunity attack and antimicrobial treatments (7–10). Recently, however, a low intracellular level of c-di-GMP has been shown to be necessary for the pathogenesis of bacteria (11, 12). The CheY-EAL-HTH domain protein VieA of Vibrio cholerae is required for the activation of certain virulence factors (13). Another EAL domain-containing protein, CdgR, has been shown to be required by Salmonella to resist phagocytosis and virulence during infection of mice (14).

Pseudomonas aeruginosa is a Gram-negative opportunistic pathogen that can cause a wide range of infections, including those in cystic fibrosis, wounds, and the urinary tract (15). The success of P. aeruginosa as a human pathogen is largely dependent on its ability to form biofilms, produce virulence factors, and launch immune protective measures in an organized fashion, as well as its notorious resistance to antimicrobial agents (16, 17), all of which may allow infections to develop into chronic conditions (16, 17). Here, we studied the effects of modulating the intracellular content of c-di-GMP in P. aeruginosa in relation to biofilm dispersal and antimicrobial peptide resistance.

MATERIALS AND METHODS

Bacteria and growth conditions.

The bacterial strains, plasmids, and primers used in the present study are listed in Table 1. Escherichia coli DH5a strain was used for standard DNA manipulations. Luria-Bertani medium (18) was used to cultivate E. coli strains. Batch cultivation of P. aeruginosa was carried out at 37°C in ABT minimal medium (19) supplemented with 5 g of glucose liter⁻¹ (ABTG) or 2 g of glucose liter⁻¹ plus 2 g of Casamino Acids liter⁻¹ (ABTGC). For plasmid maintenance in E. coli, the medium was supplemented with 100 μg of ampicillin ml⁻¹, 15 μg of gentamicin (Gm) ml⁻¹, 15 μg of tetracycline (Tc) ml⁻¹, or 8 μg of chloramphenicol ml⁻¹. For marker selection in P. aeruginosa, 30 μg of Gm ml⁻¹, 50 μg of Tc ml⁻¹, and 200 μg of carbenicillin ml⁻¹ were used, as appropriate.

Table 1.

Strains,plasmids, and primers used in this study

Strain, plasmid, or primer	Relevant characteristic(s) or sequence (5′–3′)a	Source or reference
Strains
P. aeruginosa
PAO1	Prototypic nonmucoid wild-type strain	20
PAO1ΔwspF	wspF derivative of PAO1 constructed by allelic exchange	21
PAO1/plac-yhjH	Tcr; PAO1 containing the plac-yhjH vector	This study
PAO1/pBAD-yhjH	Gmr; PAO1 containing the pBAD-yhjH vector	This study
PAO1ΔwspF/plac-yhjH	Tcr; PAO1ΔwspF containing the plac-yhjH vector	This study
PAO1ΔwspF/pBAD-yhjH	Gmr; PAO1ΔwspF containing the pBAD-yhjH vector	This study
PAO1ΔpelAΔpslBCD/pcdrA-gfp	Gmr Cbr; low intracellular c-di-GMP content derivative of PAO1	21
PAO1/pcdrA-gfp	Gmr; PAO1 containing the pcdrA-gfp vector	This study
PAO1ΔwspF/pcdrA-gfp	Gmr Cbr; PAO1ΔwspF containing the pcdrA-gfp vector	This study
PAO1/plac-yhjH/pcdrA-gfp	Tcr Cbr; PAO1/plac-yhjH containing the pcdrA-gfp vector	This study
PAO1/pBAD-yhjH/pcdrA-gfp	Gmr Cbr; PAO1/pBAD-yhjH containing the pcdrA-gfp vector	This study
PAO1-ppmr-gfp	Gmr; PAO1 tagged by miniTn7-ppmr-gfp	This study
PAO1ΔwspF/ppmr-gfp	Gmr; PAO1ΔwspF tagged by miniTn7-ppmr-gfp	This study
PAO1/plac-yhjH/ppmr-gfp	Tcr Gmr; PAO1/plac-yhjH tagged by miniTn7-ppmr-gfp	This study
PAO1/ppelA-lacZ	Tcr; PAO1 tagged by miniCTX-ppelA-lacZ	This study
PAO1ΔwspF/ppelA-lacZ	Tcr; PAO1ΔwspF tagged by miniCTX-ppelA-lacZ	This study
PAO1/pBAD-yhjH/ppelA-lacZ	Tcr Gmr; PAO1/pBAD-yhjH tagged by miniCTX-ppelA-lacZ	This study
E. coli
DH5α	F− ϕ80dlacZΔM15 Δ(lacZYA-argF)U169 deoR recA1 endA1 hsdR17(rK− mK+) phoA supE44 λ− thi-1 gyrA96 relA1	Laboratory collection
Plasmids
pUCP22	Apr Gmr; broad-host-range cloning vector	22
pJN105	Gmr; broad-host-range vector carrying the araBAD promoter	23
pBBR1MCS3	Tcr; broad-host-range ori from Bordetella bronchiseptica S87	24
miniTn7-ppmr-gfp	Apr Gmr; miniTn7 vector carrying the ppmr-gfp fusion	25
miniCTX-ppelA-lacZ	Tcr; miniCTX vector carrying the ppelA-lacZ fusion	26
plac-yhjH	Tcr; pBBR1MCS3 carrying the yhjH gene	27
pBAD-yhjH	Gmr; pJN105 carrying the yhjH gene	This study
pcdrA-gfp	Apr Gmr; pUCP22 carrying the pcdrA-gfp fusion	21
pRK600	Cmr; ori ColE1 RK2-Mob+ RK2-Tra+; helper vector for conjugation	28
Primers
yhjH-fwd	AAACTGCAGTAGTGGAGGAATTTGATGATAAGGCAGGTTATCCAGC	This study
yhjH-rev	AAATCTAGAGAAAATGAGGCAGCTTATAGCGC	This study

^aCm^r, chloramphenicol resistance; Tc^r,tetracycline resistance; Ap^r, ampicillin resistance; Gm^r,gentamicin resistance; Cb^r, carbenicillin resistance.

