In total, 24 non-repetitive strains of PA (from various specimens: tracheal aspirate 20.8% (5/24); urine 16.6% (4/24); wound 25% (6/24); sputum 25 % (6/24); blood 12.5% (3/24); were obtained from various specimens at the in-patient department at Tongji Hospital (Wuhan, China) between May and September 2012. The strains were identified by Gram staining, the oxidase test and a Vitek-2 automated microbial identification system (bioMérieux, Inc., Craponne, France), or by the API 20NE system (bioMérieux, Inc.). All PA strains produce pyocyanin pigment. Escherichia coli [23922; American Type Culture Collection (ATCC), Manassas, VA, USA], Klebsiella pneumoniae (ATCC 700603) and PA (ATCC 25923) were maintained as quality control strains. The five strains of candida were collected from clinical specimens (Candida albicans, Candida tropicalis, Candida glabrata and Candida krusei were from sputum, Candida parapsilosis from urine) and were identified by CHROMagar Candida plates (CHROMagarCompany, Paris, France) and API-20C AUX yeast-like fungus identification strips (bioMérieux, Inc.). CA (ATCC 90028) was preserved as the control strain. The isolated PA strains were stored at 4°C on nutrient agar slopes, while the Candida isolates were stored on Sabouraud dextrose agar (SDA; Oxoid Ltd., Basingstoke, UK) plates until required for study.

BALB/c mice were purchased and maintained separately at the Animal Facility of Tongji Medical College, Huazhong University of Science and Technology (Wuhan, China) under controlled conditions (specific pathogen free, 22°C, 55% humidity, and 12 h light/dark cycle). All experimental procedures on animals used in this study were performed under a protocol approved by the Institutional Animal Care and Use Committee at the Tongji Medical College. Thirty mice were purchased and were used in this experiment. Male mice (age, 8–10 weeks; weight, 25–30 g) were selected for the study. All experimental procedures on animals used in the study were performed under a protocol approved by the Institutional Animal Care and Use Committee of Tongji Medical College.

Fungi (400 μl; 1×10⁸ cells/ml) were spread on SDA plates (one plate was used for each fungi species) using a glass spreader and sterile filter paper disks were placed on the plates. Each PA strain was grown overnight at 37°C in Luria-Bertani (LB) media. Next, 3-μl specimens of the PA cultures (8×10⁹ CFU/ml) were spotted on the filter disks, which was followed by incubation at 30°C for up to 48 h. Cultures of same volume and concentration of Escherichia coli (ATCC 23922), Klebsiella pneumoniae (ATCC 700603), Pseudomonas aeruginosa (ATCC 25923) and sterile water were respectively spotted on the central filter disks as a control.

Cross-streaking was performed according to the method described by Kerr (10). A fresh 24-h plate culture of each PA strain (1×10⁸ CFU/ml) was prepared as an inoculum in 0.9% NaCl to be tested for antifungal activity. The inoculum (30 μl) was streaked diametrically at a width of 1 cm across the SDA and blood agar (BA). SDA was used since it enhances fungal growth and all the tested strains of Pseudomonas were shown to grow well on the substance. The plates were incubated at 30°C for 24 h and the macroscopic growth was then removed from the plate using a sterile glass slide.

Sterile filter paper disks (diameter, 5 cm) were soaked in chloroform and laid on a metal tray in a safety cabinet. Each plate was then placed face down, without the lid, on top of a chloroform-containing filter paper disk; the plates were left for 30 min in order to kill the microscopic remnants of the culture. The plates were then removed from the cabinet and traces of chloroform were eliminated by exposure to air for a few minutes. A fresh 24-h plate culture of each fungal strain was used to prepare an inoculum of 1×10⁶ CFU/ml. This fungal suspension was streaked onto the chloroform-treated medium at right angles to the line of the original inoculum and the plates were incubated for 24 h at 30°C. Each of the 24 PA strains was tested against each of the five fungal strains and total inhibition of fungal growth was recorded as (+), partial inhibition of fungal growth was recorded as (±) and no inhibition of fungal growth was recorded as (−).

