Accepted Manuscript Role of the Pseudomonas quinolone signal (PQS) in sensitising Pseudomonas aeruginosa to UVA radiation Magdalena Pezzoni, Martín Meichtry, Ramón A. Pizarro, Cristina S. Costa PII: DOI: Reference:
S1011-1344(14)00368-6 http://dx.doi.org/10.1016/j.jphotobiol.2014.11.014 JPB 9893
To appear in:
Journal of Photochemistry and Photobiology B: Biology
Received Date: Revised Date: Accepted Date:
9 June 2014 20 November 2014 24 November 2014
Please cite this article as: M. Pezzoni, M. Meichtry, R.A. Pizarro, C.S. Costa, Role of the Pseudomonas quinolone signal (PQS) in sensitising Pseudomonas aeruginosa to UVA radiation, Journal of Photochemistry and Photobiology B: Biology (2014), doi: http://dx.doi.org/10.1016/j.jphotobiol.2014.11.014
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Role of the Pseudomonas quinolone signal (PQS) in sensitising Pseudomonas aeruginosa to UVA radiation
by
Magdalena Pezzonia, Martín Meichtryb, Ramón A. Pizarroa and Cristina S. Costa*a
a
Comisión Nacional de Energía Atómica, Departamento de Radiobiología, Argentina b
Comisión Nacional de Energía Atómica, Gerencia de Química, Argentina
E-mail address: Magdalena Pezzoni
[email protected]
Martín Meichtry
[email protected]
Ramón A. Pizarro
[email protected]
Cristina S. Costa
[email protected]
*Corresponding author Address: Comisión Nacional de Energía Atómica, Departamento de Radiobiología, Avda. General Paz 1499, B1650KNA General San Martín, Buenos Aires, Argentina. Tel.: +54 11 6772 7011; fax +54 11 6772 7188. E-mail address: [email protected]
1
ABSTRACT One of the main stress factors that bacteria face in the environment is solar ultraviolet-A (UVA) radiation, which leads to lethal effects through oxidative damage. The aim of this work was to investigate the role of 2-heptyl-3-hydroxi-4-quinolone (the Pseudomonas quinolone signal or PQS) in the response of P. aeruginosa to UVA radiation. PQS is an intercellular quorum sensing signal associated to membrane vesicles which, among other functions, regulates genes related to iron acquisition, forms stable complexes with iron and participates in oxidative phenomena. UVA exposure of the wild-type PAO1 strain and a pqsA mutant unable to produce PQS revealed a sensitising role for this signal. Research into the mechanism involved in this phenomenon revealed that catalase, an essential factor in the UVA defence, is not related to PQS-mediated UVA sensitivity. Absorption of UVA by PQS produced its own photo-degradation, oxidation of the probe 2’,7’- dichlorodihydrofluorescein and generation of singlet oxygen and superoxide anion, suggesting that this signal could be acting as an endogenous photosensitiser. The results presented in this study could explain the high sensitivity to UVA of P. aeruginosa when compared to enteric bacteria.
Keywords Pseudomonas, Ultraviolet-A, UVA, PQS, 2-heptyl-3-hydroxi-4-quinolone, quorum sensing
2
1. Introduction One of the main stress factors that bacteria face in the environment is solar ultraviolet A (UVA) radiation (400-315 nm), representing the major fraction of UV radiation reaching the Earth’s surface. It has been largely demonstrated that the lethal effects of UVA are mainly due to the action of reactive oxygen species (ROS), which include singlet oxygen, superoxide anion, hydrogen peroxide and hydroxyl radical. ROS are generated by absorption of UVA by molecules known as photosensitisers (e.g. flavins and porphyrins) in presence of oxygen, and produce damage to proteins and lipids with consequent loss of bacterial viability [1-3]. A protective role in the response to UVA has been described in enterobacteria for different components of the anti-oxidative defence, including the regulators OxyR [4] and RpoS [5], and the ROS-detoxifying enzymes superoxide dismutase [6] and catalase [7]. In Pseudomonas aeruginosa, the mechanism of genetic regulation known as quorum sensing (QS) has also been demonstrated to be involved in the response of this microorganism to UVA radiation [8]. QS regulates biofilm formation, the production of virulence determinants and genes involved in stress adaptation, including those related to oxidative damage [9,10]. The beneficial role of QS in the response of P. aeruginosa to UVA has been attributed to its capacity to activate the katA gene, coding for the major hydrogen peroxide scavenging enzyme, KatA. This enzyme has proved to be essential in the optimal defence of P. aeruginosa from UVA, both in planktonic and biofilm cells [8,11]. The QS system acts through the action of two distinct types of signal molecules released to the environment, N-acylhomoserine lactones and 2-alkyl-4-quinolones (AHQs). Once a critical threshold concentration of these molecules is reached in the extracellular space, they diffuse into the cell and activate its cognate transcriptional regulators, controlling the expression of certain target genes [9]. The N-acylhomoserine lactones are the corresponding signals of the core of QS, the hierarchical las and rhl systems. With respect to AHQs, they include the most active signal 2-heptyl-3-hydroxi-
3
4-quinolone, also known as the Pseudomonas quinolone signal or PQS, and its precursor 2-heptyl-4-quinolone (HHQ) [12]. PQS provides a link between las and rhl, being activated by las and repressed by rhl [9]. Several functions have been attributed to the PQS signal. In addition to regulating a group of genes, such as those related to iron acquisition and oxidative stress response, iron-chelating activity has been reported for it [13]. PQS also enhances the formation of membrane vesicles (MVs), bilayered vesicles derived from the outer membrane. They are released to the external milieu transporting the PQS signal, proteins, toxins and DNA, a mechanism which has been implicated in virulence and enhanced survival to stressing agents [14-16]. In addition, a dual role was reported for PQS in the response of P. aeruginosa to oxidative stress: it acts as pro-oxidant, sensitising the bacteria to oxidative stress, but, on the other hand, it induces a protective anti-oxidative response [17]. Given the oxidative nature of the UVA damage and the PQS properties, the aim of this work was to investigate the role of this signal in the response to UVA radiation of P. aeruginosa. The results reported here demonstrate that PQS sensitises P. aeruginosa to the effects of UVA radiation, possibly acting as an endogenous photosensitiser. This could explain the high UVA sensitivity of P. aeruginosa compared to enteric bacteria such as E. coli [18] or Salmonella (unpublished results). In addition, this work could contribute to the understanding of the behaviour of P. aeruginosa in nature as well as provide useful information for the development of disinfection techniques employing solar or UVA radiation, an important subject since P. aeruginosa is an important opportunistic pathogen in humans.
