Pseudomonas aeruginosa produces aspirin insensitive eicosanoids and contributes to the eicosanoid profile of polymicrobial biofilms with Candida albicans

Author’s Accepted Manuscript Pseudomonas aeruginosa produces aspirin insensitive eicosanoids and contributes to the eicosanoid profile of polymicrobial biofilms with Candida albicans Ruan Fourie, Ruan Ells, Gabré Kemp, Olihile M. Sebolai, Jacobus Albertyn, Carolina H. Pohl www.elsevier.com/locate/plefa

PII: DOI: Reference:

S0952-3278(16)30083-7 http://dx.doi.org/10.1016/j.plefa.2017.01.008 YPLEF1803

To appear in: Prostaglandins Leukotrienes and Essential Fatty Acids Received date: 13 June 2016 Revised date: 16 December 2016 Accepted date: 24 January 2017 Cite this article as: Ruan Fourie, Ruan Ells, Gabré Kemp, Olihile M. Sebolai, Jacobus Albertyn and Carolina H. Pohl, Pseudomonas aeruginosa produces aspirin insensitive eicosanoids and contributes to the eicosanoid profile of polymicrobial biofilms with Candida albicans, Prostaglandins Leukotrienes and Essential Fatty Acids, http://dx.doi.org/10.1016/j.plefa.2017.01.008 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 galley proof before it is published in its final citable 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.

Pseudomonas aeruginosa produces aspirin insensitive eicosanoids and contributes to the eicosanoid profile of polymicrobial biofilms with Candida albicans

Ruan Fouriea, Ruan Ellsa,b, Gabré Kempa, Olihile M. Sebolaia, Jacobus Albertyna, Carolina H. Pohla* a

Department of Microbial, Biochemical and Food Biotechnology, University of the

Free State, Bloemfontein, South Africa b

National Control Laboratory for Biological Products, University of the Free State,

Bloemfontein, South Africa *

Correspondence: Department of Microbial, Biochemical and Food Biotechnology,

University of the Free State, Bloemfontein, South Africa. [email protected] (CP)

ABSTRACT The interaction of clinically relevant microorganisms is the focus of various studies, e.g. the interaction between the pathogenic yeast, Candida albicans, and the bacterium, Pseudomonas aeruginosa. During infection both release arachidonic acid, which they can transform into eicosanoids. This study evaluated the production of prostaglandin E2, prostaglandin F2α and 15-hydroxyeicosatetraenoic acid by biofilms of P. aeruginosa and C. albicans. The influence of co-incubation, acetylsalicylic acid and nordihydroguaiaretic acid on biofilm formation and eicosanoid production was evaluated. Acetylsalicylic acid decreased colony forming units of P. aeruginosa, but increased metabolic activity and eicosanoid production of the cells. In contrast to prostaglandin E2, prostaglandin F2a production by C. albicans was insensitive to acetylsalicylic acid, indicating that different enzymes are responsible for their production in this yeast. Nordihydroguaiaretic acid inhibited biofilm formation by P. aeruginosa, however co-incubation provided protection against this inhibitor. Production of these eicosanoids could affect pathogen-clearance and infection dynamics and this previously uncharacterized facet of interaction could facilitate novel therapeutic intervention against polymicrobial infection.

1

Abbreviations AA, arachidonic acid; ASA, acetylsalicylic acid; C. albicans, Candida albicans; CFU, colony forming unit; COX, cyclooxygenase; CYP450, cytochrome P450 monooxygenase; ELISA, enzyme-linked immunosorbent assay; EPS, exopolysaccharide; ExoU, exotoxin U; NDGA, nordihydroguaiaretic acid; PBS, phosphate buffered saline; PGE2, prostaglandin E2; PGF2α, prostaglandin F2α; P. aeruginosa, Pseudomonas aeruginosa; S. aureus, Staphylococcus aureus; YM agar, yeast malt agar; YNB, yeast nitrogen base; XTT, 2,3-bis (2 methoxy-4-nitro-5sulfophenyl)-5[(phenylamino)carbonyl]-2H tetrazolium hydroxide; 15-HETE, 15hydroxyeicosatetraenoic acid Keywords Acetylsalicylic acid, biofilm, Candida albicans, eicosanoid, nordihydroguaiaretic acid, Pseudomonas aeruginosa.

1. INTRODUCTION Candida albicans is an opportunistic pathogen frequently isolated from healthy individuals as part of the normal commensal microbiota [1]. During compromised immunity, C. albicans causes superficial to systemic infections. Yeast, pseudohyphae and true hyphae are frequently found at infection sites. In addition to this, C. albicans is able to form biofilms on biotic (i.e. epithelial surfaces) as well as abiotic surfaces (i.e. catheters and other indwelling devices) [2–5]. This opportunistic pathogen is rarely found alone and is the subject of study for interkingdom interactions with bacteria, such as Staphylococcus aureus and Pseudomonas aeruginosa [6,7]. The latter is a Gram-negative opportunistic pathogen, frequently infecting immunocompromised individuals [8,9]. Pseudomonas aeruginosa also possesses the ability to form antibiotic resistant biofilms [10,11]. A large amount of literature is available describing the antagonistic interaction between C. albicans and P. aeruginosa in vitro [1,7,12–15]. During close proximity, such as during the formation of polymicrobial biofilms, P. aeruginosa has been shown to kill C. albicans

2

hyphae through the involvement of various virulence factors, including the production of a phenazine pigment, pyocyanin [16,17]. The interaction is generally described as antagonistic with the dynamic shifted towards fungal growth inhibition [18–23]. During infection, C. albicans and P. aeruginosa elicit the release of arachidonic acid (AA) from host cells [24,25]. Arachidonic acid is the substrate for the production of various eicosanoids, with crucial involvement in the innate and adaptive immunity of hosts [26]. During C. albicans infection, AA liberation is induced by cell wall components as well as phospholipases produced by the yeast [24,27]. This leads to the increased production of eicosanoids, especially prostaglandin E2 (PGE2), by the host. The production of prostaglandins in mammalian systems, is due to the action of cyclooxygenase (COX) enzymes with additional modification by synthases [28]. Prostaglandin E2 can inhibit Th1- and promote Th2-type responses in the host, ultimately affecting the ability of the host to clear C. albicans infection [27,29]. Interestingly, C. albicans is also able to produce a significant amount of PGE2 from exogenous/host derived AA [30,31]. Curiously, this PGE2 production by C. albicans is inhibited by the COX inhibitor, acetylsalicylic acid (ASA), although C. albicans does not possess COX homologs [32]. However, other enzymes have been shown to possibly play a role in PGE2 production by C. albicans, namely, a multicopper oxidase (Fet3p), a fatty acid desaturase (Ole2p), as well as monooxygenases (CYP450s) [30,33,34]. Prostaglandin E2 may also play an important role in interkingdom interactions between C. albicans and bacteria, as PGE2 significantly increased S. aureus biofilm growth [34]. Pseudomonas aeruginosa elicits the release of large amounts of AA through exotoxin U (ExoU) and the subsequent production of PGE2 by the host [25]. Pyocyanin, 3-oxo-homoserine lactone and lipopolysaccharides also induce the production of PGE2 by the host [35,36]. Pseudomonas aeruginosa possesses a secretable 15-lipoxygenase capable of converting AA to 15-hydroxyeicosatetraenoic acid (15-HETE), which has a dramatic anti-inflammatory effect on the host [37,38]. In addition, P. aeruginosa also has the ability to produce prostaglandins and prostaglandin-like molecules with potential effects on host cells during infection [39]. Although the production of eicosanoids by individual pathogens has been the subject of various studies, the metabolism of AA and production of eicosanoids in

3

polymicrobial interaction has not been addressed. Therefore, the aim of this study is to determine how AA is metabolized by C. albicans and P. aeruginosa polymicrobial biofilms compared to the monomicrobial counterparts in vitro. In addition, the effect of inhibitors, that have been shown to affect PGE2 production by C. albicans, is assessed for their ability to influence eicosanoid production by P aeruginosa as well as polymicrobial biofilms. This will aid in the understanding of possible mechanisms involved in eicosanoid production by these pathogens.

