Saprolegniosis is a fungal disease that occurs in freshwater fish. Saprolegnia parasitica is known to be a major cause of saprolegniosis (Lone and Manohar, 2018; Anokhina et al., 2021). It is not only a problem for the fish farming industry, but also partly responsible for the decline in natural populations of salmonids and other freshwater fish (Neitzel et al., 2004; van West, 2006). Neitzel et al. (2004) have reported that S. parasitica is associated with pre-spawning mortality in wild salmon populations. Especially, during incubation of fish eggs, Saprolegnia spp. can affect damaged and unfertilized eggs. After they are gradually transferred to live eggs, they can decrease the hatchability of fertilized eggs and survival of fry (Anokhina et al., 2021). Therefore, it is very important to control Saprolegnia outbreaks during incubation of fish eggs.

Malachite green (MG) has been used to control Saprolegnia sp. infection in aquaculture for a long time (Máchova et al., 1996). However, its use in food fishes has been banned worldwide since 2002 due to its toxicological effects and possible carcinogenicity in mammals (Srivastava et al., 2004; Culp et al., 2006). Therefore, discovery and application of safe and appropriate alternatives are needed. Chemical products (Ali et al., 2014), bacterial isolates, UV irradiation (Heikkinen et al., 2016), and plant extracts (Emara et al., 2020; Velichkova and Sirakov, 2021) have been studied to find replacement or alternative of MG that is currently banned. Although hydrogen peroxide (H₂O₂) (Barnes et al., 1998) and formalin (37% formaldehyde) (Gieseker et al., 2006) have been reported to show efficacies of controlling Saprolegnia sp. infection, they have human health risks such as eye inflammation and respiratory diseases caused by smoke. Several potential alternatives of MG such as bronopol, sodium chloride, and iodine have been proposed (Jee et al., 2007; Jee and Lee, 2009; Abd El-Gawad et al., 2016). However, materials with effects comparable to MG are still unavailable (van West, 2006; Sudova et al., 2007). In addition, synthetic organic dyes such as MG can contaminate the aquatic environment and potentially harm aquatic biota and threaten health of animals or human at higher orders (Tkaczyk et al., 2020). Therefore, interest in the possibility of using biological or natural substances for disease control instead of using antibiotics or chemotherapeutic agents is growing in the field of fish and shellfish farming (Emara et al., 2020; González-Palacios et al., 2020).

Pseudomonads are widely distributed in soil, freshwater, and marine environments. They produce extracellular inhibitory substances such as phenazine compounds (Chin-A-Woeng et al., 2003; Saosoong et al., 2009), hydrogen cyanides (Dubuis et al., 2007), siderophore (González-Palacios et al., 2020), and so on. Pyocyanin, phenazine-1-carboxylic acid, hydrogen cyanide, and so on have been found to be antifungal compounds produced by pseudomonads (Chin-A-Woeng et al., 2003). Bly et al. (1997) have reported that Pseudomonas fluorescens exhibits antifungal activity against fish pathogenic Saprolegnia sp. whereas its inhibitory activity is not associated with its culture supernatant. Carbajal-González et al. (2011) have reported in vitro inhibitory activities of bacteria including P. fluorescens found in the skin of rainbow trout against S. parasitica.

Our previous study has reported that Pseudomonas aeruginosa MB1-3 has antibacterial activities against fish pathogenic vibrios (Byon and Kim, 2000; Lee et al., 2014). In the present study, we evaluated antifungal effects of MB1-3 and an ethyl acetate extract of its culture supernatant (MB-EA) against fish pathogenic fungus Saprolegnia sp. FP9001 and identified its active compound for the purpose of developing a new substance capable of controlling fungal infection of fertilized fish eggs.