2. Materials and Methods
2.1. Bacterial strains, plasmids and growth conditions The P. aeruginosa strains used in this study were PAO1, referred to as the wild-
4
type, and isogenic derivatives carrying insertions in the pqsA gene (PW2798 strain) or in the katA gene (PW8190 strain). Mutant strains were obtained from the Transposon mutant library generated by the Washington Genome Centre [19]. In order to evaluate the effect of PQS in the UVA response of a heterologous host, the strain Escherichia coli K12 (Laboratory collection) was employed. pLG10 plasmid, containing the pqsApqsE operon into pUCP18 vector [20], was employed in complementation assays. This plasmid carries the origin of replication of the multicopy Pseudomonas plasmid pRO1600 [21]. Conventional recombinant DNA techniques [22] were employed for plasmid manipulation, which was introduced by transformation into the PW2798 strain. The strains were routinely grown at 37ºC in Luria Bertani (LB) broth; 15 mg l-1 agar was added when solid medium was used. When required, the following compounds were added to the culture medium: PQS (Sigma), 50 µM for P. aeruginosa and 20 µM for E. coli; 2'-2' dipyridyl (Sigma), 250 µM; FeSO4.7H2O, 100 µM; carbenicillin, 200 µg ml-1.
2.2. Irradiation source Cell suspensions were irradiated with a bench with two Philips TDL 18W/08 tubes (more than 95% of the UVA emission at 365 nm), mounted on aluminium anodised reflectors enhancing the fluence rate on the section to be irradiated. The incident fluence was measured at the surface of the suspension with a 9811.58 ColeParmer Radiometer (Cole-Parmer Intruments Co., Chicago, IL). The fluence rate employed (20 W m-2) may be normally encountered in the environment [23], and it was obtained by placing the tubes about 18 cm from the suspension surface.
2.3. Irradiation procedure and viability assays Bacteria were grown to stationary growth phase, washed once and suspended in saline solution (NaCl 0.1 M) at OD650 0.4. The suspensions were divided into two 50 ml fractions, each of which was placed in a glass beaker (diameter of the exposed surface 9 cm) open to air. One of these fractions was irradiated from above at a fluence
5
rate of 20 W m-2 at the level of the free surface while the other one remained in the dark. Both suspensions were maintained in an ice bath under slow magnetic stirring throughout the procedure. Samples of cell suspensions exposed to UVA radiation or maintained in the dark were taken at the indicated times and plated on LB solid medium after dilution with 0.1 M NaCl. Plates were incubated in the dark immediately after irradiation to prevent light-induced DNA repair and the colonies were counted after 24-48 h at 37ºC. Survival was expressed as a fraction of the number of colony forming units per ml at time 0. To evaluate the effect of external PQS in the UVA response of E. coli, cell suspensions in saline solution were added before exposure with methanol-diluted PQS (final concentration 50 µM) or the solvent alone (control). Control and PQS-added suspensions were exposed to UVA or maintained in the dark according to the procedure described above.
2.4. Acridine orange staining Acridine orange was added to 10 µl of cell suspensions at a final concentration of 0.1%. After 10 min, 2 ml of 0.1 M NaCl were added and the suspensions were filtered through 0.22 µm polycarbonate black filter. The filters were placed on a slide and the cells were visualised by epifluorescence microscopy.
2.5. Siderophore quantification Cells from stationary phase cultures were removed by centrifugation and the supernatants were filtered through 0.22 µm nitrocellulose filters for siderophore quantification. Absorbance was measured at 405 nm for pyoverdine quantification [24]. For pyochelin quantification, fluorescence (360 excitation, 535 emission) was measured with a microplate reader [24]. The values obtained were referred to the OD650 of the corresponding cultures.
6
2.6. Hydrogen peroxide and streptonigrin sensitivity To assay sensitivity to hydrogen peroxide, 100 µl of stationary phase cultures were diluted in 4 ml of molten LB containing 0.75% agar, and layered on LB agar plates. Sterile filter paper disks saturated with 8 µl of 30% hydrogen peroxide were placed on the layer. Sensitivity was recorded as the diameter of growth inhibition after 24 h growth at 37ºC. To assay sensitivity to streptonigrin by the disk diffusion assay, approximately 5x107 viable cells from control and irradiated suspensions were diluted in 4 ml of molten LB and layered on LB plates as described above. Sterile filter paper disks saturated with 5 µl of 1 mg ml-1 streptonigrin diluted in dimethyl sulfoxide were placed on the cell layer. Sensitivity was recorded as the diameter of the growth inhibition halo after 24 h at 37ºC. To assay sensitivity to streptonigrin in liquid medium, cultures grown to OD650 1 were divided into three 1 ml fractions and added with 20 µM streptonigrin, 20 µM streptonigrin plus 100 µM FeCl3.6H2O or 10 µl dimethyl sulfoxide (control). Samples were taken at time 0 and after 60 min incubation at 37ºC, and plated on LB solid medium after dilution with 0.1 M NaCl. Plates were incubated at 37ºC and the colonies were counted after 24 h. Survival was expressed as the fraction of the number of colony forming units per ml at time 0.