Construction of p_BAD-yhjH vector.

Plasmid pJN105 contains an araC-P_BAD promoter, which has been well studied and induced in the presence of l-arabinose (23). The yhjH gene of E. coli MG1655 was amplified by PCR using primers yhjH-rev and yhjH-fwd. The PCR product was cloned into the vector pJN105 by restriction with PstI and XbaI. DNA restriction enzyme digestions and modifications were performed according to the manufacturer's instructions (Fermentas and Invitrogen). The resulting plasmid p_BAD-yhjH was transferred into E. coli DH5α by electroporation. Correct insertion of the yhjH gene into the vector pJN105 was verified by sequencing. The p_BAD-yhjH plasmid was transformed into E. coli S17-1 by electroporation and thereafter conjugated into P. aeruginosa.

CdrA-gfp assay.

P. aeruginosa strains containing p_cdrA-gfp reporter were cultivated in ABTGC medium at 37°C with shaking. Portions (200 μl) of overnight cultures were transferred into each of the wells of a 96-well microplate. The expression of p_cdrA-gfp in P. aeruginosa was measured using a Tecan Infinite Pro2000 microplate reader. The optical density at 600 nm (OD₆₀₀) and green fluorescent protein (GFP) fluorescence (in relative fluorescence units) were recorded for each well of the 96-well microplate.

For measuring p_cdrA-gfp expression in biofilm cells of the PAO1 strain, the P. aeruginosa PAO1/p_cdrA-gfp strain were cultivated in 50-ml BD Falcon tubes containing 15 ml of ABTGC medium. A sterile glass cover slide (24 by 60 mm) was inserted into each Falcon tube to support biofilm growth. After overnight incubation, PAO1 biofilms on the slides were washed twice with 1 ml of 0.9% NaCl and imaged using fluorescence microscopy (Carl Zeiss). The planktonically growing PAO1/p_cdrA-gfp strain and strain PAO1ΔwspF/p_cdrA-gfp were also imaged using fluorescence microscopy for comparison.

Pel-lacZ assay.

The mini-CTX-p_pel-lacZ reporter fusion (26) was inserted into the chromosomes of P. aeruginosa PAO1, PAO1ΔwspF, and PAO1/p_BAD-yhjH strains by triparental mating with the help of pRK600 vectors as previously described (29). PAO1/p_BAD-yhjH biofilms were cultivated in ABTGC medium in a 24-well plate (Nunc) overnight at 37°C. The biofilms were washed twice with 1 ml of 0.9% NaCl and supplemented with ABTGC medium containing 0.25, 0.5, or 1% arabinose for 5 h to induce dispersal. Biofilms formed by the PAO1 and PAO1ΔwspF strains were dispersed by 5 μM NO donor sodium nitroprusside (SNP; Sigma). As controls, PAO1 and PAO1/p_BAD-yhjH planktonic cultures were diluted 10 times to fresh ABTGC medium and incubated for 5 h at 37°C with shaking. The OD₆₀₀ values of planktonic cells were measured and normalized to the same OD₆₀₀ values of dispersed cells. A classical β-galactosidase assay was used to measure expression of the p_pel-lacZ fusion in P. aeruginosa cells (30).

Intracellular c-di-GMP concentration in biofilm cells.

To assay c-di-GMP concentrations of biofilm cells, the glass slide biofilm assay was performed as previously reported (31). The p_cdrA-gfp containing P. aeruginosa PAO1 and PAO1/p_BAD-yhjH strain were cultivated in 50-ml BD Falcon tubes containing 15 ml of ABTGC medium. A sterile glass cover slide (24 by 60 mm) was inserted into each Falcon tube to support biofilm growth. After overnight incubation, slide biofilms were washed twice with 0.9% NaCl and imaged by using fluorescence microscopy (Carl Zeiss).

Microplate biofilm formation assay.

A microplate biofilm formation assay was carried out in ABTGC medium as previously described (32).

iTRAQ-based proteomics analyses.

P. aeruginosa cells were harvested after 48 h of cultivation in AB minimal medium supplemented with 5 g of glucose liter⁻¹ and subjected to iTRAQ-based proteomics analyses (additional details for these analyses are provided in the supplemental material).

Pyoverdine quantification.

PAO1, PAO1ΔwspF, and PAO1/p_lac-yhjH strains were grown in ABTGC medium overnight. The pyoverdine fluorescence (excitation wavelength, 400 nm; emission wavelength, 450 nm) of each supernatant of P. aeruginosa overnight cultures was recorded by using the Tecan Infinite Pro2000 microplate reader as previously reported (33).

Pmr-gfp assay.