2. MATERIALS AND METHODS 2.1. Strains used Candida albicans strain CBS 8758 (SC5314) was maintained on yeast malt (YM) agar (3 g/L malt extract, 3 g/L yeast extract, 5 g/L peptone, 10 g/L glucose, 16 g/L agar) at 30 °C. Pseudomonas aeruginosa strain PAO1, provided by Professor Hancock from the Department of Microbiology and Immunology at the University of British Columbia, was revived according to American Type Culture Collection method from lyophilized strain ampoules, and maintained on nutrient agar (1 g/L malt extract, 2 g/L yeast extract, 5 g/L peptone, 8 g/L sodium chloride and 20 g/L agar) at 37 °C. 2.2. Formation of mono- and polymicrobial biofilms 2.2.1. Monomicrobial biofilm formation by Candida albicans Candida albicans was grown on YM agar for 24 h at 30 °C and was inoculated into 10 mL Yeast Nitrogen Base (YNB) broth (10 g/L glucose, 16 g/L YNB) and incubated at 30 °C for 24 h. Cells were harvested at 1878 g for 5 minutes and the supernatant removed. This was followed by washing the cells twice with phosphate buffered saline (PBS) (Sigma-Aldrich, USA). The cells were then counted with a haemocytometer and diluted to 1 x 10 6 cells/ml in 20 mL filter sterilized (0.22 μm nitrocellulose filter, Merck Millipore, Ireland) RPMI1640 medium (Sigma-Aldrich, USA) and dispensed into 90 mm polystyrene petri dishes (Merck, Germany). In addition to the cells, 500 μM AA (Sigma-Aldrich, USA) (Stock of 1 g in 25 mL of ethanol reaching a concentration of 0.1314 M) was added to each petri dish containing medium and cells [33]. Petri dishes were incubated for 4

48 h at 37 °C to allow biofilm formation [40]. The final ethanol concentration in the biofilms before incubation was 0.38 %. 2.2.2. Monomicrobial biofilm formation by Pseudomonas aeruginosa Pseudomonas aeruginosa was grown on nutrient agar for 24 h at 37 °C and was inoculated into 5 mL nutrient broth and incubated at 37 °C for 24 h with shaking (150 rpm). Cells were diluted to an optical density (OD600) of approximately 0.005 in 20 mL filter sterilized RPMI-1640 medium and dispensed into 90 mm polystyrene petri dishes. Arachidonic acid (500 μM) was added to each petri dish and they were incubated for 48 h at 37 °C to allow biofilm formation. The final ethanol concentration in the biofilms before incubation was 0.38 %. 2.2.3. Polymicrobial biofilm formation The formation of polymicrobial biofilms follows the combination of the protocols for monomicrobial biofilm formation of both C. albicans and P. aeruginosa. Briefly, C. albicans and P. aeruginosa were grown as described above, and diluted to 20 mL RPMI-1640 medium. The medium thus contains 1 x 106 cells/mL C. albicans cells, as well as P. aeruginosa cells (approximately 0.005 OD600). In addition to the cells in the medium, 500 μM AA was added to each petri dish. Petri dishes were incubated for 48 h at 37 °C to allow biofilm formation. The final ethanol concentration in the biofilms before incubation was 0.38 %. 2.3. Effect of inhibitors Prior to biofilm formation, 100 µM of the inhibitors, acetylsalicylic acid (SigmaAldrich, USA) (dissolved in ethanol, final ethanol concentration in medium 0.45 %) and nordihydroguaiaretic acid (Sigma-Aldrich, USA) (dissolved in ethanol, final ethanol concentration in medium 0.30 %) were added together with 500 µM AA. Biofilms were incubated for 48 h at 37 °C [33]. 2.4. Determination of metabolic activity Mono- and polymicrobial biofilms were prepared as described above in a 96-well plate (Corning Incorporated, USA) with the volume of medium adjusted to 100 μL. The plate was incubated for 48 h at 37 °C to allow the formation of biofilms. Following incubation, the supernatant from each well was removed and the biofilms were washed twice with sterile PBS. The XTT assay was performed according to 5

Kuhn et al. [41]. Briefly, 50 μL of 2,3-bis (2-methoxy-4-nitro-5-sulfophenyl)5[(phenylamino) carbonyl]-2H tetrazolium hydroxide (XTT) (Sigma-Aldrich, USA) (1 g XTT in 1 L PBS, filter sterilized, aliquoted and stored at -20 °C) containing 0.08 mM menadione (Sigma-Aldrich, USA) (stock solution of 10 mM menadione in acetone) was added to each well and incubated for 3 h in the dark at 37 °C. Following incubation, the optical density of each well was measured at 492 nm on a Labotec Spectramax M2 microplate reader (Molecular devices). Appropriate controls including medium containing only AA were included. The same assay was also performed on biofilms grown in 90 mm polystyrene petri dishes. These experiments were performed in triplicate. 2.5. Determination of dry biomass Biofilms were prepared as described above in 90 mm polystyrene petri dishes. After incubation, the biofilms were scraped off and washed twice with sterile PBS, where after the resuspended cells were filtered through pre-weighed filters (0.22 μm nitrocellulose). The filters were dried at 37 °C overnight and the dry biomass of the mono- and polymicrobial biofilms determined [42]. This experiment was performed in triplicate. 2.6. Determination of colony forming units (CFUs) Mono- and polymicrobial biofilms were prepared as described above. After incubation, the biofilms were scraped off and washed twice with sterile PBS. To disrupt biofilms and to remove adherent cells from one another, cells were vortexed three times for 1 minute. For monomicrobial biofilms of C. albicans as well as polymicrobial biofilms, serially diluted cells were plated on YM medium acidified with tartaric acid (final concentration 0.08 %) to inhibit bacterial growth. For monomicrobial biofilms of P. aeruginosa as well as polymicrobial biofilms, cells were serially diluted and plated onto Luria-Bertani plates (5 g/L yeast extract, 10 g/L tryptone, 10 g/L NaCl and 15 g/L agar) supplemented with 10 μg/mL amphotericin B (Sigma-Aldrich, USA) to inhibit fungal growth [43]. Plates were incubated overnight at 37 °C and CFUs were counted. This experiment was performed in triplicate. 2.7. Quantification of exopolysaccharide material Mono- and polymicrobial biofilms were prepared as described above in 90 mm polystyrene petri dishes and incubated for 48 h. Biofilms were scraped off and 6