2.7. Catalase assay To determine total catalase activity, stationary phase cultures were centrifuged at 10,000 g for 10 min at 4ºC. The cell pellet was suspended in ice-cold 50 mM sodium phosphate buffer (pH 7), sonicated in an ice-water bath, and clarified by centrifugation at 12,000 g for 10 min at 4º C to obtain a cell extract. Total catalase activity was monitored by following the decomposition of 10 mM H2O2 according to Aebi [25]. One unit of activity was that which decomposes 1 µmol of hydrogen peroxide per min per mg of protein. Protein content was determined by Lowry’s method [26].
7
2.8. Chemiluminescence assay Photo-emissive species were followed by means of a liquid scintillation system in the “out of coincidence” mode [27]. For this purpose, during the UVA treatment, 5 ml aliquots were transferred to the scintillation system, equipped with photomultipliers sensitive in the blue region up to 600-650 nm (Tri-Carb, Model 1500, Packard Instruments Co.). Chemiluminescence values were expressed as counts per min (cpm) per mg of protein.
2.9. PQS extraction and quantification PQS was extracted from cells and culture supernatants as described by Fletcher et al. [28]. Briefly, 25 ml stationary phase cultures were centrifuged at 10,000 g for 10 min at 4ºC. For cell extraction, 25 ml of methanol were added to cell pellets, which were suspended by vigorous vortexing and maintained at room temperature for 10 min. This suspension was centrifuged at 10,000 g for 10 min at 4ºC, the pellet was discarded, and the extract was filtered through 0.22 µm filters. For supernatant extraction, supernatants from stationary phase cultures were filtered through 0.22 µm filters to remove unpelleted cells and extracted twice by addition of 25 ml acidified ethyl acetate. The extraction mixtures from both fractions were evaporated to dryness with a rotavapor and recovered with two sequential additions of methanol. The solvent was dried under a stream of nitrogen gas, and the extracts suspended in 50 µl methanol were stored at -20ºC until required. The presence of AHQs (PQS and its precursor HHQ) in supernatants and cell fractions was first evaluated by a bioassay, analyzing the production of β-galactosidase activity by the strain PW2798, which can be employed as AHQs biosensor [28]. PW2798 does not synthesise AHQs because it carries an in-frame insertion of the transposable element ISlacZ/hah-tc in the pqsA gene, the first gene of operon pqsABCDE, responsible for the synthesis of PQS and HHQ. Both signals bind to the
8
regulator PqsR (MvfR) to promote their biosynthesis from the pqsA promoter, so that in presence of exogenous PQS we observed that PW2798 synthesises β-Galactosidase in a dose-dependant way (data not shown). To quantify AHQs by this indirect method, 5 ml of logarithmic cultures of PW2798 strain were added with 10 µl of extracts and grown until stationary phase, when the β-Galactosidase activity was assayed. Methanol and PQS 20 µM were employed as negative and positive controls, respectively. βGalactosidase activity was assayed by following the decomposition of o-nitrophenyl-βD-galactoside at 420 nm as described by Miller in cells treated with SDS and chloroform [29]. Specific activities are expressed in Miller units referred to OD650. The same extracts employed in the bioassay were analyzed by thin layer chromatography (TLC) to detect PQS, using synthetic PQS as standard [28]. For this purpose, 5 µl extracts were spotted onto TLC plates prepared with silica gel 60 F254 and separated using 95:5 dicholoromethane:methanol as solvent. The spots were visualised under ultraviolet light and the plate was photographed.
2.10. DCFH assays The probe 2’,7’-dichlorodihydrofluorescein (DCFH) was employed to evaluate the oxidative potential of PQS exposed to UVA radiation. Different concentrations of PQS were added to a solution containing Tris HCl 50 mM (pH 8) and 10 µM DCFH, in a final volume of 200 µl, in duplicate. Negative controls without PQS were performed. The samples were placed in a 96-well microplate and exposed to UVA at a fluence rate of 20 W m-2 for 90 min (total UVA dose 108 kJ m-2) or maintained in the dark, open to air. To analyze the effect of the UVA dose on the oxidative potential of PQS, a similar assay was performed, but PQS was used at a final concentration of 20 µM and the samples were exposed at the same fluence rate for 30, 60 and 90 minutes (total UVA dose 36, 72 and 108 kJ m-2) or maintained in the dark. The fluorescence produced by 2’,7’-dichlorofluorescein (DCF) generated by DCFH oxidation was measured with a
9
microplate reader (485 nm excitation, 535 nm emission) after incubation at 37º C for 1 h.
2.11. Quantification of singlet oxygen and superoxide anion The production of singlet oxygen under aerobic conditions was measured as described [30]. Briefly, a 10 ml assay solution of 4 µM N,N-dimethyl-p-nitrosoaniline (RNO) and 10 mM L-histidine in 0.01 M potassium phosphate buffer (pH 7) without or with 50 µM PQS, was irradiated at a fluence rate of 20 W m-2 for 120 min or maintained in the dark in small Petri dishes (diameter 4.5 cm). The production of singlet oxygen was monitored by measuring the decrease in RNO absorbance at 440 nm. Sodium azide (NaN3) 2 mM was used as specific quencher. The production of superoxide anion was evaluated by a similar system, recording the reduction of nitro-blue tetrazolium (NBT) to nitro-blue formazan (NBF) spectrophotometrically [30]. For this purpose, a 10 ml solution containing 0.167 mM NBT in 0.01 sodium carbonate buffer (pH 10) without or with 50 µM PQS, was irradiated at a fluence of 20 W m-2 for 120 min or maintained in the dark. The production of NBF was evaluated by recording the increase in absorbance at 560 nm. Superoxide dismutase (Cayman, 25 U ml-1) was employed as quencher. Riboflavine 50 µM exposed to the same total UVA doses was used as positive control of singlet oxygen and superoxide anion production.