The miniTn7-Gm-p_pmr-gfp fusion was inserted into the chromosomes of PAO1, PAO1ΔwspF, and PAO1/p_lac-yhjH strains by four-parental mating with the help of pBF13 and pRK600 vectors as previously described (29). P. aeruginosa PAO1, PAO1ΔwspF, and PAO1/p_lac-yhjH strains were grown in ABTGC medium overnight. The cultures were then diluted 10-fold into fresh ABTGC medium with or without 1 μg of colistin ml⁻¹. Cultures, 3 μl for each condition, were spotted onto cover slides after 7 h growth for fluorescence microscopy imaging (Carl Zeiss). The level of fluorescence of 30 individual p_pmr-gfp-tagged bacterial cells was measured for each sample by using ImageJ (http://rsbweb.nih.gov/ij/). The corrected total cell fluorescence of each cell was calculated as the sum of the fluorescence intensity within the region of interest minus the background intensity.

Antimicrobial peptide resistance assay of planktonic cells.

For comparison of the resistance of strains PAO1, PAO1ΔwspF, and PAO1/p_lac-yhjH to colistin, the growth curves of the three strains in the presence of 0, 0.125, and 2 μg of colistin ml⁻¹ were produced in triplicate, as previously described (34). Colistin (0.25 μg ml⁻¹) was selected to represent a concentration lower than the MIC of PAO1 (which is 1 μg ml⁻¹), and 2 μg of colistin ml⁻¹ was chosen for a concentration higher than the MIC. Overnight cultures were diluted to an OD₆₀₀ of 0.15 with ABTGC minimal medium containing the appropriate concentrations of colistin. The OD₆₀₀ was recorded every hour for 9 h using the Tecan Infinite Pro2000 microplate reader.

A time-kill kinetic assay was also performed to compare the resistance of PAO1, PAO1ΔwspF, and PAO1/p_lac-yhjH to colistin to concentrations of 2, 4, and 8 μg ml⁻¹, respectively. Overnight cultures of PAO1, PAO1ΔwspF, and PAO1/p_lac-yhjH strains were diluted to an OD₆₀₀ of ∼0.2 in fresh ABTGC medium containing 2, 4, and 8 μg of colistin ml⁻¹, respectively. The absorbance of the surviving bacterial cells was monitored by using the Tecan Infinite Pro2000 microplate reader and Live/Dead BacLight bacterial viability kits (Invitrogen).

Colistin resistance assay of dispersed cells.

In order to compare the tolerance of cells that dispersed from biofilms with tube-cultivated planktonic cells to colistin, biofilms of PAO1 and PAO1/p_BAD-yhjH were cultivated in ABTGC medium in a 24-well plate (Nunc) overnight at 37°C. The biofilms were washed twice with 1 ml of 0.9% NaCl and supplemented with ABTGC medium containing 5 μM SNP (PAO1 biofilms) or 0, 0.5, or 1% arabinose (PAO1/p_BAD-yhjH biofilms) for 5 h to induce dispersal. Biofilms were stained with 0.01% crystal violet as previously described (32). As controls, biofilms of PAO1 and PAO1/p_BAD-yhjH were washed twice with 1 ml of 0.9% NaCl and supplemented with ABTGC medium for 5 h. Planktonic cells were derived from both biofilms. The OD₆₀₀ of dispersed cells and planktonic cells was measured and adjusted to an OD₆₀₀ of 0.15. The growth curves of planktonic PAO1/p_BAD-yhjH cells and dispersed biofilm cells from PAO1/p_BAD-yhjH and PAO1 biofilms were measured in ABTGC medium containing 0, 0.125, and 2 μg of colistin ml⁻¹. The OD₆₀₀ was recorded every 15 min for 5 h using the Tecan Infinite Pro2000 microplate reader.

Biofilm colistin treatment assay.

The P. aeruginosa PAO1 strain was cultivated in 50-ml BD Falcon tubes containing 15 ml of ABTGC medium. A sterile glass cover slide (24 by 60 mm) was inserted into each Falcon tube to support biofilm growth. After overnight incubation, slide PAO1 biofilms were washed twice with 1 ml of 0.9% NaCl and supplemented with ABTGC medium containing 4 μg of colistin ml⁻¹. PAO1 biofilms incubated in ABTGC medium without colistin were used as a control. After 2 h of treatment, the biofilms were washed twice with 0.9% NaCl, stained by using a Live/Dead bacterial viability kit (Invitrogen), and imaged using fluorescence microscopy (Carl Zeiss).

RESULTS

Construction of P. aeruginosa cells with different intracellular c-di-GMP levels.

We constructed P. aeruginosa strains with controllable intracellular levels of c-di-GMP so that we can mimic the three phases of the biofilm life cycle: planktonic cells (PCells), biofilm cells (BCells), and dispersed cells (DCells). The PAO1ΔwspF strain, which overexpresses the diguanylatecyclase WspR (21), is known to contain a high intracellular level of c-di-GMP (11) and can be used to mimic BCells. The PAO1/p_lac-yhjH strain contains a PBBRMCS-2 plasmid carrying the phosphodiesterase gene yhjH fused to a lac promoter, which is constitutively expressed in Pseudomonas (27), and can be used to mimic DCells. The PAO1/p_BAD-yhjH strain contains a pJN105 plasmid carrying yhjH fused to an arabinose-inducible promoter. The intracellular content of c-di-GMP of the PAO1/p_BAD-yhjH strain can be adjusted by arabinose and can therefore be used to mimic all of the three phases dependent on the conditions.