collected by centrifugation (5750 g for 10 minutes at 4 °C, Eppendorf). The biomass was washed twice with sterile PBS and polysaccharides were quantified according to Kannan & Gautum [44]. Briefly, biofilms were resuspended in 1 mL of 30 % (w/v) sodium hydroxide [44,45] and samples were boiled for 15 min. Cells were removed by centrifugation (5750 g for 15 minutes) and polysaccharides were precipitated with 60 % ethanol (10 min, 40 °C). The precipitate was redissolved in 2 mL sterile water, after which 8 volumes of sulphuric acid (98 %, v/v) was used to hydrolyse polysaccharides. One volume of cold tryptophan (1 % w/v) was added to mixture and boiled for 15 min. The absorbance (500 nm) was read for each sample (Jenway 6400 Spectrophotometer, Keison products). 2.8. Morphology of polymicrobial biofilms Polymicrobial biofilms were prepared as described above in flat bottom 6 well culture plates (Corning Incorporated, USA) in 3 mL medium with the addition of inhibitors. After incubation, the supernatants were removed and 5 mm rectangular sections of the bottom of the wells were cut out and washed with PBS and fixed overnight with the primary fixative, 3 % (v/v) gluterdialdehyde (Merck, Germany) in phosphate buffer (pH 7.0) [46]. The cells were then washed with PBS and fixed with the secondary fixative, 1 % (v/v) osmium tetroxide (Merck, Germany) for 1 h at room temperature. The biofilms were then sequentially dehydrated with 50 %, 70 % and 95 % ethanol for 15 min each and twice for 1 h with 100 % ethanol. After critical point drying (Samdri-795 Critical point dryer, Tousimis, USA), the biofilms were sputter coated with gold using a SEM coating system (Bio-Rad Microscience division, UK) and examined using a Shimadzu SSX-550 Superscan scanning electron microscope. 2.9. Quantification of eicosanoid production Mono- and polymicrobial biofilms were prepared in 90 mm polystyrene petri dishes as described before. Supernatant and cells were collected in 50 mL Falcon tubes (Corning Incorporated, USA) and supernatants recovered by centrifugation (5750 g for 5 minutes at 4 °C). After centrifugation the supernatants were filtered (0.2 μm nitrocellulose filter). Extraction of eicosanoids were performed according to a modified protocol proposed by Cayman chemicals for PGE2 purification for enzyme linked immunosorbent assay (ELISA). Briefly, supernatants were acidified to a pH of 7

approximately 4 with the addition of 1 M formic acid. Solid phase extraction (SPE) classic C18 cartridges (Waters, South Africa) were prepared with 5 mL methanol (Merck, Germany), followed by 5 mL deionized water. Samples (10 mL) were applied to cartridges and subsequently washed with 5 mL deionized water to remove impurities. Eicosanoids were then eluted from the SPE cartridges with 5 mL ethyl acetate containing 1 % methanol and collected in pre-washed poly top glass vials. The eluent was dried under a stream of N2 and stored at -80 °C until use. Samples were dissolved in eicosanoid affinity (EIA) buffer provided by the manufacturer and samples were assayed for PGE2 (Cayman Chemicals, USA), PGF2α (Cayman Chemicals, USA) and 15-HETE (Cayman Chemicals, USA) using ELISA according to manufacturer’s specifications. Samples were assayed in two dilutions in duplicate. Values obtained from control biofilms (consisting of only AA without cells) were subtracted from experimental samples to remove values of nonenzymatic oxidation of AA. This experiment was done in triplicate. Data was analysed according to manufacturer’s specifications.

2.10. Identification of P. aeruginosa prostaglandin E2 by LC-MS/MS In order to confirm authentic PGE2 production, intracellular PGE2 extraction was used to limit the detection of AA autoxidation products. Single species biofilms of P. aeruginosa with AA were prepared in the presence and absence of ASA and NDGA as described with the exception that these biofilms were incubated for 24 h [47]. Biofilms were scraped off and washed three times with PBS. The biomass of three biofilms were pooled to obtain sufficient biomass. Washed biofilms were vortexed twice for 1 min to disperse biofilms and centrifuged (5750 g for 5 min at 4 °C, Eppendorf) to collect cells. The cells were resuspended in 1 mM ethylenediaminetetraacetic acid (EDTA) with 1 N citric acid (160 μL) and 20 μL (10 %) butylated hydroxytoluene (BHT, Sigma-Aldrich) as antioxidant. Samples were sonicated according to a modified protocol described for P. aeruginosa biofilms [48]. Briefly, cell suspensions were sonicated at 60 % power (Bandalin Sonopuls, Germany) for only 30 seconds, to limit thermal degradation of eicosanoids, and incubated on ice for 30 seconds. This was repeated for 10 cycles. After breaking the cells, eicosanoids were extracted with 2 mL hexane:ethyl acetate (1:1) [47]. After 8

addition of hexane:ethyl acetate, samples were vortexed for 1 minute and centrifuged (1878 g for 10 minutes at 4 °C) to remove cellular debris after which the organic phase was removed. This was done 3 times and the organic phase was pooled and evaporated under a stream of N2 in pre-washed glass vials. Samples were stored at -80 °C until use. Prior to LC-MS/MS analysis, samples were divided into two aliquots. One aliquot was spiked with 1 ng of PGE2 (Sigma-Aldrich) to determine retention time of authentic PGE2 in samples. Each sample was separated on a C18 (Luna 3µm C18 (2), 150 x 3 mm, Phenomenex) column at a flow rate of 200 µL/min using 0.1 % formic acid (mobile phase A) and acetonitrile with 0.1 % formic acid (mobile phase B). The column was equilibrated and loaded at 20 % B, increasing to 42.5 % B over 50 minutes, 95 % B for 10 minutes, followed by re-equilibration at 20 % B for a total run time of 70 minutes [49]. The targeted analysis for the extracted PGE2 used 5 MRM transitions: 351.17 > 315.2; 351.17 > 271.2; 351.17 > 333.3; 351.17 > 189.0; 351.17 > 235.1. Only if all 5 transitions were recorded at the same retention time would the presence of PGE2 be confirmed. 2.11. Statistical analysis To evaluate significant differences between mono- and polymicrobial biofilms, unpaired t-test was used (P < 0.05). In addition, the t-test was used to determine significant differences between control biofilms and biofilms treated with inhibitors. Significant differences are indicated with ‘ * ’.

3. RESULTS AND DISCUSSION 3.1. Effect of co-incubation on biofilm formation and eicosanoid production In order to discuss the production of eicosanoids by mono- and polymicrobial biofilms, it is first necessary to evaluate the effect of co-incubation on biofilm growth and metabolic activity in the presence of exogenously added AA. Metabolic activity, as measured by the XTT assay, of monomicrobial C. albicans biofilms and polymicrobial biofilms do not differ significantly (Fig 1 A), however a reduction in dry biomass is observed in polymicrobial biofilms compared to C. albicans monomicrobial biofilms (Fig 1 B). In contrast, the metabolic activity and dry biomass

9

of P. aeruginosa monomicrobial biofilms is significantly lower compared to polymicrobial biofilms (Fig 1 A). It has to be noted that exopolysaccharide (EPS), produced by biofilms, may contribute significantly to the dry biomass (Fig 1 D). In addition, although the use of XTT in the determination of fungal and bacterial biofilm metabolic activity has been proven to be useful, previous research indicates that additional electron carriers (in addition to menadione used in this study), such as phenazine methosulphate, may facilitate the assay for metabolic activity by P. aeruginosa biofilms and give higher values [41,50]. Due to the possibility that the smaller culture size may influence the results of the XTT assay, this assay was repeated for biofilms in petri dishes. The results obtained were similar to the XTT assay in 96 well plates (Figure S1). To circumvent possible irregularities arising from normalizing eicosanoid production according to biomass or metabolic activity, colony forming units (CFUs) were determined for respective biofilms as previously described [51]. Fig 1 C indicates the effect of co-incubation of C. albicans and P. aeruginosa on the respective CFUs. A significant reduction in C. albicans CFUs is seen in the polymicrobial biofilm compared to the monomicrobial counterpart. This is expected given the known inhibitory effect of P. aeruginosa on C. albicans [1,12,16,17]. Interestingly, a significant increase in P. aeruginosa CFUs is seen in the polymicrobial biofilm compared to the monomicrobial counterpart. This resulted in a dramatic increase in the total number of CFUs in the polymicrobial biofilm compared to either monomicrobial biofilm. This increase in P. aeruginosa cell numbers may be due to ethanol production by C. albicans, previously indicated to increase P. aeruginosa biofilm growth [52]. Fig 2 represents eicosanoid concentration (pg/mL) obtained with the use of ELISA specific for PGE2, prostaglandin F2α (PGF2α) and 15-HETE, after subtraction of values obtained for AA autoxidation controls. Our results confirm previous research indicating the production of PGE2 by C. albicans biofilms (Fig 2 A) as well as the presence of PGF2α (Fig 2 C) and significant quantities of 15-HETE (Fig 2 E) from exogenously added AA [30,31]. Pseudomonas aeruginosa is also capable of producing significant quantities of all three these eicosanoids from AA. Although P. aeruginosa has previously been shown to produce 15-HETE [38] as well as prostaglandins and prostaglandin-like molecules [39], the production of PGE2 and 10