2.12. Fe2+ release by exposure of the PQS-Fe3+ complex to UVA The PQS-Fe3+ complex was prepared by mixing methanol diluted PQS and a solution of Fe2(SO4)3 at a ratio 3:1 (PQS:Fe3+) [31] . The colour of the solution changing to reddish-pink confirmed the formation of the complex [13]. Release of Fe2+ from the complex by UVA exposure was analyzed following Aubailly et al. [32] on ferritin photoreduction. For this purpose, a 4 ml PQS-Fe3+ solution (final PQS concentration 50 µM) was placed in small Petri dishes (diameter 4.5 cm) and exposed to UVA (20 W m-2 for 120 min) or maintained in the dark. Aliquots were removed at intervals and Fe2+ was
10
quantified by a colorimetric method based on the use of o-phenanthroline [33]. A Fe(NH4)2(SO4)2.6H2O solution was employed for calibration curves.
2.13. Absorption spectrum of PQS and photodegradation UV-Vis spectra of PQS 100 µM in methanol was done in 1 cm pathlenght quartz cells using a Hewlett Packard 8453 diode-array UV-Vis spectrophotometer. The decomposition of PQS under UVA radiation was studied by direct exposure in the quartz cells of the corresponding methanol solution, using the same irradiation setup indicated in Section 2.2.
2.14. Photodegradation of 2’deoxyguanosine (2’-dGuO) and DNA fragmentation The photo-oxidative degradation of 2’dGuO was evaluated following a previous paper [30]. For this purpose, a solution containing 25 µg ml-1 2’dGuO in sodium carbonate buffer 0.01 M (pH10) was exposed to UVA (fluence rate 20 W m-2) for 300 min in presence or absence of PQS 50 µM, in small Petri dishes (diameter 4.5 cm). For control assays, the dishes were maintained in the dark. Samples were taken at 0, 100, 200 and 300 min to monitor 2’dGuO content, measured by absorbance at 260 nm. Oxidative DNA fragmentation was evaluated by exposing to UVA (fluence rate 20 W m2
) 0.5 ml of aqueous dilutions of bacterial DNA, in presence or absence of PQS 50 µM,
in a 4-well microplate. Identical assays were maintained in the dark as control. Samples were taken at the same times as indicated above and subjected to agarose gel electrophoresis using a 1% agarose gel stained with Gel Green and photographed over ultraviolet light. Riboflavine 50 µM exposed to the same UVA doses was employed as positive control of 2’dGuO photodegradation and DNA fragmentation.
3. Results
11
3.1. Role of PQS on the response of P. aeruginosa to UVA radiation As reported by Häusler and Becker [17], PQS has a primary anti-oxidant activity; however, it also exhibits a pro-oxidant activity enhancing the susceptibility of bacteria to external stress. These properties and its ability to form complexes with iron [13], a potent inducer of oxidative stress, posed a question with regard to the role of this signal in the response of P. aeruginosa to UVA radiation. In order to analyse the issue, the strains PAO1 and an isogenic non-PQS producer derivative (PW2798) were exposed to UVA for 300 min at a fluence rate of 20 W m-2. Survival was evaluated by plating on solid medium. As shown in Fig. 1, the wild-type maintained its survival for about 120 min, after which a marked decay of cell viability was observed. In contrast, the pqsA strain required a longer irradiation time to show a lethal effect. After 5 hs irradiation, the wild-type was about two orders more sensitive than the pqsA mutant. Since PQS promotes cell aggregation [17], samples of both strains were taken during the UVA exposure, stained with acridine orange and observed under epifluorescence microscopy to verify that the decreased cell count of PAO1 strain is not related to the formation of clumps. No relevant difference was observed between strains (data not shown), indicating differential sensitivity to the radiation. To verify that the UVA-resistant phenotype is due to the pqsA mutation, pLG10 plasmid (carrying the pqsA-E operon) was introduced into the mutant strain. The survival curve of the complemented strain showed a biphasic behaviour (fast inactivation followed by slow inactivation) (Fig. 1). The low viability of the pqsA pLG10 strain at low UVA doses compared to the wild-type could be attributed to high PQS levels produced by the multicopy plasmid [20]. The second part of the curve could be explained by the presence of an UVA-resistant subpopulation with lower levels of PQS able to be reached by the radiation. The results suggest that the quorum sensing signal PQS sensitises the bacteria to the effects of UVA radiation.