The cdrA and pel genes are both positively regulated by c-di-GMP in P. aeruginosa (35), and the fusions p_cdrA-gfp and p_pel-lacZ can therefore be used to monitor intracellular c-di-GMP levels in P. aeruginosa (21, 26). We measured the expression of the c-di-GMP biosensor p_cdrA-gfp (21) in the PAO1 (PCells), PAO1ΔwspF strain (BCells), PAO1/p_lac-yhjH strain (DCells) and found that the intracellular level of c-di-GMP in the PAO1ΔwspF strain (BCells) was significantly higher than in the PAO1 strain (PCells) and the PAO1/p_lac-yhjH strain (DCells) (Fig. 1A). The wspF mutation was shown before to increase the intracellular content of c-di-GMP of P. aeruginosa up to 7-fold in planktonic growth (21). Expression of the p_lac-yhjH in PAO1ΔwspF strain was found to decrease the p_cdrA-gfp fluorescence intensity (Fig. 1). PAO1 cells from biofilms had a high level of cdrA-gfp fluorescence intensity close to that of the PAO1ΔwspF cells (see Fig. S1 in the supplemental material). However, the p_cdrA-gfp expression in PAO1 PCells and PAO1/p_lac-yhjH DCells was too low to indicate differences in the c-di-GMP level (Fig. 1). Nevertheless, the PAO1/p_lac-yhjH strain (DCells) was unable to form biofilms similar to the PAO1 (PCells) and PAO1ΔwspF strain during static cultivation (Fig. 1B), indicating that it had a low c-di-GMP level.

Fig 1.

(A) Expression of p_cdrA-gfp fusion in P. aeruginosa PAO1 (PCells), PAO1ΔwspF (BCells), PAO1ΔwspF/p_lac-yhjH, and PAO1/p_lac-yhjH (DCells) strains. Means and standard deviations (SD) in relative fluorescence units (RFU) from triplicate experiments are shown. *, P < 0.01. (B) Biofilm formation of P. aeruginosa PAO1 (PCells), PAO1ΔwspF (BCells), and PAO1/p_lac-yhjH (DCells) strains in microplates. Means and SD from triplicate experiments are shown. *, P < 0.01.

Due to the detection limit of the p_cdrA-gfp biosensor, we then used the p_pel-lacZ reporter gene (26) to compare the intracellular levels of c-di-GMP in SNP-dispersed biofilm cells (DCells*) and yhjH-dispersed biofilm cells (DCells). SNP reduces the intracellular c-di-GMP level in P. aeruginosa through activation of the DipA PDE (36), as an alternative to direct induction of the ectopically expressed YhjH PDE (27). The PAO1ΔwspF BCells showed a higher β-galactosidase activity than that of the PAO1 PCells (Fig. 2). SNP-dispersed PAO1 biofilm cells (DCells*) showed a level of β-galactosidase activity similar to that of PAO1/p_BAD-yhjH cells dispersed by the addition of 0.5% arabinose (DCells) (Fig. 2). The dispersed cells (DCells* and DCells) expressed lower levels of β-galactosidase activity than did the planktonic PAO1 cells (PCells) (Fig. 2). This finding suggests that freshly dispersed P. aeruginosa cells (DCells) from the biofilms had lower levels of c-di-GMP than the planktonic cells (PCells).

Fig 2.

β-Galactosidase activity of P. aeruginosa strains grown as planktonic cells or biofilm cells containing the pel-lacZ biosensors. SNP was added to both PAO1 and the PAO1ΔwspF (BCells) at final concentration of 5 μM, whereas 0.25, 0.5, and 1% arabinose was added to the PAO1/p_BAD-yhjH strain. For the biofilm cells, the β-galactosidase activity was measured in the dispersed cells. Means and SD in β-galactosidase activity from triplicate experiments are shown. *, P < 0.01.

Proteomics analysis of P. aeruginosa cells with different intracellular c-di-GMP levels.

Proteomics analysis of P. aeruginosa cells with different intracellular c-di-GMP levels was performed. Using a P value cutoff of 0.05, the abundances of 116 proteins were found to be significantly affected by low intracellular levels of c-di-GMP; the abundance of 44 proteins was upregulated, while the abundance of 72 proteins was downregulated (shown in Tables 2 and 3, respectively). As expected (3), extracellular matrix proteins were expressed more abundantly in PAO1ΔwspF strain (BCells) (Table 3), while motility and chemotaxis proteins were more abundant in PAO1/p_lac-yhjH (DCells) (Table 2).

Table 2.

Proteins whose abundance in P. aeruginosa PAO1 increased significantly in conditions of low intracellular levels of c-di-GMP