PGF2α specifically has not been reported previously. Interestingly, co-incubation of C. albicans and P. aeruginosa caused a significant increase in the concentration of PGE2, PGF2α and 15-HETE in the supernatant compared to monomicrobial biofilms (Fig 2 A, C, E). However, since these values only represent the total production of eicosanoids per volume, this increase may be due to the observed increased number of CFUs in the polymicrobial biofilm. In order to evaluate the contribution to eicosanoid production per CFU, the eicosanoid concentrations were normalized against CFU count for both mono- and polymicrobial biofilms (Fig 2 B, D, F). From this data it can be seen that the ability of C. albicans CFUs to produce these eicosanoids exceed that of P. aeruginosa CFUs. A significant reduction in PGF2 and 15-HETE production by individual CFUs in polymicrobial biofilms is seen, compared to monomicrobial biofilms consisting of either C. albicans or P. aeruginosa. This may be partly due to the lower C. albicans CFUs in the polymicrobial biofilm, but may also indicate that the ability of P. aeruginosa to produce these two eicosanoids in the presence of C. albicans is slightly inhibited. In the case of PGE2, a reduction was only seen compared to monomicrobial biofilms of C. albicans. Possibly indicating that most of the PGE2 produced by polymicrobial biofilms is due to P. aeruginosa. The high concentration of PGE2 produced by polymicrobial biofilms, raises questions of the effect on the host, as inhibition of COX-2 and thus PGE2 production in the host, leads to increased clearance of bacterial cells [53,54]. In addition, both organisms produce significant quantities of 15-HETE, which can be converted to anti-inflammatory lipoxins in the host. The combination of these eicosanoids produced by both species, may ultimately decrease the host response during coinfection. 3.2. Effect of inhibitors of PGE2 production on biofilm formation Although eicosanoid metabolism in mammalian systems has been well studied, the same cannot be said for the production of eicosanoids by microorganisms. Previous studies have indicated that several inhibitors decrease the production of PGE 2 by C. albicans [30,33,34]. These include ASA and nordihydroguaiaretic acid (NDGA). These inhibitors are responsible for the inhibition of various enzyme groups. Acetylsalicylic acid acetylates COX enzymes in mammalian systems, inhibiting the production of prostaglandins and thromboxanes [55,56]. However, the mechanism of inhibition of C. albicans PGE2 synthesis is unknown. Nordihydroguaiaretic acid has a 11

broad inhibitory effect, affecting lipoxygenase, multicopper oxidase and monooxygenase activity [57]. This is due to the potent antioxidant activity of NDGA. The effect of these inhibitors on P. aeruginosa eicosanoid production has not yet been studied. Fig 3 indicates the effect of ASA and NDGA on metabolic activity (Fig 3 A), dry biomass (Fig 3 B), EPS (Fig 3 C) as well as CFU counts (Fig 3 D) of C. albicans and P. aeruginosa mono- and polymicrobial biofilms compared to control biofilms without inhibitors. As can be seen, ASA and NDGA did not influence the metabolic activity, biomass production or CFU count of C. albicans monomicrobial biofilms, however, a reduction in EPS was seen in the presence of NDGA. Interestingly, ASA resulted in an increase in metabolic activity, but decreased the CFU count of P. aeruginosa in the monomicrobial biofilms, although none of these changes are reflected in the biomass or EPS. Notably, ASA was shown to reduce extracellular material production by P. aeruginosa at higher concentrations [58]. Previous research also indicates that ASA inhibits quorum sensing, virulence factors and biofilm formation in P. aeruginosa [58]. However, this study utilized much higher concentrations of ASA (>50 fold) than the 100 μM used in our study. It is also known that salicylic compounds inhibits ATP synthesis and is an electron uncoupler [59]. A similar effect may occur in P. aeruginosa, where the bacterial cells compensate for reduced ATP synthesis, due to the disruption of electron flow through the electron transport chain, by increasing metabolic activity. Therefore, although less cells may be present (as ATP synthesis is inhibited), they appear more metabolically active due to overcompensation. Similar results were also seen for salicylic acid [60]. This finding should be further investigated for possible clinical significance. Nordihydroguaiaretic acid lead to an almost complete inhibition of metabolic activity of P. aeruginosa monomicrobial biofilm and a significant reduction in CFU count. However, similar to ASA these changes were not significantly reflected in the dry biomass or EPS. This may indicate that both of these inhibitors have as yet unknown effects on other unquantified compounds in the extracellular matrix, that contribute to the biomass. The polymicrobial biofilms grown in the presence of ASA or NDGA did not significant differ in metabolic activity (Fig 3A) or biomass (Fig 3B) compared to the control, although a slight decrease in EPS was observed in the presence of ASA (Fig 3C). In addition, although a decrease in CFU count of P. aeruginosa in these biofilms were 12

seen, the decrease was not statistically significant (P > 0.05). A consequence of NDGA on polymicrobial biofilms was the observed increase in EPS as well as C. albicans CFUs. In order to determine the influence of possible changes in morphology due to the presence of ASA or NDGA on the results, the effect of inhibitors on the morphology of polymicrobial biofilms were evaluated by scanning electron microscopy (Fig 4). The control polymicrobial biofilm grown only in the presence of AA, consisted of C. albicans cells, predominantly in the yeast morphology with a small number of hyphae. In addition, P. aeruginosa cells can be observed entangled in extracellular matrix (Fig 4A). Thus, AA did not noticeably influence the interaction between C. albicans and P. aeruginosa in terms of morphology of the biofilm, as this morphology is similar to polymicrobial biofilms not exposed to AA [58]. The inhibition of filamentation, usually observed in C. albicans single species biofilms, is as expected due to the presence of P. aeruginosa. Interestingly, this inhibition of filamentation of C. albicans is abolished in the biofilm grown in the presence of AA and ASA, with C. albicans forming hyphae even in the presence of large numbers of P. aeruginosa cells (Fig 4 B). This may indicate that the P. aeruginosa inhibitory factors, such as the quorum sensing molecule 3-oxohomoserine lactone and pyocyanin, are inhibited by ASA [56]. Although this increase in hyphae may cause a decrease in CFUs compared to the control, this was not observed for C. albicans in polymicrobial biofilms exposed to ASA (Fig 3 D). Interestingly, a decrease in extracellular material is observed by P. aeruginosa in the presence of ASA, confirmed by quantification of EPS production (Fig 3 C). A significant influence on C. albicans morphology can be seen for biofilms grown in the presence of NDGA, with no hyphae observed (Fig 4C). This effect of NDGA on C. albicans morphology has not previously been reported and could explain the increased CFUs of C. albicans observed in these polymicrobial biofilms (Fig 3 D). In addition, very little P. aeruginosa cells are observable in the presence of NDGA. Interestingly, a large amount of extracellular material is observed. This increase in extracellular material compared to control biofilms was confirmed by quantification of EPS (Fig 3 C). The low metabolic activity may correlate with the low P. aeruginosa cell numbers observed, with the bulk of the dry biomass possibly reflecting extracellular material, rather than metabolically active cells.