3.2. Role of PQS in iron homeostasis and catalase production: relationship with the
12
UVA response Transcriptomic studies in P. aeruginosa revealed that PQS induces genes involved in siderophore-mediated iron uptake [13,34,35]. Siderophores are iron-binding compounds whose expression is blocked when the level of intracellular Fe2+ reaches a certain threshold. In a study analysing the signalling pathways of PQS, it was demonstrated that induction of siderophore expression by PQS is due to the iron deprivation state produced by its iron-chelating activity [34]. Stable complexes of two or three PQS molecules bound to one Fe3+ atom were obtained in vitro and isolated from PQS-supplemented PAO1 cultures [13]. On the basis of these data, it was suggested that when PQS binds Fe3+, P. aeruginosa senses a lack of available iron even in the iron-sufficient LB broth [13,34]. Iron is required for the expression and activity of P. aeruginosa KatA [36,37], an essential factor in the defence of P. aeruginosa against UVA [8,11]. Then, we analyzed whether the role of PQS in UVA response could be related to a decrease in catalase activity produced by a state of iron deprivation. Since PQS causes iron depletion, we first investigated the concentration of siderophores pyoverdine and pyochelin in supernatants of stationary phase cultures of the PAO1 and pqsA strains grown in absence or presence of PQS; 2'2-dipyridyl, a known iron chelator, was also tested as a control for iron depletion. As shown in Fig. 2, a significant (P
S1011-1344(14)00368-6 http://dx.doi.org/10.1016/j.jphotobiol.2014.11.014 JPB 9893
To appear in:
Journal of Photochemistry and Photobiology B: Biology
Received Date: Revised Date: Accepted Date:
9 June 2014 20 November 2014 24 November 2014
Please cite this article as: M. Pezzoni, M. Meichtry, R.A. Pizarro, C.S. Costa, Role of the Pseudomonas quinolone signal (PQS) in sensitising Pseudomonas aeruginosa to UVA radiation, Journal of Photochemistry and Photobiology B: Biology (2014), doi: http://dx.doi.org/10.1016/j.jphotobiol.2014.11.014
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Role of the Pseudomonas quinolone signal (PQS) in sensitising Pseudomonas aeruginosa to UVA radiation
by
Magdalena Pezzonia, Martín Meichtryb, Ramón A. Pizarroa and Cristina S. Costa*a
a
Comisión Nacional de Energía Atómica, Departamento de Radiobiología, Argentina b
Comisión Nacional de Energía Atómica, Gerencia de Química, Argentina
E-mail address: Magdalena Pezzoni
[email protected]
Martín Meichtry
[email protected]
Ramón A. Pizarro
[email protected]
Cristina S. Costa
[email protected]
*Corresponding author Address: Comisión Nacional de Energía Atómica, Departamento de Radiobiología, Avda. General Paz 1499, B1650KNA General San Martín, Buenos Aires, Argentina. Tel.: +54 11 6772 7011; fax +54 11 6772 7188. E-mail address: [email protected]
1
ABSTRACT One of the main stress factors that bacteria face in the environment is solar ultraviolet-A (UVA) radiation, which leads to lethal effects through oxidative damage. The aim of this work was to investigate the role of 2-heptyl-3-hydroxi-4-quinolone (the Pseudomonas quinolone signal or PQS) in the response of P. aeruginosa to UVA radiation. PQS is an intercellular quorum sensing signal associated to membrane vesicles which, among other functions, regulates genes related to iron acquisition, forms stable complexes with iron and participates in oxidative phenomena. UVA exposure of the wild-type PAO1 strain and a pqsA mutant unable to produce PQS revealed a sensitising role for this signal. Research into the mechanism involved in this phenomenon revealed that catalase, an essential factor in the UVA defence, is not related to PQS-mediated UVA sensitivity. Absorption of UVA by PQS produced its own photo-degradation, oxidation of the probe 2’,7’- dichlorodihydrofluorescein and generation of singlet oxygen and superoxide anion, suggesting that this signal could be acting as an endogenous photosensitiser. The results presented in this study could explain the high sensitivity to UVA of P. aeruginosa when compared to enteric bacteria.
Keywords Pseudomonas, Ultraviolet-A, UVA, PQS, 2-heptyl-3-hydroxi-4-quinolone, quorum sensing
2
1. Introduction One of the main stress factors that bacteria face in the environment is solar ultraviolet A (UVA) radiation (400-315 nm), representing the major fraction of UV radiation reaching the Earth’s surface. It has been largely demonstrated that the lethal effects of UVA are mainly due to the action of reactive oxygen species (ROS), which include singlet oxygen, superoxide anion, hydrogen peroxide and hydroxyl radical. ROS are generated by absorption of UVA by molecules known as photosensitisers (e.g. flavins and porphyrins) in presence of oxygen, and produce damage to proteins and lipids with consequent loss of bacterial viability [1-3]. A protective role in the response to UVA has been described in enterobacteria for different components of the anti-oxidative defence, including the regulators OxyR [4] and RpoS [5], and the ROS-detoxifying enzymes superoxide dismutase [6] and catalase [7]. In Pseudomonas aeruginosa, the mechanism of genetic regulation known as quorum sensing (QS) has also been demonstrated to be involved in the response of this microorganism to UVA radiation [8]. QS regulates biofilm formation, the production of virulence determinants and genes involved in stress adaptation, including those related to oxidative damage [9,10]. The beneficial role of QS in the response of P. aeruginosa to UVA has been attributed to its capacity to activate the katA gene, coding for the major hydrogen peroxide scavenging enzyme, KatA. This enzyme has proved to be essential in the optimal defence of P. aeruginosa from UVA, both in planktonic and biofilm cells [8,11]. The QS system acts through the action of two distinct types of signal molecules released to the environment, N-acylhomoserine lactones and 2-alkyl-4-quinolones (AHQs). Once a critical threshold concentration of these molecules is reached in the extracellular space, they diffuse into the cell and activate its cognate transcriptional regulators, controlling the expression of certain target genes [9]. The N-acylhomoserine lactones are the corresponding signals of the core of QS, the hierarchical las and rhl systems. With respect to AHQs, they include the most active signal 2-heptyl-3-hydroxi-
3
4-quinolone, also known as the Pseudomonas quinolone signal or PQS, and its precursor 2-heptyl-4-quinolone (HHQ) [12]. PQS provides a link between las and rhl, being activated by las and repressed by rhl [9]. Several functions have been attributed to the PQS signal. In addition to regulating a group of genes, such as those related to iron acquisition and oxidative stress response, iron-chelating activity has been reported for it [13]. PQS also enhances the formation of membrane vesicles (MVs), bilayered vesicles derived from the outer membrane. They are released to the external milieu transporting the PQS signal, proteins, toxins and DNA, a mechanism which has been implicated in virulence and enhanced survival to stressing agents [14-16]. In addition, a dual role was reported for PQS in the response of P. aeruginosa to oxidative stress: it acts as pro-oxidant, sensitising the bacteria to oxidative stress, but, on the other hand, it induces a protective anti-oxidative response [17]. Given the oxidative nature of the UVA damage and the PQS properties, the aim of this work was to investigate the role of this signal in the response to UVA radiation of P. aeruginosa. The results reported here demonstrate that PQS sensitises P. aeruginosa to the effects of UVA radiation, possibly acting as an endogenous photosensitiser. This could explain the high UVA sensitivity of P. aeruginosa compared to enteric bacteria such as E. coli [18] or Salmonella (unpublished results). In addition, this work could contribute to the understanding of the behaviour of P. aeruginosa in nature as well as provide useful information for the development of disinfection techniques employing solar or UVA radiation, an important subject since P. aeruginosa is an important opportunistic pathogen in humans.