PA no.	Gene	Description of product	Peptides (95%)	Coverage (95%)	Ratio (115:114)	Pa (115:114)
PA2452		Enterochelin esterase	43	53.42	99.08	5.22E–03
PA1092	fliC	Flagellin type B	146	54.3	99.08	3.19E–02
PA3531	bfrB	Bacterioferritin	33	34.18	22.28	2.82E–04
PA4777	pmrB	PmrB: two-component regulator system signal sensor kinase PmrB	2	5.451	11.91	3.28E–02
PA1596	htpG	Heat shock protein 90	30	22.4	10.57	1.58E–02
PA2821	gstA	Glutathione S-transferase	5	20.91	10.28	3.68E–02
PA4386	groES	Cochaperonin GroES	2	45.36	9.20	3.15E–02
PA3126	ibpA	Heat shock protein Hsp20	15	24.16	8.95	1.20E–02
PA4228	pchD	Pyochelin biosynthesis protein PchD	14	13.71	8.24	3.35E–03
PA1039	ychJ	Hypothetical protein O1Q_07577	2	15.29	7.94	3.79E–02
PA3552	arnB	UDP-4-amino-4-deoxy-l-arabinose–oxoglutarate aminotransferase	4	13.09	7.38	2.85E–02
PA4942	hflK	Protease subunit HflK	6	17	7.31	2.24E–03
PA1534	recR	Recombination protein RecR	2	21.21	7.31	3.72E–02
PA4670	prs	Ribose-phosphate pyrophosphokinase	43	56.55	7.11	8.95E–05
PA4710	phuR	Heme/hemoglobin uptake outer membrane receptor PhuR	20	25.52	6.55	1.57E–04
PA4227	pchR	Transcriptional regulator PchR	3	15.54	6.31	2.34E–02
PA4224	pchG	Pyochelin biosynthetic protein PchG	16	30.37	6.25	9.10E–03
PA3158	wbpB	UDP-2-acetamido-2-deoxy-d-glucuronic acid 3-dehydrogenase, WbpB	10	22.78	6.08	3.03E–02
PA0427	oprM	Major intrinsic multiple antibiotic resistance efflux outer membrane protein OprM	13	32.16	5.40	1.56E–03
PA0018	fmt	Bifunctional UDP-glucuronic acid decarboxylase/UDP-4-amino-4-deoxy-l-arabinose formyltransferase	3	8.006	5.15	2.80E–02
PA1803	lon	Putative ATP-dependent protease	23	25.28	4.79	1.21E–04
PA3135		Putative transcriptional regulator	2	11.11	4.61	3.86E–02
PA4231	pchA	Salicylate biosynthesis isochorismate synthase	4	12.39	4.57	4.12E–02
PA3114	truA	tRNA pseudouridine synthase A	3	17.54	3.98	3.39E–02
PA3831	pepA	Multifunctional aminopeptidase A	42	38.59	3.80	2.66E–02
PA3159	wbpA	UDP-glucose/GDP-mannose dehydrogenase	17	31.42	3.66	5.69E–03
PA5054	hslU	ATP-dependent protease ATP-binding subunit HslU	6	13.87	3.44	4.85E–02
PA4225	pchF	pchF gene product	24	16.8	3.25	5.80E–06
PA0426	mexB	RND multidrug efflux transporter MexB	6	5.067	3.25	3.05E–03
PA1288	ompP1	Putative outer membrane protein	8	24.76	3.19	2.64E–02
PA4749	glmM	Phosphoglucosamine mutase	5	8.09	3.16	3.36E–02
PA5213	gcvP1	Glycine dehydrogenase	3	3.967	2.91	8.00E–03
PA3478	rhlB	Rhamnosyltransferase chain B	2	7.512	2.86	3.40E–03
PA2086	ephx	Epoxide hydrolase	3	19.33	2.75	3.59E–02
PA4336		Hypothetical protein O1Q_03368	9	23.32	2.73	3.78E–03
PA4595	yjjK	Putative ABC transporter ATP-binding protein	15	24.37	2.68	9.43E–04
PA4476		Hypothetical protein O1Q_04078	6	5.956	2.51	9.67E–03
PA2290	gcd	Glucose dehydrogenase	9	9.34	2.42	3.72E–02
PA5237	yigC	3-Octaprenyl-4-hydroxybenzoate carboxy-lyase	3	6.967	2.42	4.15E–02
PA2302	ambE	Protein AmbE	20	10.34	2.27	1.66E–02
PA4307	pctC	pctC gene product	20	31.65	2.11	2.50E–03
PA3707	wspB	Hypothetical protein O1Q_02948	2	14.79	2.07	3.07E–02
PA4588	gdhA	Glutamate dehydrogenase	30	41.57	2.05	3.19E–02
PA4226	pchE	pchE gene product	78	30.6	2.03	1.85E–05

^aSignificance was defined as having a 115:114 abundance score of >2.0 and a P value for 115:114 of <0.05. “115:114” refers to the ratio of the protein's abundance in the low c-di-GMP PAO1/p_lac-yhjH strain (strain 115) versus the high c-di-GMP PAO1ΔwspFstrain (strain 114).

Table 3.

Proteins whose abundance in P. aeruginosa PAO1 decreased significantly in conditions of low intracellular levels of c-di-GMP