13

3.3. Inhibitors of PGE2 production affect mono- and polymicrobial biofilm eicosanoid production 3.3.1. Effect of acetylsalicylic acid Fig 5 A indicates the change in concentration (% pg/mL) of PGE2, PGF2α and 15HETE produced by mono- and polymicrobial biofilms of C. albicans and P. aeruginosa in the presence of ASA. In addition to this, the change in eicosanoid production in the presence of ASA normalized against CFU (% pg/CFU) is indicated in Fig 5 B. A significant decrease in C. albicans PGE2 production by ASA has previously been reported [33]. In the present study, PGE2 production by C. albicans monomicrobial biofilms was also significantly inhibited. However, PGF2α concentration as well as PGF2α production normalized against CFUs was significantly increased. In mammalian systems, AA is converted to the unstable intermediate PGH2 by COX. This intermediate acts as substrate for synthases, to form PGE2 and PGF2α [28,61]. As ASA inhibits COX in mammalian systems, inhibition thereof would cause a decrease in PGF2α as well [61]. This phenomenon is not seen with C. albicans, indicating that PGF2α production is not dependent on the same enzyme as PGE2 production and is not ASA sensitive. It can also be speculated that the reduction in PGE2 production makes more AA available to be converted to PGF2. Interestingly, although it is known that ASA triggers increased production of 15-HETE in humans [62], ASA had no effect on 15-HETE production in C. albicans. This is the first report of the effect of ASA on production of eicosanoids other than PGE2 in C. albicans. Unexpectedly, a significant increase in PGE2, PGF2α and 15-HETE concentration is observed in P. aeruginosa monomicrobial biofilms in the presence of ASA. This is also evident when the production is normalized against CFUs. This non-ASA sensitive production of PGE2 was confirmed with LC-MS/MS (Fig 6), where PGE2 was verified with a PGE2-isomer separation method [49]. Samples were spiked with PGE2 to identify retention times and minor peak shifting occurred due to the increased total prostaglandin concentration after spiking. The increase in eicosanoid production might be due to the increased metabolic activity observed for P. aeruginosa monomicrobial biofilms during ASA treatment (Fig 3 A). In polymicrobial 14

biofilms, ASA also caused a significant increase in eicosanoid production (pg/CFU), similar to P. aeruginosa monomicrobial biofilms. This similarity between eicosanoid profiles may indicate the contribution of high numbers of bacterial cells in the polymicrobial biofilms exposed to ASA. 3.3.2. Effect of nordihydroguaiaretic acid Nordihydroguaiaretic acid, an inhibitor of multiple enzyme classes, including lipoxygenase, multicopper oxidase and CYP450, significantly decreased PGE 2 production by C. albicans monomicrobial biofilms (Fig 7), confirming previous observations [33]. In addition, this inhibitor completely inhibited 15-HETE production by C. albicans monomicrobial biofilms. Although the decrease in concentration of PGF2α in the media was not statistically significant (due to high standard deviations), normalization against CFU did reveal a significant reduction in PGF2 production by C. albicans CFUs. Similarly to C. albicans monomicrobial biofilms, the concentration of all three eicosanoids produced by P. aeruginosa monomicrobial biofilms was significantly lower (Fig 7A). However, the results normalized against CFUs indicate that this apparent inhibition is due to inhibition of the number of CFUs, and that NDGA actually increases the ability of P. aeruginosa cells to produce these eicosanoids (Fig 7B). Notably, the presence of authentic PGE2 could not be confirmed through LC/MS/MS, with PGE2 possibly falling below the detection limit of the instrument. This may be due to the fact that LC-MS/MS analysis was done on 24 hour biofilms of P. aeruginosa according to previously described methods [47], rather than the 48 hour biofilm supernatant utilized in ELISA. In the case of the polymicrobial biofilms, the eicosanoid profiles are a combination of those of the monomicrobial biofilms, with lower concentrations of PGE2 and PGF2 and no 15HETE (Fig 7A). 4. CONCLUSIONS The identification of immune-modulating compounds produced by microorganisms, previously thought to be restricted to animals, may lead to alternative therapeutic options for treating infections. Since it is known that many infections are polymicrobial in nature it is also important to study this aspect of the interaction between different pathogens. This was done by comparing eicosanoid production of monomicrobial biofilms with polymicrobial biofilms. The results of this study confirms 15

the production of various eicosanoids from exogenous AA by C. albicans. In addition, it was the first report of the ability of P. aeruginosa to produce significant amounts of PGE2 and PGF2. This study confirms previous observations that C. albicans PGE2 production is dependent on acetylsalicylic acid-sensitive enzymes. In addition, results obtained suggest that synthesis of PGE2 and PGF2α may be due to different enzyme groups and that PGF2 synthesis is not dependent on ASA-sensitive enzymes. Interestingly, the production of eicosanoids, including PGE 2, by P. aeruginosa is not sensitive to and may even be increased in the presence of low concentrations of ASA. It is known that co-incubation of C. albicans and P. aeruginosa results in a decrease in C. albicans and an increase in P. aeruginosa CFUs in biofilms. This was also observed in our study. Interestingly, co-incubation resulted in a significant increase in the concentration of all three eicosanoids studied. This increase, possibly due to increased cell numbers, could influence the capability of hosts to clear infection of polymicrobial biofilms composed of C. albicans and P. aeruginosa. The PGE2 produced by P. aeruginosa and C. albicans may initiate the Th2-type immune response, hampering clearance of pathogens from host infection sites. In addition, significant concentration of 15-HETE is produced by mono- and polymicrobial biofilms. This 15-HETE may act as substrate for the formation of anti-inflammatory lipoxins. The effect of large amounts of PGE2, PGF2α and 15-HETE in combination has not been studied, and may influence pathogen clearance. Although ASA did not have a significant effect on C. albicans biofilm formation, it increased the metabolic activity but decreased the CFU count of P. aeruginosa. This phenomenon can be explained by the fact that cells try to compensate for the lack of ATP production (due to uncoupling of electron flow), by increasing metabolic activity. This increase in metabolic activity could also be a possible explanation for the increased production of all three eicosanoids in the presence of ASA. This increase was also seen in the polymicrobial biofilms. Nordihydroguaiaretic acid, which also did not influence C. albicans biofilm formation (except EPS production), lead to a decrease in both metabolic activity and CFUs of P. aeruginosa. Interestingly, co-incubation seemed to protect P. aeruginosa from the harmful effects of NDGA and also resulted in an observed increase of C. albicans 16

CFUs in the presence of NDGA, however, this increase in CFU may be due to more cells in the yeast morphology. This inhibitor decreased the production of all three eicosanoids by C. albicans, with 15-HETE being the most sensitive. Although it also decreased the concentration of eicosanoids produced by P. aeruginosa, this inhibitory effect seems to be due to reduction in CFUs. The polymicrobial biofilm exhibited an eicosanoid profile which is a combination of the monomicrobial profiles.

FUNDING INFORMATION The financial assistance of the National Research Foundation (NRF) towards this research (grant number 93485) is hereby acknowledged. Opinions expressed and conclusions arrived at, are those of the authors and are not to be attributed to the NRF. Funding This work was supported by the South African National Research Foundation [grant number 93485]. Conflict of interest The authors declare that they have no conflict of interests.