Sterile eppendorf tubes (EPs) were filled with 1 ml LB and each aforementioned PA and fungal suspension (50 μl) was added. Each fungal suspension (50 μl) was also added as a control. The EPs were agitated at 100 rpm at 30°C for 48 h.

PA1206 and PA1215 strains, which exhibit a strong inhibitory effect on fungi, as well as the PA1201 and PA1222 strains, which present no inhibitory effect, were analyzed with SDS-PAGE. The four strains of PA were cultured in EPs containing 1 ml LB under conditions of 100 rpm at 30°C for 24 h (two replicates for each PA strain were prepared). The two replicates were boiled at 100°C for 10 min, then centrifuged at 14,500 × g for 1 min and the supernatant and sediment were extracted for SDS-PAGE. The steps of SDS-PAGE were performed according to Thermo Fisher Scientific Inc. (Waltham, MA, USA). The results were observed following bleaching.

A model of blood infection was applied to evaluate the anticandidal activity of PA in mice. PA and Candida species are often identified on skin and in the mucosa of healthy individuals. When the host defenses falter, PA and Candida initiate invasive growth that leads to severe diseases. BALB/c mice weighing 25–30 g were used in this study. All animals received humane care. The mice were randomly assigned to the following three groups. Group 1 was the total inhibition positive group (n=5+5), where five mice applied with PA1206 and five mice applied with PA1215 were tested against CA (ATCC 90028). Group two was the no inhibition group (n=5+5), where five mice applied with PA1201 and five mice applied with PA1222 were tested against CA (ATCC 90028). Finally, group three was the control group (n=10) and only CA (ATCC 90028) was applied. The bacterial and yeast suspensions (0.2 ml; 1×10⁸ CFU/ml) were injected into the caudal vein of the mice in groups one and two, while only the yeast suspension (0.2 ml; 1×10⁸ CFU/ml) was injected into the caudal vein of the mice in group three. At 24 h after the injections, blood samples were collected from the mice by the method of retro-orbital blood collection. The blood drops were incubated in BA and SDA at 30°C for 48 h, and the growth of the PA and CA strains was evaluated.

PA strains 1206, -15, -16, -17, -18, -19 and -20 demonstrated clear zones of inhibition, while strains 1203, -09 and -23 presented partial zones of inhibition that were not very large. The remaining strains and the control strain exhibited no inhibition. Diameters of the zones of inhibition were measured; (−) no zone of inhibition; (±) zone of inhibition: 7–10 mm; (+) zone of inhibition: >10 mm (Fig. 1 and Table I).

The cross-streak method results were consistent with those of the disk diffusion method (Fig. 2).

When the fungi were co-incubated with known inhibitory PA strains, a significant reduction in the production of fungal hyphae was observed. However, when the fungi were co-incubated with non-inhibitory PA strains, the production of fungal hyphae was similar to that of fungi that had been cultured alone in LB medium (Fig. 3).

The two replicates of each PA strain produced the same pattern. The supernatants produced few bands, but the sediments presented the same bands and a number of different bands. Almost all the strains had one band in common at ~35 kDa. However, the inhibitory PA strains (1206 and 1215) produced distinct strips at 38, 35, 27 and 24 kDa, while the non-inhibitory PA strains (1201 and 1222) presented no bands at those points according to the marker. These observations indicate that the PA1206 and PA1215 strains secreted a greater variety of proteins compared with the PA1201 and PA1222 strains. An association between these bands and the growth inhibition effect on pathogenic fungi may exist, however, this requires further study (Fig. 4).

The mouse model of blood infection with CA and PA revealed that in group 1, no yeast was recovered from the infected mice, despite PA having 100% detection. In group 2, PA and yeast were recovered on the plate, while in group 3, 100% yeast was recovered.