PA no.	Gene	Description of product	Peptides (95%)	Coverage (95%)	Ratio (115:114)	Pa (115:114)
PA1245	aprX	Hypothetical protein O1Q_25902	21	33.82	0.03	2.74E–02
PA3064	pelA	PelA protein	2	2.743	0.04	1.71E–02
PA2395	pvdO	Protein PvdO	3	16.2	0.05	8.64E–03
PA4554	pilY1	Type 4 fimbrial biogenesis protein PilY1	6	7.666	0.05	3.52E–02
PA3613		Hypothetical protein O1Q_02478	12	20.85	0.05	6.41E–03
PA2398	fpvA	fpvA gene product	40	25.64	0.06	8.02E–06
PA2394	pvdN	Protein PvdN	57	50.82	0.06	1.24E–02
PA0059	osmC	Osmotically inducible protein OsmC	2	26.24	0.07	3.22E–02
PA5192	pckA	Phosphoenolpyruvate carboxykinase	8	18.13	0.07	7.83E–03
PA0423	pasP	Hypothetical protein O1Q_22583	17	59.16	0.08	1.49E–03
PA0781		Hypothetical protein O1Q_01027	4	6.259	0.08	1.13E–02
PA5171	arcA	Arginine deiminase	68	50.72	0.09	2.68E–05
PA5427	adhA	Alcohol dehydrogenase	18	47.08	0.10	2.33E–02
PA2397	pvdE	Pyoverdine biosynthesis protein PvdE	3	9.107	0.10	3.47E–02
PA2392	pvdP	pvdP gene product	15	28.31	0.11	4.16E–04
PA3117	asd	Aspartate-semialdehyde dehydrogenase	11	24.32	0.11	8.29E–03
PA0764	mucB	Sigma E regulatory protein, MucB/RseB	4	21.52	0.13	2.89E–02
PA3313		Hypothetical protein	17	44.48	0.13	6.00E–03
PA3190	gltB	Putative binding protein component of ABC sugar transporter	45	50	0.14	3.53E–03
PA4624	cdrB	Hypothetical protein O1Q_15965	5	12.16	0.14	3.22E–02
PA5046	maeB	Malic enzyme	22	41.23	0.15	1.34E–03
PA5312	pauC	Putative aldehyde dehydrogenase	15	16.1	0.15	4.37E–02
PA3330		Putative short-chain dehydrogenase	21	36.18	0.15	1.05E–02
PA0895	aruC	Bifunctional N-succinyldiaminopimelate-aminotransferase/acetylornithine transaminase protein	23	49.26	0.15	1.40E–02
PA4450	murA	Bifunctional cyclohexadienyl dehydrogenase/3-phosphoshikimate 1-carboxyvinyltransferase	32	35.12	0.15	9.69E–03
PA3327		Nonribosomal peptide synthetase	31	15.48	0.15	2.35E–05
PA0482	glcB	Malate synthase G	33	24.69	0.16	3.22E–02
PA3769	guaA	GMP synthase	31	35.62	0.17	1.49E–02
PA3977	hemL	Glutamate-1-semialdehyde aminotransferase	27	37	0.18	6.48E–03
PA2119	adh	Alcohol dehydrogenase	6	24.59	0.18	3.22E–02
PA0552	pgk	Phosphoglycerate kinase	12	30.23	0.19	2.51E–02
PA4687	hitA	Ferric iron-binding periplasmic protein HitA	10	29.55	0.19	2.03E–02
PA2413	pvdH	Diaminobutyrate–2-oxoglutarate aminotransferase	22	41.36	0.19	7.34E–04
PA4448	hisD	Bifunctional histidinal dehydrogenase/histidinol dehydrogenase	11	29.55	0.20	3.40E–02
PA5172	arcB	Ornithine carbamoyltransferase	67	44.94	0.20	3.04E–02
PA3686	adk	Adenylate kinase	4	25.58	0.20	3.45E–02
PA3922		Hypothetical protein O1Q_15760	7	21.54	0.21	4.15E–02
PA2385	pvdQ	3-Oxo-C12-homoserine lactone acylase PvdQ	19	22.31	0.21	4.23E–04
PA3729		Hypothetical protein O1Q_03058	3	3.634	0.22	2.48E–02
PA0314	fliY	l-Cysteine transporter of ABC system FliY	7	30.86	0.24	1.02E–02
PA2445	gcvP2	Glycine dehydrogenase	21	24.19	0.24	2.85E–04
PA0084	tssC1	Hypothetical protein O1Q_08024	8	19.48	0.24	2.89E–02
PA3452	mqoA	Malate:quinone oxidoreductase	4	10.52	0.26	7.25E–03
PA0400	metB	Putative cystathionine gamma-lyase	26	48.48	0.26	7.74E–03
PA4138	tyrS	Tyrosyl-tRNA synthetase	7	19.8	0.26	4.93E–02
PA4236	katA	Catalase	14	28.42	0.27	8.08E–03
PA0139	ahpC	Alkyl hydroperoxide reductase subunit C	16	40.64	0.27	2.34E–02
PA5322	algC	Phosphomannomutase	12	16.94	0.27	1.63E–02
PA2944	cobN	Cobaltochelatase subunit CobN	8	9.936	0.29	1.96E–02
PA3186	oprB	Glucose-sensitive porin, partial	7	7.432	0.29	3.49E–02
PA0036	trpB	Tryptophan synthase subunit beta	12	25.87	0.30	2.98E–02
PA4560	ileS	Isoleucyl-tRNA synthetase	20	16.65	0.32	4.14E–03
PA4938	purA	Adenylosuccinate synthetase	17	41.86	0.34	5.56E–03
PA0077	tssM1	Hypothetical protein O1Q_28197	5	6.378	0.34	4.85E–02
PA4266	fusA1	Elongation factor G	75	44.62	0.34	8.93E–04
PA4829	lpd3	Dihydrolipoamide dehydrogenase	20	45.91	0.35	6.32E–03
PA3328		Putative FAD-dependent monooxygenase	12	19.85	0.35	1.96E–02
PA3666	dapD	2,3,4,5-Tetrahydropyridine-2-carboxylate N-succinyltransferase	8	28.49	0.36	1.20E–02
PA2391	opmQ	Hypothetical protein O1Q_22777	7	23.94	0.37	7.47E–03
PA0956	proS	Prolyl-tRNA synthetase	14	26.8	0.37	8.13E–03
PA5131	pgm	Phosphoglyceromutase	5	11.46	0.37	3.54E–02
PA3213		Hypothetical protein O1Q_21511	7	46.4	0.41	2.14E–02
PA1833	yhfP	Putative oxidoreductase	13	34.24	0.42	4.81E–03
PA0548	tktA	Transketolase	10	17.74	0.42	4.74E–03
PA5497	nrdJa	nrdJa gene product	9	13.08	0.43	1.94E–02
PA1342		Putative binding protein component of ABC transporter	10	31.79	0.44	1.39E–02
PA0067	prlC	Oligopeptidase A	13	19.82	0.44	4.51E–02
PA1588	sucC	Succinyl coenzyme A synthetase subunit beta	52	42.53	0.45	6.66E–03
PA1010	dapA	Dihydrodipicolinate synthase	21	45.89	0.47	6.41E–03
PA3790	oprC	Outer membrane copper receptor OprC	15	24.76	0.48	7.21E–03
PA3194	edd	Phosphogluconate dehydratase	36	36.84	0.49	3.74E–03
PA1583	sdhA	Succinate dehydrogenase flavoprotein subunit	25	22.88	0.50	3.35E–02