REFERENCES [1]

F.F. Rinzan, Pseudomonas aeruginosa-Candida albicans interactions from ecological and molecular perspectives, 2009. http://scholarworks.gsu.edu/cgi/viewcontent.cgi?article=1057&context=biology _diss.

[2]

D. Andes, J. Nett, P. Oschel, Development and characterization of an in vivo central venous catheter Candida albicans biofilm model, Infect Immun 72 (2004) 6023–6031. doi:10.1128/IAI.72.10.6023.

[3]

I. Bruzual, P. Riggle, S. Hadley, C.A. Kumamoto, Biofilm formation by fluconazole-resistant Candida albicans strains is inhibited by fluconazole, J. Antimicrob. Chemother. 59 (2007) 441–450. doi:10.1093/jac/dkl521. 17

[4]

G.E. Pierce, Pseudomonas aeruginosa, Candida albicans, and device-related nosocomial infections: Implications, trends, and potential approaches for control, J. Ind. Microbiol. Biotechnol. 32 (2005) 309–318. doi:10.1007/s10295005-0225-2.

[5]

G. Ramage, K. Vandewalle, B.L. Wickes, J.L. López-Ribot, Characteristics of biofilm formation by Candida albicans, Rev. Iberoam. Micol. Organo La Asoc. Esp. Espec. En Micol. 18 (2001) 163–170. doi:200118163 [pii].

[6]

P.I. Diaz, L.D. Strausbaugh, A. Dongari-Bagtzoglou, Fungal-bacterial interactions and their relevance to oral health: linking the clinic and the bench, Front. Cell. Infect. Microbiol. 4 (2014) 101. doi:10.3389/fcimb.2014.00101.

[7]

A.K. Lindsay, D.A. Hogan, Candida albicans: Molecular interactions with Pseudomonas aeruginosa and Staphylococcus aureus, Fungal Biol. Rev. 28 (2014) 85–96. doi:10.1016/j.fbr.2014.10.002.

[8]

G.B. Pier, Pulmonary disease associated with Pseudomonas aeruginosa in cystic fibrosis: current status of the host-bacterium interaction, J. Infect. Dis. 151 (1985) 575–80. doi:10.1093/infdis/151.4.575.

[9]

M.W. Tan, S. Mahajan-Miklos, F.M. Ausubel, Killing of Caenorhabditis elegans by Pseudomonas aeruginosa used to model mammalian bacterial pathogenesis., Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 715–20. doi:10.1073/pnas.96.2.715.

[10] A. Bragonzi, M. Paroni, A. Nonis et al., Pseudomonas aeruginosa microevolution during cystic fibrosis lung infection establishes clones with adapted virulence, Am. J. Respir. Crit. Care Med. 180 (2009) 138–145. doi:10.1164/rccm.200812-1943OC. [11] E. Drenkard, Antimicrobial resistance of Pseudomonas aeruginosa biofilms, Microbes Infect. 5 (2003) 1213–1219. doi:10.1016/j.micinf.2003.08.009. [12] D.A. Hogan, R. Kolter, Pseudomonas-Candida interactions: An ecological role for virulence factors, Science 296 (2002) 2229–2232. doi:10.1126/science.1070784. [13] J.B. Méar, E. Kipnis, E. Faure et al., Candida albicans and Pseudomonas aeruginosa interactions: More than an opportunistic criminal association?,

18

Med. Mal. Infect. 43 (2013) 146–151. doi:10.1016/j.medmal.2013.02.005. [14] M.E. Shirtliff, B.M. Peters, M.A. Jabra-Rizk, Cross-kingdom interactions: Candida albicans and bacteria, FEMS Microbiol. Lett. 299 (2009) 1–8. doi:10.1111/j.1574-6968.2009.01668.x. [15] L. Xu, F. Wang, Y. Shen et al., Pseudomonas aeruginosa inhibits the growth of pathogenic fungi: In vitro and in vivo studies, Exp. Ther. Med. 7 (2014) 1516– 1520. doi:10.3892/etm.2014.1631. [16] A. Brand, J.D. Barnes, K.S. Mackenzie, F.C. Odds, N.A.R. Gow, Cell wall glycans and soluble factors determine the interactions between the hyphae of Candida albicans and Pseudomonas aeruginosa, FEMS Microbiol. Lett. 287 (2008) 48–55. doi:10.1111/j.1574-6968.2008.01301.x. [17] J.R. Kerr, G.W. Taylor, A. Rutman, N. Høiby, P.J. Cole, R. Wilson, Pseudomonas aeruginosa pyocyanin and 1-hydroxyphenazine inhibit fungal growth, J. Clin. Pathol. 52 (1999) 385–7. doi:10.1136/jcp.52.5.385. [18] H.M.H.N. Bandara, B.P. K Cheung, R.M. Watt, L.J. Jin, L.P. Samaranayake, Pseudomonas aeruginosa lipopolysaccharide inhibits Candida albicans hyphae formation and alters gene expression during biofilm development, Mol. Oral Microbiol. 28 (2013) 54–69. doi:10.1111/omi.12006. [19] C. Cugini, M.W. Calfee, J.M. Farrow, D.K. Morales, E.C. Pesci, D.A. Hogan, Farnesol, a common sesquiterpene, inhibits PQS production in Pseudomonas aeruginosa, Mol. Microbiol. 65 (2007) 896–906. doi:10.1111/j.13652958.2007.05840.x. [20] F.G. Purschke, E. Hiller, I. Trick, S. Rupp, Flexible survival strategies of Pseudomonas aeruginosa in biofilms result in increased fitness compared with Candida albicans, Mol. Cell. Proteomics. 11 (2012) 1652–69. doi:10.1074/mcp.M112.017673. [21] B. Sapaar, A. Nur, K. Hirota et al., Effects of extracellular DNA from Candida albicans and pneumonia-related pathogens on Candida biofilm formation and hyphal transformation, J. Appl. Microbiol. 116 (2014) 1531–1542. doi:10.1111/jam.12483. [22] A. Trejo-Hernández, A. Andrade-Domínguez, M. Hernández, S. Encarnación,

19

Interspecies competition triggers virulence and mutability in Candida albicansPseudomonas aeruginosa mixed biofilms, ISME J. 8 (2014) 1974–1988. doi:10.1038/ismej.2014.53. [23] X.L. Xu, R.T.H. Lee, H.M. Fang et al., Bacterial peptidoglycan triggers Candida albicans hyphal growth by directly activating the adenylyl cyclase Cyr1p, Cell Host Microbe 4 (2008) 28–39. doi:10.1016/j.chom.2008.05.014. [24] M. Castro, N. V. Ralston, T.I. Morgenthaler, M.S. Rohrbach, A.H. Limper, Candida albicans stimulates arachidonic acid liberation from alveolar macrophages through alpha-mannan and beta-glucan cell wall components, Infect. Immun. 62 (1994) 3138–3145. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=302938&tool=pmce ntrez&rendertype=abstract. [25] A.M. Saliba, D.O. Nascimento, M.C.A. Silva et al., Eicosanoid-mediated proinflammatory activity of Pseudomonas aeruginosa ExoU, Cell. Microbiol. 7 (2005) 1811–1822. doi:10.1111/j.1462-5822.2005.00635.x. [26] E.A. Dennis, P.C. Norris, Eicosanoid storm in infection and inflammation., Nat. Rev. Immunol. 15 (2015) 511–23. doi:10.1038/nri3859. [27] S.P. Smeekens, F.L. van de Veerdonk, J.W.M. van der Meer, B.J. Kullberg, L. A. B. Joosten, M.G. Netea, The Candida Th17 response is dependent on mannan- and beta-glucan-induced prostaglandin E2., Int. Immunol. 22 (2010) 889–895. doi:10.1093/intimm/dxq442. [28] V.C. Tam, Lipidomic profiling of bioactive lipids by mass spectrometry during microbial infections, Semin. Immunol. 25 (2013) 240–248. doi:10.1016/j.smim.2013.08.006. [29] L. Romani, Innate and adaptive immunity in Candida albicans infections and saprophytism, J. Leukoc. Biol. 68 (2000) 175–179. http://www.ncbi.nlm.nih.gov/pubmed/10947060. [30] J.R. Erb-Downward, M.C. Noverr, Characterization of prostaglandin E2 production by Candida albicans, Infect. Immun. 75 (2007) 3498–3505. doi:10.1128/IAI.00232-07. [31] M.C. Noverr, S.M. Phare, G.B. Toews, J. Michael, G.B. Huffnagle, M.J. Coffey,