Organisms and their culture conditions

Pseudomonas aeruginosa MB1-3 strain (hereinafter MB1-3) was isolated from seawater and identified in our laboratory based on its phenotypic and biochemical properties (Byon and Kim, 2000). For further identification, 16S rRNA gene sequencing was carried out using universal primers, fD1 (5'-AGAGTTTGATCCTGGCTCAG-3') and rP2 (5'-ACGGCTA CCTTGTTACGACTT-3') (Weisburg et al., 1991). Sequences were compared with those deposited into National Center for Biotechnology Information (NCBI, USA) database using BLASTN and completely corresponded (100%) to those of P. aeruginosa (GenBank accession number: KM232911). Fish pathogenic Saprolegnia sp. FP9001 (hereinafter Saprolegnia sp.) was donated by the National Institute of Fisheries Science (NIFS). MB1-3 was grown on tryptic soy agar (TSA, BD) containing 1.5% NaCl at 25°C. Saprolegnia sp. was cultured in Sabouraud’ dextrose agar (SDA, BD), or glucose-yeast extract agar medium (GY agar; 1% glucose, 0.25% yeast extract, 1.5% agar) added with penicillin-streptomycin solution (P/S, Sigma) at 20°C. Two fish cell lines, a rainbow trout gonadal (RTG-2) cell line and a Chinook salmon embryonic (CHSE-214) cell line, were donated by the Laboratory of Fish Prevention of Chonnam National University. These fish cell lines were cultured in Dulbeccós modified Eagle medium (DMEM, Gibco) supplemented with 10% fetal bovine serum (FBS, Sigma) and 1% stabilized penicillin-streptomycin solution (P/S, Sigma). DMEM with 2% FBS (MEM₂) and DMEM without FBS (MEM₀) were also used. Fertilized rainbow trout eggs were purchased from a fish farm in Pyeongchang, Gangwon-do, South Korea.

Separation of antifungal fraction from MB1-3 culture supernatant

The antifungal fraction of MB1-3 culture supernatant was separated using the method of Lategan et al. (2006) with slight modifications. Briefly, pre-cultured MB1-3 was inoculated into 1 L tryptic soy broth (TSB, BD) and incubated at 25°C with shaking (180 rpm) for 24 h, followed by heat-shock at 70°C for 1 h. After centrifugation at 6000 rpm for 15 min (Supra 25K, Hanill), only the supernatant was collected. The culture supernatant was then fractionated with ethyl acetate (Daejung) at supernatant to ethyl acetate ratios of 1:0.5, 1:0.3, and 1:0.2. These fractions were concentrated under reduced pressure using a rotary vacuum concentrator (BUCHI V-805, Buchi). Approximately 0.1 g of antifungal fraction (MB-EA) was obtained. MB-EA was dissolved in 10% dimethyl sulfoxide (DMSO, Junsei) when necessary.

Inhibition test of MB1-3 and MB-EA on growth of Saprolegnia sp.

Inhibition activities of MB1-3 and MB-EA against hyphal growth of Saprolegnia sp. were determined using a plate assay. Briefly, marginal hyphae of Saprolegnia sp. cultured in GY agar for 2 days was cut with a cork borer into 8 mm in diameter. Agar blocks of the mycelium, MB1-3, and 20 mg/disc of MB-EA absorbed into paper discs (8 mm in diameter, ADVANTEC) were loaded onto SDA medium (Fig. 1). Antifungal effect was evaluated based on the degree of growth inhibition after culturing for 24 h according to the method described by Bly et al. (1997).

Fig. 1. Antifungal activities of Pseudomonas aeruginosa MB1-3 and ethyl acetate extract of its culture supernatant (MB-EA) against Saprolegnia sp. after 24 h culture on Sabouraud dextrose agar medium. (A) Mycelia growing from an agar block of Saprolegnia sp.; (B) Fungal mycelia inhibited by MB1-3 in co-culture; (C) Mycelia inhibited by 20 mg/disc of MB-EA.

Comparison of mycelial growth according to MB-EA concentration

Mycelial growth inhibition effect according to concentration of MB-EA was evaluated by measuring the size of mycelia grown in liquid medium (Jee and Lee, 2009; Miura et al., 2009). GY broth (10 ml each) was used to adjust MB-EA to 0, 50, 100, 200, 300, 400, and 500 mg/L in 15 ml test tubes. In addition, 50 mg/L of formalin (37% formaldehyde solution, Junsei) and 0.001 mg/L of MG (MBcell, Kisanbio) were prepared. Concentrations of formalin and MG were determined by referring to various results of long-term bath treatment (van Heerden et al., 1995; Máchova et al., 1996; Sudova et al., 2007; Jee and Lee, 2009). MG and formalin are excellent antifungal substances that show good effects at relatively low concentrations, they have been used as reference substances frequently to evaluate antifungal effects of new substances. MG and formalin were used as positive controls in this study. An agar block of 8 mm in diameter having mycelia was inoculated into each test tubes. The length of hyphae was measured while incubating at 20°C for 5 days. Effect of MB-EA was evaluated by comparing results with results of 50 mg/L formalin and 0.001 mg/L MG. The concentration that inhibited growth by 50% of the maximum mycelial growth (IC₅₀) was obtained using a four-parameter logistic (4PL) curve calculator (https://www.aatbio.com)