^aSignificance was defined as having a 115:114 abundance score of <0.5 and a P value for 115:114 of <0.05. “115:114” refers to the ratio of the protein's abundance in the low c-di-GMP PAO1/p_lac-yhjH strain (strain 115) versus the high c-di-GMP PAO1ΔwspF strain (train 114).

High intracellular levels of c-di-GMP were correlated with the increased expression of proteins for synthesis of the major iron siderophore, pyoverdine (Table 3). The data were corroborated by pyoverdine fluorescence measurements showing that the production of pyoverdine in the P. aeruginosa PAO1, PAO1ΔwspF, and PAO1/p_lac-yhjH strains was in accordance with the proteomics analysis (Fig. 3).

Fig 3.

Pyoverdine production by P. aeruginosa PAO1 (PCells), PAO1ΔwspF (BCells), and PAO1/p_lac-yhjH (DCells). The pyoverdine fluorescence levels (excitation wavelength, 400 nm; emission wavelength, 450 nm) of supernatants of P. aeruginosa overnight cultures were recorded using the Tecan Infinite Pro2000 microplate reader.

Low intracellular levels of c-di-GMP were found to favor the expression of a set of virulence-associated proteins (Table 2). Surprisingly, we found that dispersal correlated with the expression of proteins that contributes to the antimicrobial peptide resistance of P. aeruginosa. Antimicrobial peptides (AMPs; e.g., defensins) are secreted by a wide-range of host cells as a response to microbial infections and act by disrupting the bacterial cell (37). Bacteria have evolved a set of inducible AMP-sensing systems (38, 39). In P. aeruginosa, the PhoP/PhoQ system and the PmrA/PmrB two-component systems can sense the presence of AMPs and upregulate genes involved in AMP resistance, including lipopolysaccharide modification (40). The arn operon (PA3552-PA3559) can also be induced by AMPs, and its expression is partially regulated by the PmrA/PmrB two-component system (40). PmrB and ArnB, typically induced by antimicrobial peptides (41, 42), were observed to be induced here by low intracellular levels of c-di-GMP (Table 2).

To examine whether the c-di-GMP effect found by proteomic analysis is on the level of transcription, we analyzed the expression of a p_pmrA-gfp transcriptional fusion (25) in PAO1 (PCells), PAO1ΔwspF (BCells), and PAO1/p_lac-yhjH (DCells). The p_pmrA-gfp fusion was expressed in all three strains in the presence of sublethal concentrations of colistin, but in the absence of any colistin, the fusion was only expressed in the PAO1/p_lac-yhjH (DCells) (Fig. 4 and see Fig. S2 in the supplemental material).

Fig 4.

P_pmr-gfp expression in P. aeruginosa PAO1 (PCells), PAO1ΔwspF (BCells), and PAO1/p_lac-yhjH (DCells) strains. Overnight cultures were diluted 10-fold into fresh ABTGC medium with or without 1 μg of colistin ml⁻¹. Portions (3 μl) of cultures representing each condition were spotted onto cover slides after 7 h of growth for imaging by fluorescence microscopy. The level of fluorescence of 30 individual p_pmr-gfp tagged bacterial cells was measured for each sample by using ImageJ. Means and SD in relative fluorescence intensity units (RFU) from 30 individual cells are shown. *, P < 0.01.

Antimicrobial peptide resistance of P. aeruginosa cells with different intracellular c-di-GMP levels.

Growth monitored in the presence of different concentrations of colistin revealed that the PAO1/p_lac-yhjH DCells were more resistant to colistin than PAO1 PCells and PAO1ΔwspF BCells during planktonic growth (Fig. 5A, B, and C). Colistin is a fast-killing bactericidal agent, and we thus measured the killing kinetics of 4 μg of colistin ml⁻¹ in PAO1 PCells and PAO1/p_lac-yhjH DCells. PAO1 PCells and PAO1ΔwspF BCells were killed rapidly by 4 μg of colistin ml⁻¹, whereas PAO1/p_lac-yhjH DCells were able to survive in the presence of 4 μg of colistin ml⁻¹ (Fig. 5D and see Fig. S3 in the supplemental material).