20

Pathogenic yeasts Cryptococcus neoformans and Candida albicans produce immunomodulatory prostaglandins, Infect Immun. 69 (2001) 2957-2963. doi:10.1128/IAI.69.5.2957. [32] M.A.S. Alem, L.J. Douglas, Prostaglandin production during growth of Candida albicans biofilms, J. Med. Microbiol. 54 (2005) 1001–1005. doi:10.1099/jmm.0.46172-0. [33] R. Ells, J.L. Kock, J. Albertyn, G. Kemp, C.H. Pohl, Effect of inhibitors of arachidonic acid metabolism on prostaglandin E2 production by Candida albicans and Candida dubliniensis biofilms, Med Microbiol Immunol. 200 (2011) 23–28. doi:10.1007/s00430-010-0169-7 [doi]. [34] J. Krause, G. Geginat, I. Tammer, Prostaglandin E2 from Candida albicans stimulates the growth of Staphylococcus aureus in mixed biofilms, PLoS One. 10 (2015) e0135404. doi:10.1371/journal.pone.0135404. [35] C. McDermott, R. Chess-Williams, K.A. Mills et al., Alterations in acetylcholine, PGE2 and IL6 release from urothelial cells following treatment with pyocyanin and lipopolysaccharide, Toxicol. Vitr. 27 (2013) 1693–1698. doi:10.1016/j.tiv.2013.04.015. [36] R.S. Smith, S.G. Harris, R. Phipps, B. Iglewski, The Pseudomonas aeruginosa quorum sensing molecule N-(3-Oxododecanoyl ) homoserine lactone contributes to virulence and induces inflammation in vivo, J Bacteriol. 184 (2002) 1132-1139. doi:10.1128/JB.184.4.1132. [37] C.N. Serhan, Lipoxins and aspirin-triggered 15-epi-lipoxin biosynthesis: An update and role in anti-inflammation and pro-resolution, Prostaglandins Other Lipid Mediat. 68-69 (2002) 433–455. doi:10.1016/S0090-6980(02)00047-3. [38] R.E. Vance, S. Hong, K. Gronert, C.N. Serhan, J.J. Mekalanos, The opportunistic pathogen Pseudomonas aeruginosa carries a secretable arachidonate 15-lipoxygenase., Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 2135–2139. doi:10.1073/pnas.0307308101. [39] M. Lamacka, J. Sajbidor, The occurrence of prostaglandins and related compounds in lower organisms, Prostaglandins, Leukot. Essent. Fat. Acids. 52 (1995) 357–364. doi:10.1016/0952-3278(95)90062-4.

21

[40] G. Ramage, K. VandeWalle, B.L. Wickes, J.L. López-Ribot, Standardized method for in vitro antifungal susceptibility testing of Candida albicans biofilms, Antimicrob. Agents Chemother. 45 (2001) 2475–2479. doi:10.1128/AAC.45.9.2475. [41] D.M. Kuhn, M. Balkis, J. Chandra, P.K. Mukherjee, M.A. Ghannoum, Uses and limitations of the XTT assay in studies of Candida growth and metabolism, J. Clin. Microbiol. 41 (2003) 506–508. doi:10.1128/JCM.41.1.506. [42] V.S. Thibane, J.L.F. Kock, R. Ells, P.W.J. Van Wyk, C.H. Pohl, Effect of marine polyunsaturated fatty acids on biofilm formation of Candida albicans and Candida dubliniensis, Mar. Drugs. 8 (2010) 2597–2604. doi:10.3390/md8102597. [43] D.P. Pires, S. Silva, C. Almeida et al., Evaluation of the ability of C. albicans to form biofilm in the presence of phage-resistant phenotypes of P. aeruginosa, Biofouling. 29 (2013) 1169–80. doi:10.1080/08927014.2013.831842. [44] A. Kannan, P. Gautum, A quantitative study on the formation of Pseudomonas aeruginosa biofilm, SpringerPlus. 4 (2015). doi:10.1186/540064-015-1029-0. [45] A. Forde, G.F. Fitzgerald, Analysis of exopolysaccharide (EPS) production mediated by the bacteriophage adsorption blocking plasmid pCI658, isolated from Lactococcus lactis spp. cremoris HO2, Int. Dairy J. 9 (1999) 465-492. [46]

S.P. Hawser, L.J. Douglas, Biofilm formation by Candida species on the surface of catheter material in vitro. Infect Immun 62 (1994) 915-921.

[47] R. Ells, G. Kemp, J. Albertyn et al., Phenothiazine is a potent inhibitor of prostaglandin E2 production by Candida albicans biofilms, FEMS Yeast Res. 13 (2013) 849-855. doi:10.1111/1567-1364.12093. [48] V. Chaturvedi, A. Kumar, Optimization of protein extraction protocol for 2D PAGE analysis of biofilm forming strain of Pseudomonas aeruginosa, Rom Biotechnol Lett. 17 (2012) 7529-7532. [49] S.A. Brose, B.T. Thuen, M.Y. Golovko, LC/MS/MS method for analysis of E2 series prostaglandins and isoprostanes. J Lipid Res 52 (2011). doi:10.1194/jlr.DOI3441 [50] I. Koban, R. Matthes, N.-O. Hübner et al., XTT assay of ex vivo saliva biofilms 22