2. Materials and Methods
2.1. Bacterial strains, plasmids and growth conditions The P. aeruginosa strains used in this study were PAO1, referred to as the wild-
4
type, and isogenic derivatives carrying insertions in the pqsA gene (PW2798 strain) or in the katA gene (PW8190 strain). Mutant strains were obtained from the Transposon mutant library generated by the Washington Genome Centre [19]. In order to evaluate the effect of PQS in the UVA response of a heterologous host, the strain Escherichia coli K12 (Laboratory collection) was employed. pLG10 plasmid, containing the pqsApqsE operon into pUCP18 vector [20], was employed in complementation assays. This plasmid carries the origin of replication of the multicopy Pseudomonas plasmid pRO1600 [21]. Conventional recombinant DNA techniques [22] were employed for plasmid manipulation, which was introduced by transformation into the PW2798 strain. The strains were routinely grown at 37ºC in Luria Bertani (LB) broth; 15 mg l-1 agar was added when solid medium was used. When required, the following compounds were added to the culture medium: PQS (Sigma), 50 µM for P. aeruginosa and 20 µM for E. coli; 2'-2' dipyridyl (Sigma), 250 µM; FeSO4.7H2O, 100 µM; carbenicillin, 200 µg ml-1.
2.2. Irradiation source Cell suspensions were irradiated with a bench with two Philips TDL 18W/08 tubes (more than 95% of the UVA emission at 365 nm), mounted on aluminium anodised reflectors enhancing the fluence rate on the section to be irradiated. The incident fluence was measured at the surface of the suspension with a 9811.58 ColeParmer Radiometer (Cole-Parmer Intruments Co., Chicago, IL). The fluence rate employed (20 W m-2) may be normally encountered in the environment [23], and it was obtained by placing the tubes about 18 cm from the suspension surface.
2.3. Irradiation procedure and viability assays Bacteria were grown to stationary growth phase, washed once and suspended in saline solution (NaCl 0.1 M) at OD650 0.4. The suspensions were divided into two 50 ml fractions, each of which was placed in a glass beaker (diameter of the exposed surface 9 cm) open to air. One of these fractions was irradiated from above at a fluence
5
rate of 20 W m-2 at the level of the free surface while the other one remained in the dark. Both suspensions were maintained in an ice bath under slow magnetic stirring throughout the procedure. Samples of cell suspensions exposed to UVA radiation or maintained in the dark were taken at the indicated times and plated on LB solid medium after dilution with 0.1 M NaCl. Plates were incubated in the dark immediately after irradiation to prevent light-induced DNA repair and the colonies were counted after 24-48 h at 37ºC. Survival was expressed as a fraction of the number of colony forming units per ml at time 0. To evaluate the effect of external PQS in the UVA response of E. coli, cell suspensions in saline solution were added before exposure with methanol-diluted PQS (final concentration 50 µM) or the solvent alone (control). Control and PQS-added suspensions were exposed to UVA or maintained in the dark according to the procedure described above.
2.4. Acridine orange staining Acridine orange was added to 10 µl of cell suspensions at a final concentration of 0.1%. After 10 min, 2 ml of 0.1 M NaCl were added and the suspensions were filtered through 0.22 µm polycarbonate black filter. The filters were placed on a slide and the cells were visualised by epifluorescence microscopy.
2.5. Siderophore quantification Cells from stationary phase cultures were removed by centrifugation and the supernatants were filtered through 0.22 µm nitrocellulose filters for siderophore quantification. Absorbance was measured at 405 nm for pyoverdine quantification [24]. For pyochelin quantification, fluorescence (360 excitation, 535 emission) was measured with a microplate reader [24]. The values obtained were referred to the OD650 of the corresponding cultures.
6
2.6. Hydrogen peroxide and streptonigrin sensitivity To assay sensitivity to hydrogen peroxide, 100 µl of stationary phase cultures were diluted in 4 ml of molten LB containing 0.75% agar, and layered on LB agar plates. Sterile filter paper disks saturated with 8 µl of 30% hydrogen peroxide were placed on the layer. Sensitivity was recorded as the diameter of growth inhibition after 24 h growth at 37ºC. To assay sensitivity to streptonigrin by the disk diffusion assay, approximately 5x107 viable cells from control and irradiated suspensions were diluted in 4 ml of molten LB and layered on LB plates as described above. Sterile filter paper disks saturated with 5 µl of 1 mg ml-1 streptonigrin diluted in dimethyl sulfoxide were placed on the cell layer. Sensitivity was recorded as the diameter of the growth inhibition halo after 24 h at 37ºC. To assay sensitivity to streptonigrin in liquid medium, cultures grown to OD650 1 were divided into three 1 ml fractions and added with 20 µM streptonigrin, 20 µM streptonigrin plus 100 µM FeCl3.6H2O or 10 µl dimethyl sulfoxide (control). Samples were taken at time 0 and after 60 min incubation at 37ºC, and plated on LB solid medium after dilution with 0.1 M NaCl. Plates were incubated at 37ºC and the colonies were counted after 24 h. Survival was expressed as the fraction of the number of colony forming units per ml at time 0.