Inhibition assay of MB-EA on spore germination

Inhibitory effect of MB-EA on spore germination was determined according to the method of Miura et al. (2009). Briefly, Saprolegnia sp. agar block was inoculated into a Petri-dish containing 20 ml of GY broth medium and incubated at 20°C for 48 h. After washing with sterile groundwater, the mycelium was transferred to a Petri-dish containing fresh sterile groundwater and left at 20°C for 24 h to form spores. Mycelium was removed using filter paper (Whatman No. 541). Spore suspension at a concentration of 1.0 × 10⁵ spores/ml was then prepared using a hemocytometer.

After 100 μl of spore suspension was inoculated into each well of 96-well plates containing 100 μl of MB-EA solution with sterilized ground water adjusted to 0, 100, 200, 400, 800, 1000, and 2000 mg/L, plates were incubated at 20°C for 24 h. Then 3 μl of GY broth was added to promote germination of zoospores (Miura et al., 2009). After 3 h, the germination rate of spores was determined by observing under a microscope (× 400 magnification). Spore germination inhibition rate of MB-EA was calculated as {1 - (the rate of germinated zoospore in tested/the rate of germinated zoospore in control)} × 100.

Cytotoxicity assay

Cytotoxicity was evaluated by MTT (Thiazolyl blue tetrazolium bromide, Sigma) assay (Park et al., 2008). CHSE-214 and RTG-2 cells were adjusted to a density of 6.0 × 10⁴ cells/ml in DMEM. After they were seeded (100 μl each) into 96-well plates, they were then incubated at 20°C and 15°C for 24 h, respectively. After removing the medium, 62.5, 125, 250, 500, 1000, and 2000 mg/L of MB-EA (100 μl each) and 100 μl of MEM₂ were added to wells followed by incubation at 20°C and 15°C for 48 h, respectively. After washing with 200 µl of MEM₀, 180 μl of MEM₀ and 20 μl of MTT (2.5 mg/ml PBS) were added to each well. Formation of formazan was confirmed after reaction at 20°C for 4 h. The medium was completely removed under dark conditions. After 75 μl of DMSO was added, the plate was incubated at 20°C for 2 h to dissolve formazan. The absorbance was then measured at 570 nm using a microplate reader (SpectraMax 340, Molecular Devices).

Examination of MB-EA effect on hatching of rainbow trout eggs (small in vivo experiment)

Hatching and mortality according to treatment time were examined to investigate effect of MB-EA on hatching of rainbow trout fertilized eggs. Fifty rainbow trout eyed eggs were accommodated in a net installed in 1 L beaker. After, 600 ml of UV-treated groundwater was supplied, the temperature was maintained at 13°C. In experimental groups treated with 250 mg/L of MB-EA for 30 min/day for 2 days or 24 h, the hatching rate and deformity rate were determined until all eyed eggs hatched. The control group proceeded in the same way as the experimental group using distilled water (DW).

Isolation and identification of an active compound from MB-EA

To separate active subfraction from MB-EA (1.0353 g), reverse pressure liquid chromatography (RP-MPLC, Combi Flash graduate [Isco]) was performed using a gradient with increasing amount of acetonitrile in H₂O (5→100% aqueous acetonitrile) and C₁₈ column (43 g, RediSep). Thin-layer chromatography (TLC) was then performed for obtained eluates. Eluates showing similar patterns of spots were collected into seven subfractions (Fr. I–VII) (Fig. 4). Antifungal activities of subfractions were evaluated by disc diffusion method. After dissolving fractions in DMSO (final concentration 10%), 100 μg/disc was adsorbed to a sterile paper disc of 8 mm in diameter. The diameter of the growth inhibitory zone was measured after co-incubation with an agar block (Φ8 mm) having mycelia of Saprolegnia sp. at 20°C for 48 h. As controls, discs adsorbed with DW and DMSO were also used.