to test antimicrobial influences, GMS Krankenhaushygiene Interdiszip. 7 (2012) Doc06. doi:10.3205/dgkh000190. [51] B.M. Peters, R.M. Ward, H.S. Rane et al., Efficacy of ethanol against Candida albicans and Staphylococcus aureus polymicrobial biofilms, Antimicrob. Agents Chemother. 57 (2013) 74-82. doi:10.1128/AAC01599-12. [52] A.I. Chen, E.F. Dolben, C. Okegbe et al., Candida albicans ethanol stimulates Pseudomonas aeruginosa WspR-controlled biofilm formation as part of a cyclic relationship involving phenazines, PLoS Pathog. 10 (2014) e1004480. doi:10.1371/journal.ppat.1004480. [53] G.Y. Park, N. Hu, X. Wang, R.T. Sadikot, F.E. Yull, M. Joo, R. Stokes Peebles, T.S. Blackwell, J.W. Christman, Conditional regulation of cyclooxygenase-2 in tracheobronchial epithelial cells modulates pulmonary immunity, Clin. Exp. Immunol. 150 (2007) 245–254. doi:10.1111/j.1365-2249.2007.03478.x. [54] R.T. Sadikot, H. Zeng, A.C. Azim et al., Bacterial clearance of Pseudomonas aeruginosa is enhanced by the inhibition of COX-2, Eur. J. Immunol. 37 (2007) 1001–1009. doi:10.1002/eji.200636636. [55] C. Schneider, A.R. Brash, Stereospecificity of hydrogen abstraction in the conversion of arachidonic acid to 15R-HETE by aspirin-treated cyclooxygenase-2. Implications for the alignment of substrate in the active site, J. Biol. Chem. 275 (2000) 4743–4746. doi:10.1074/jbc.275.7.4743. [56] J.R. Vane, R.M. Botting, The mechanism of action of aspirin, Thromb. Res. 110 (2003) 255–258. doi:10.1016/S0049-3848(03)00379-7. [57] J.-M. Lü, J. Nurko, S.M. Weakley et al., Molecular mechanisms and clinical applications of nordihydroguaiaretic acid (NDGA) and its derivatives: an update, Med. Sci. Monit. 16 (2010) RA93–100. doi:10.1111/j.16006143.2008.02497.x.Plasma. [58] Á. Dergez, H. Füvesi, Á. Koós, P. Sebestyén, P. Kesseru, Inhibition of exopolysaccharide biopolymer and pyocyanin virulence factors produced by Pseudomonas aeruginosa 1604 by salicylic compounds, Period. Polytech. Chem. Eng. 58 (2014) 75–80. doi:10.3311/PPch.7207. [59] C. Norman, K.A. Howell, A.H. Millar, J.M. Whelan, D.A. Day, Salicylic acid is

23

an uncoupler and inhibitor of mitochondrial electron transport, Plant Physiol. 134 (2004) 492–501. doi:10.1104/pp.103.031039. [60] S.A. El-Mowafy, K.H. Abd El Galil, S.M. El-Messery, M.I. Shaaban, Aspirin is an efficient inhibitor of quorum sensing, virulence and toxins in Pseudomonas aeruginosa, Microb. Pathog. 74 (2014) 25–32. doi:10.1016/j.micpath.2014.07.008. [61] B. Cryer, M. Goldschmiedt, J.S. Redfern, M. Feldman, Comparison of salsalate and aspirin on mucosal injury and gastroduodenal mucosal prostaglandins, Gastroenterology. 99 (1990) 1616–1621. http://view.ncbi.nlm.nih.gov/pubmed/2227277. [62] P. Korosec, U. Tisler, N. Bajrovic, M. Silar, A. Mrhar, M. Kosnik, Acetylsalicylic acid-triggered 15-HETE generation by peripheral leukocytes for identifying ASA sensitivity, Respir. Med. 105 (2011) S81–S83. doi:10.1016/S09546111(11)70017-8.

LEGENDS TO FIGURES

Fig 1. Formation of mono- and polymicrobial biofilms in the presence of 500 μM arachidonic acid. Metabolic activity (A), dry biofilm biomass (B), colony forming units (C) and exopolysaccharide material (D) of mono- and polymicrobial biofilms consisting of Candida albicans (CA) and Pseudomonas aeruginosa (PAO1). * Significantly different from polymicrobial biofilms (P < 0.05).

Fig 2. Eicosanoid concentration in pg/mL as well as normalized against colony forming units (pg/CFU) produced by mono- and polymicrobial biofilms. Production of 24

(A, B) prostaglandin E2 (PGE2), (C, D) prostaglandin F2α (PGF2α) and (E,F) 15hydroxyeicosatetraenoic acid (15-HETE) by mono- and polymicrobial biofilms of Candida albicans (CA) and Pseudomonas aeruginosa (PAO1). * Significantly different from polymicrobial biofilms (P < 0.05).

Fig 3. Formation of mono- and polymicrobial biofilms in the presence of 500 μM arachidonic acid and inhibitors. Percentage metabolic activity (A), dry biofilm biomass (B), exopolysaccharide (EPS) material (C) and percentage colony forming units (CFU) (D) of mono- and polymicrobial biofilms consisting of Candida albicans (CA) and Pseudomonas aeruginosa (PAO1) of mono- and polymicrobial biofilms consisting of Candida albicans (CA) and Pseudomonas aeruginosa (PAO1) in the presence of 100 M acetylsalicylic acid (ASA) or 100 M nordihydroguaiaretic acid (NDGA) compared to control biofilms. “Co-incubation” identifies yeast or bacterial CFU in polymicrobial biofilms. * Significantly different from control biofilms (P < 0.05).

Fig 4. Scanning electron micrographs of polymicrobial biofilms in the presence of 500 μM arachidonic acid (AA) and 100 μM inhibitors. Polymicrobial biofilms consisting of Candida albicans and Pseudomonas aeruginosa in the presence of AA 25

only (A), AA and acetylsalicylic acid (B), AA and nordihydroguaiaretic acid (C). Scale bars represent 10 μm.

Fig 5. Eicosanoid production of mono- and polymicrobial biofilms in the presence 500 μM arachidonic acid and acetylsalicylic acid. Percentage pg/mL (A) and pg/CFU (B) production of prostaglandin E2 (PGE2), prostaglandin F2α (PGF2α) and 15hydroxyeicosatetraenoic acid (15-HETE) by mono- and polymicrobial biofilms of Candida albicans (CA) and Pseudomonas aeruginosa (PAO1) in the presence of 100 M acetylsalicylic acid compared to control biofilms. Striped bars indicate respective control biofilms. * Significantly different from control biofilms (P < 0.05).

Fig 6. Confirmation of intracellular prostaglandin E2 (PGE2) production by single species biofilms of Pseudomonas aeruginosa by LC-MS/MS. Pseudomonas aeruginosa with arachidonic acid (A, B), P. aeruginosa with arachidonic acid and acetylsalicylic acid (C, D) and P. aeruginosa with arachidonic acid and nordihydroguaiaretic acid (E, F). Profiles on left (A, C and E) are samples spiked with 1 ng PGE2 prior to analysis. Profiles indicate transitions (351.17/351.2; 351.17/271.2; 26

351.17/333.3; 351.17/189.0 and 351.1/235.1) of PGE2 in each sample. Corresponding colours of transitions are indicated on the bottom of the figure. Arrows indicate peaks/retention times corresponding to PGE2 in samples.

Fig 7. Eicosanoid production of mono- and polymicrobial biofilms in the presence of 500 μM arachidonic acid and nordihydroguaiaretic acid. Percentage pg/mL (A) and pg/CFU (B) production of prostaglandin E2 (PGE2), prostaglandin F2α (PGF2α) and 15-hydroxyeicosatetraenoic acid (15-HETE) by mono- and polymicrobial biofilms of Candida albicans (CA) and Pseudomonas aeruginosa (PAO1) in the presence of 100 M nordihydroguaiaretic acid compared to control biofilms. Striped bars indicate respective control biofilms. * Significantly different from control biofilms (P < 0.05).

27

Supplementry Figure Fig S1. Metabolic activity of mono- and polymicrobial biofilms in 90 mm polystyrene petri dishes in the presence of 500 μM arachidonic acid. Polymicrobial biofilms consist of Candida albicans (CA) and Pseudomonas aeruginosa (PAO1). * Significantly different from polymicrobial biofilms (P < 0.05).

HIGHLIGHTS Candida albicans and Pseudomonas aeruginosa can form polymicrobial biofilms and often co-infect the host Both can release and metabolise arachidonic acid Acetyl salicylic acid increases metabolic activity and eicosanoid production by P. aeruginosa Prostaglandin F2 production by C. albicans is also increased by acetyl salicylic acid, indicating that different enzymes may be responsible for PGF2 production and PGE2 production in this yeast.

28