2.7. Catalase assay To determine total catalase activity, stationary phase cultures were centrifuged at 10,000 g for 10 min at 4ºC. The cell pellet was suspended in ice-cold 50 mM sodium phosphate buffer (pH 7), sonicated in an ice-water bath, and clarified by centrifugation at 12,000 g for 10 min at 4º C to obtain a cell extract. Total catalase activity was monitored by following the decomposition of 10 mM H2O2 according to Aebi [25]. One unit of activity was that which decomposes 1 µmol of hydrogen peroxide per min per mg of protein. Protein content was determined by Lowry’s method [26].
7
2.8. Chemiluminescence assay Photo-emissive species were followed by means of a liquid scintillation system in the “out of coincidence” mode [27]. For this purpose, during the UVA treatment, 5 ml aliquots were transferred to the scintillation system, equipped with photomultipliers sensitive in the blue region up to 600-650 nm (Tri-Carb, Model 1500, Packard Instruments Co.). Chemiluminescence values were expressed as counts per min (cpm) per mg of protein.
2.9. PQS extraction and quantification PQS was extracted from cells and culture supernatants as described by Fletcher et al. [28]. Briefly, 25 ml stationary phase cultures were centrifuged at 10,000 g for 10 min at 4ºC. For cell extraction, 25 ml of methanol were added to cell pellets, which were suspended by vigorous vortexing and maintained at room temperature for 10 min. This suspension was centrifuged at 10,000 g for 10 min at 4ºC, the pellet was discarded, and the extract was filtered through 0.22 µm filters. For supernatant extraction, supernatants from stationary phase cultures were filtered through 0.22 µm filters to remove unpelleted cells and extracted twice by addition of 25 ml acidified ethyl acetate. The extraction mixtures from both fractions were evaporated to dryness with a rotavapor and recovered with two sequential additions of methanol. The solvent was dried under a stream of nitrogen gas, and the extracts suspended in 50 µl methanol were stored at -20ºC until required. The presence of AHQs (PQS and its precursor HHQ) in supernatants and cell fractions was first evaluated by a bioassay, analyzing the production of β-galactosidase activity by the strain PW2798, which can be employed as AHQs biosensor [28]. PW2798 does not synthesise AHQs because it carries an in-frame insertion of the transposable element ISlacZ/hah-tc in the pqsA gene, the first gene of operon pqsABCDE, responsible for the synthesis of PQS and HHQ. Both signals bind to the
8
regulator PqsR (MvfR) to promote their biosynthesis from the pqsA promoter, so that in presence of exogenous PQS we observed that PW2798 synthesises β-Galactosidase in a dose-dependant way (data not shown). To quantify AHQs by this indirect method, 5 ml of logarithmic cultures of PW2798 strain were added with 10 µl of extracts and grown until stationary phase, when the β-Galactosidase activity was assayed. Methanol and PQS 20 µM were employed as negative and positive controls, respectively. βGalactosidase activity was assayed by following the decomposition of o-nitrophenyl-βD-galactoside at 420 nm as described by Miller in cells treated with SDS and chloroform [29]. Specific activities are expressed in Miller units referred to OD650. The same extracts employed in the bioassay were analyzed by thin layer chromatography (TLC) to detect PQS, using synthetic PQS as standard [28]. For this purpose, 5 µl extracts were spotted onto TLC plates prepared with silica gel 60 F254 and separated using 95:5 dicholoromethane:methanol as solvent. The spots were visualised under ultraviolet light and the plate was photographed.
2.10. DCFH assays The probe 2’,7’-dichlorodihydrofluorescein (DCFH) was employed to evaluate the oxidative potential of PQS exposed to UVA radiation. Different concentrations of PQS were added to a solution containing Tris HCl 50 mM (pH 8) and 10 µM DCFH, in a final volume of 200 µl, in duplicate. Negative controls without PQS were performed. The samples were placed in a 96-well microplate and exposed to UVA at a fluence rate of 20 W m-2 for 90 min (total UVA dose 108 kJ m-2) or maintained in the dark, open to air. To analyze the effect of the UVA dose on the oxidative potential of PQS, a similar assay was performed, but PQS was used at a final concentration of 20 µM and the samples were exposed at the same fluence rate for 30, 60 and 90 minutes (total UVA dose 36, 72 and 108 kJ m-2) or maintained in the dark. The fluorescence produced by 2’,7’-dichlorofluorescein (DCF) generated by DCFH oxidation was measured with a
9
microplate reader (485 nm excitation, 535 nm emission) after incubation at 37º C for 1 h.
2.11. Quantification of singlet oxygen and superoxide anion The production of singlet oxygen under aerobic conditions was measured as described [30]. Briefly, a 10 ml assay solution of 4 µM N,N-dimethyl-p-nitrosoaniline (RNO) and 10 mM L-histidine in 0.01 M potassium phosphate buffer (pH 7) without or with 50 µM PQS, was irradiated at a fluence rate of 20 W m-2 for 120 min or maintained in the dark in small Petri dishes (diameter 4.5 cm). The production of singlet oxygen was monitored by measuring the decrease in RNO absorbance at 440 nm. Sodium azide (NaN3) 2 mM was used as specific quencher. The production of superoxide anion was evaluated by a similar system, recording the reduction of nitro-blue tetrazolium (NBT) to nitro-blue formazan (NBF) spectrophotometrically [30]. For this purpose, a 10 ml solution containing 0.167 mM NBT in 0.01 sodium carbonate buffer (pH 10) without or with 50 µM PQS, was irradiated at a fluence of 20 W m-2 for 120 min or maintained in the dark. The production of NBF was evaluated by recording the increase in absorbance at 560 nm. Superoxide dismutase (Cayman, 25 U ml-1) was employed as quencher. Riboflavine 50 µM exposed to the same total UVA doses was used as positive control of singlet oxygen and superoxide anion production.