Fig. 4. Fractionation scheme to isolate active compound from parental fraction, MB-EA. Compound 1 with the highest activity was obtained from subfraction IV.

Subfraction Fr. IV (17 mg) showing higher activity than MB-EA (Table 3) was selected and then subjected to semi- preparative high-performance liquid chromatography (HPLC) using a Gilson 151 UV-VIS detector and a 321 pump, and equipped with a C₁₈ column (250 × 10 mm, 4 µm, J’sphere ODS-H80, YMC) with isocratic elution in 51% aqueous acetonitrile (2 ml/min) to give a pure compound, compound 1 (1 mg). To elucidate the chemical structure of compound 1, mass spectrometry (MS, JMS-700 MStation mass spectrometer, JEOL), ¹H and ¹³C nuclear magnetic resonance (NMR) (in CD₃OD, Varian VNMRS spectrometer 600MHz, Varian) were performed. Spectral data were then analyzed by comparison with literature values (Rattanachuay et al., 2011; Youn et al., 2018). MS data were obtained from the National Center for Inter-University Research Facilities of Seoul National University. NMR data were obtained from Gwangju Center of Korea Basic Science Institute.

Table 3 Antifungal activities of subfractions (100 μg/disc) separated from the ethyl acetate fraction of Pseudomonas aeruginosa MB1-3 supernatant (MB-EA) against Saprolegnia sp.

Culture time (h)	Inhibition zone (mm)
	Control	DMSO	MB-EA	Fr.Ⅰ	Fr.Ⅱ	Fr.Ⅲ	Fr.Ⅳ	Fr.Ⅴ	Fr.Ⅵ	Fr.Ⅶ
2	0	0	0	0	0	0	0	0	0	0
24	0	0	3	0	2	2	8	3	1	1
36	0	0	5	0	4	4	10	7	4	2
48	0	0	6	0	5	5	20	7	5	3

2-Heptyl-4-quinolone (1). White amorphous powder; C₁₆H₂₁NO, FAB-MS m/z 244 [M+H]⁺. ¹H-NMR (600 MHz, CD₃OD): δ 8.20 (1H, dd, J=8.4, 1.2 Hz, H-6), 7.69 (1H, td, J=6.6, 1.2, H-), 7.58 (1H, d, J=7.8), 7.39 (1H, td, J=6.6, 1.2, H-), 6.23 (1H, s, H-3), 2.72 (2H, t, J=7.8 Hz, H-1') 1.77 (2H, quint, J=7.2, H-2'), 1.43-1.28 (8H, m, H-3' to H-6') 0.89 (3H, t, J=7.2 Hz, H-7'). ¹³C-NMR (150 MHz, CD₃OD): δ 180.7 (C-4), 157.3 (C-2), 141.7 (C-10), 133.5 (C-8), 126.1 (C-6), 125.6 (C-5), 125.2 (C-7), 119.2 (C-9), 108.9 (C-3), 35.1 (C-1'), 33.0 (C-2'), 30.3-30.2 (C-3' to C-5'), 23.8 (C-6'), 14.5 (C-7').

Determination of MIC and MFC of compound 1

Minimum inhibitory concentration (MIC) was determined by a micro-dilution assay as described by Fujita et al. (2008) with modification. Briefly, after adjusting final concentrations of compound 1 to 3.125 mg/L to 3200 mg/L in 300 μl of GY broth with a 2-fold dilution (DMSO final concentration: 1% or less), agar block of mycelia (Φ5 mm) was inoculated into GY broth. After culturing at 20°C for 48 h, MIC value was determined as the lowest concentration of compound 1 that resulted in no visible growth. After determining MIC, mycelia of agar block were washed with GY broth three times to remove antifungal substance and inoculated into 300 μl of new GY broth. After incubation at 20°C for 48 h, the lowest concentration at which the mycelium did not grow was determined as the minimum fungicidal concentration (MFC) of compound 1. MIC and MFC of formalin as control were determined with the same concentration gradient.