Fig 5.

Colistin resistance assay. P. aeruginosa PAO1 (PCells) (A), PAO1ΔwspF (BCells) (B), and PAO1/p_lac-yhjH (DCells) (C) were cultivated at 37°C in ABTGC medium with 0, 0.25, or 2 μg of colistin ml⁻¹. The OD₆₀₀ was monitored for 10 h. Means and SD from triplicate experiments are shown. (D) Fast-kill assay of P. aeruginosa PAO1 (PCells), PAO1ΔwspF (BCells), and PAO1/p_lac-yhjH (DCells) by 4 μg of colistin ml⁻¹. The proportion of dead bacterial cells was monitored by using the Live/Dead BacLight bacterial viability kits (Invitrogen) after 10 min of treatment. *, P < 0.01.

To examine whether cells that dispersed from biofilms were also more resistant to colistin than planktonic cells, the dispersal cells from biofilms of PAO1 and PAO1/p_BAD-yhjH were tested for colistin resistance, and it was observed that PAO1/p_BAD-yhjH cells dispersed from biofilms by the expression of yhjH were more resistant to colistin based on differences in growth rates (Fig. 6A and B). It was also observed that P. aeruginosa biofilms treated with dispersing agents (either arabinose or SNP) were more resistant to colistin than planktonic cells (Fig. 6C and D). In contrast, exposure of biofilms to colistin resulted in the killing of most biofilm cells and showed that a large fraction of the biofilm cells remained sensitive to colistin (Fig. 6E and F).

Fig 6.

Colistin resistance of planktonic cells (PCells), biofilm cells (BCells), and dispersed cells (DCells). Planktonic cells (PCells) of PAO1 (A), biofilm-dispersed cells (BCells) from PAO1 biofilm by 5 μM SNP (B), planktonic cells of (PCells) PAO1/p_BAD-yhjH (C), and biofilm-dispersed cells (DCells) from PAO1/p_BAD-yhjH biofilms by 1% arabinose (D) were cultivated at 37°C in ABTGC medium with 0 or 4 μg of colistin/ml. The OD₆₀₀ was monitored for 300 min. Means of three replicates are shown. (E and F) Biofilms formed by PAO1 strain on glass slides were submerged into ABTGC medium with 0 (E) and 4 (F) μg of colistin ml⁻¹ for 2 h. Live and dead cells in treated biofilms were stained by using Live/Dead BacLight bacterial viability kits, followed by confocal laser scanning microscopy imaging.

DISCUSSION

In this work, P. aeruginosa strains were constructed with a controllable intracellular c-di-GMP content to mimic the natural biofilm cells (BCells) and dispersed cells (DCells) from biofilms. Unlike the natural biofilm cells with a high level of physiological heterogeneity (43), our cells are cultivated as homogeneous planktonic cultures and are easy to manipulate. These P. aeruginosa strains thus enable us to study the overall impact of c-di-GMP on P. aeruginosa cells. Of course, the P. aeruginosa PAO1ΔwspF cells in planktonic cultures cannot functionally mimic the late stage biofilm cells since cells from mature biofilms have a slow growth rate, oxygen limitation, and a large amount of extracellular matrix material around them. Nevertheless, we showed here that the PAO1/p_lac-yhjH cells (DCells) have an intracellular c-di-GMP content similar to that of chemically dispersed cells (DCells*), which have a distinct physiology compared to planktonic cells (PCells). In fact, the PAO1/p_lac-yhjH cells were unable to form normal amounts of biofilms compared to the PAO1 cells (Fig. 1B).

It was also observed that the intracellular c-di-GMP level plays an important role in production of pyoverdine by P. aeruginosa. Pyoverdine is the major siderophore of P. aeruginosa and is required for subpopulation interactions and biofilm maturation (44). Previous work showed that pyoverdine is mainly produced by the nonmotile subpopulation at the bottom part of mature P. aeruginosa biofilms (44). The present result suggests that the nonmotile subpopulation might have a higher intracellular level of c-di-GMP compared to the motile subpopulation at the top part of mature P. aeruginosa biofilms. Further studies will be carried out to study the detailed regulation mechanism exerted by c-di-GMP on pyoverdine production.

Induction of expression of arnB and PA4773 (from the pmr operon) with 2 μg of polymyxin B ml⁻¹ but not with 0.125 μg of polymyxin B ml⁻¹ was reported to increase the polymyxin B resistance of P. aeruginosa clinical isolates from cystic fibrosis patients (34). A number of unique mutations in the pmrAB and phoPQ operons enable these clinical isolates to show an adaptive growth in medium containing 2 μg of polymyxin B ml⁻¹ after a long lag phase (34). Our study has shown for the first time that c-di-GMP signaling plays a role in AMP resistance in P. aeruginosa. Reduced c-di-GMP levels were found to induce the expression of PmrB and AnrB even without the presence of AMPs. PhoP was recently found to be able to bind c-di-GMP (45); thus, it might be an effector of c-di-GMP in regulation of AMP resistance. However, further studies are needed to elucidate the mechanistic basis of induction of the pmr and arn genes by low levels of c-di-GMP. The reported induced resistance thus confers a “protection in advance” mechanism to protect dispersed cells from the otherwise detrimental action of antibiotics on planktonic cells and may be the first finding to contradict the current dogma stating that dispersed cells are inevitably more susceptible than their sessile counterparts.