2.12. Fe2+ release by exposure of the PQS-Fe3+ complex to UVA The PQS-Fe3+ complex was prepared by mixing methanol diluted PQS and a solution of Fe2(SO4)3 at a ratio 3:1 (PQS:Fe3+) [31] . The colour of the solution changing to reddish-pink confirmed the formation of the complex [13]. Release of Fe2+ from the complex by UVA exposure was analyzed following Aubailly et al. [32] on ferritin photoreduction. For this purpose, a 4 ml PQS-Fe3+ solution (final PQS concentration 50 µM) was placed in small Petri dishes (diameter 4.5 cm) and exposed to UVA (20 W m-2 for 120 min) or maintained in the dark. Aliquots were removed at intervals and Fe2+ was
10
quantified by a colorimetric method based on the use of o-phenanthroline [33]. A Fe(NH4)2(SO4)2.6H2O solution was employed for calibration curves.
2.13. Absorption spectrum of PQS and photodegradation UV-Vis spectra of PQS 100 µM in methanol was done in 1 cm pathlenght quartz cells using a Hewlett Packard 8453 diode-array UV-Vis spectrophotometer. The decomposition of PQS under UVA radiation was studied by direct exposure in the quartz cells of the corresponding methanol solution, using the same irradiation setup indicated in Section 2.2.
2.14. Photodegradation of 2’deoxyguanosine (2’-dGuO) and DNA fragmentation The photo-oxidative degradation of 2’dGuO was evaluated following a previous paper [30]. For this purpose, a solution containing 25 µg ml-1 2’dGuO in sodium carbonate buffer 0.01 M (pH10) was exposed to UVA (fluence rate 20 W m-2) for 300 min in presence or absence of PQS 50 µM, in small Petri dishes (diameter 4.5 cm). For control assays, the dishes were maintained in the dark. Samples were taken at 0, 100, 200 and 300 min to monitor 2’dGuO content, measured by absorbance at 260 nm. Oxidative DNA fragmentation was evaluated by exposing to UVA (fluence rate 20 W m2
) 0.5 ml of aqueous dilutions of bacterial DNA, in presence or absence of PQS 50 µM,
in a 4-well microplate. Identical assays were maintained in the dark as control. Samples were taken at the same times as indicated above and subjected to agarose gel electrophoresis using a 1% agarose gel stained with Gel Green and photographed over ultraviolet light. Riboflavine 50 µM exposed to the same UVA doses was employed as positive control of 2’dGuO photodegradation and DNA fragmentation.
3. Results
11
3.1. Role of PQS on the response of P. aeruginosa to UVA radiation As reported by Häusler and Becker [17], PQS has a primary anti-oxidant activity; however, it also exhibits a pro-oxidant activity enhancing the susceptibility of bacteria to external stress. These properties and its ability to form complexes with iron [13], a potent inducer of oxidative stress, posed a question with regard to the role of this signal in the response of P. aeruginosa to UVA radiation. In order to analyse the issue, the strains PAO1 and an isogenic non-PQS producer derivative (PW2798) were exposed to UVA for 300 min at a fluence rate of 20 W m-2. Survival was evaluated by plating on solid medium. As shown in Fig. 1, the wild-type maintained its survival for about 120 min, after which a marked decay of cell viability was observed. In contrast, the pqsA strain required a longer irradiation time to show a lethal effect. After 5 hs irradiation, the wild-type was about two orders more sensitive than the pqsA mutant. Since PQS promotes cell aggregation [17], samples of both strains were taken during the UVA exposure, stained with acridine orange and observed under epifluorescence microscopy to verify that the decreased cell count of PAO1 strain is not related to the formation of clumps. No relevant difference was observed between strains (data not shown), indicating differential sensitivity to the radiation. To verify that the UVA-resistant phenotype is due to the pqsA mutation, pLG10 plasmid (carrying the pqsA-E operon) was introduced into the mutant strain. The survival curve of the complemented strain showed a biphasic behaviour (fast inactivation followed by slow inactivation) (Fig. 1). The low viability of the pqsA pLG10 strain at low UVA doses compared to the wild-type could be attributed to high PQS levels produced by the multicopy plasmid [20]. The second part of the curve could be explained by the presence of an UVA-resistant subpopulation with lower levels of PQS able to be reached by the radiation. The results suggest that the quorum sensing signal PQS sensitises the bacteria to the effects of UVA radiation.
3.2. Role of PQS in iron homeostasis and catalase production: relationship with the
12
UVA response Transcriptomic studies in P. aeruginosa revealed that PQS induces genes involved in siderophore-mediated iron uptake [13,34,35]. Siderophores are iron-binding compounds whose expression is blocked when the level of intracellular Fe2+ reaches a certain threshold. In a study analysing the signalling pathways of PQS, it was demonstrated that induction of siderophore expression by PQS is due to the iron deprivation state produced by its iron-chelating activity [34]. Stable complexes of two or three PQS molecules bound to one Fe3+ atom were obtained in vitro and isolated from PQS-supplemented PAO1 cultures [13]. On the basis of these data, it was suggested that when PQS binds Fe3+, P. aeruginosa senses a lack of available iron even in the iron-sufficient LB broth [13,34]. Iron is required for the expression and activity of P. aeruginosa KatA [36,37], an essential factor in the defence of P. aeruginosa against UVA [8,11]. Then, we analyzed whether the role of PQS in UVA response could be related to a decrease in catalase activity produced by a state of iron deprivation. Since PQS causes iron depletion, we first investigated the concentration of siderophores pyoverdine and pyochelin in supernatants of stationary phase cultures of the PAO1 and pqsA strains grown in absence or presence of PQS; 2'2-dipyridyl, a known iron chelator, was also tested as a control for iron depletion. As shown in Fig. 2, a significant (P