Data analysis

Data were subjected to unpaired T-test or one-way analysis of variance (ANOVA) using SPSS 27.0 software followed by post-hoc analysis using the Tukey HSD module. Significant differences were defined at p < 0.05.

Antifungal activities of MB1-3 and MB-EA

Figure 1 shows results of 24 h of culture after inoculation with Saprolegnia sp., MB1-3, or MB-EA. Mycelia of control Saprolegnia sp. extended from the center of the agar block to the edge of the medium (plate A). However, they did not expand around the MB1-3 strain but in the opposite direction (plate B). Mycelial growth was inhibited by disc soaked with MB-EA. However, DMSO solvent did not affect mycelial growth (plate C). Inhibition of mycelial growth by MB1-3 was higher than that of MB-EA, meaning that MB1-3 bacteria continuously produced inhibitory substances as they grew.

Inhibitory effects of different concentrations of MB-EA on growth of Saprolegnia sp.

As a result of observing the degree of mycelial growth inhibition according to the concentration of MB-EA (Fig. 2), the higher the concentration, the more the mycelial growth was inhibited. All experimental groups treated with MB-EA at different concentration showed significant differences in mycelial growth inhibition compared to the control. MB-EA at 50 mg/L and 500 mg/L reduced hyphal growth by 18.3% and 55.9%, respectively. IC₅₀ against mycelial growth was calculated as 352.24 mg/L in broth culture. In particular, 300, 400, and 500 mg/L of MB-EA showed no difference in inhibitory effect by days for 5 days. Moreover, MB-EA at 200 mg/L or higher inhibited mycelial growth, which was significantly different from 50 mg/L formalin and 0.001 mg/L MG (all p < 0.05). The length of mycelia grown for 3 days in 50 mg/L formalin (25 mm) showed about 43% inhibition compared to the controls (43.75 ± 1.5 mm), which was similar to the 200 mg/L MB-EA result (25.8 ± 4.35 mm). The growth inhibitory effect of formalin on mycelia decreased over time. Mycelial growth on the fifth day was similar to that of the controls. However, the mycelial inhibitory effect of MB-EA tended to be maintained.

Fig. 2. Comparison of mycelial growth according to various concentrations of ethyl acetate fraction of Pseudomonas aeruginosa MB1-3 culture supernatant (MB-EA) (50 mg/L formalin and malachite green (MG) of 0.001 mg/L were used for effect comparison). Different letters above the corresponding columns indicate statistically significant difference between treatments (p < 0.05).

Effect of MB-EA on spore germination

The spore germination inhibition rate of MB-EA was expressed as a relative inhibition rate (Table 1). MB-EA at concentrations of 50 mg/L and 100 mg/L showed inhibition rates of 24.3 ± 3.2% and 25.0 ± 13.0%, respectively. As the concentration of MB-EA increased, the inhibition rate increased reaching 65.0 ± 10.8% at 1000 mg/L. Spore germination inhibition rate was not significantly different for MB-EA at 0, 50, or 100 mg/L. IC₅₀ of MB-EA based on relative inhibition of spore germination was calculated to be 551.6 mg/L.

Table 1 Effect of ethyl acetate fraction of Pseudomonas aeruginosa MB1-3 culture supernatant (MB-EA) on spore germination of Saprolegnia sp.

MB-EA concentration (mg/L)	Spore germination (%)*	Relative inhibition ratio (%)†
0	59.3 ± 36.1	0a
50	45.2 ± 28.0	24.3 ± 3.2ab
100	45.6 ± 32.0	25.0 ± 13.0ab
200	42.3 ± 32.5	33.0 ± 12.0b
400	37.8 ± 28.3	40.0 ± 8.9bc
500	33.8 ± 27.7	47.3 ± 12.5bc
1000	19.7 ± 10.7	65.0 ± 10.8c

* Spore germination (%) = (Number of germinated zoospore/Number of total spore) ×100

^† Relative inhibition ratio (%) = [1 - (the rate of germinated zoospore in tested/the rate of germinated zoospore in control)]

^a,b,c Significantly different in spore germination inhibition rate at p < 0.05.

Cytotoxicity of MB-EA

Cytotoxicities of MB-EA to fish cell lines are shown in Fig. 3. The relative survival rate compared to the control was more than 90% when CHSE-214 cells were treated with MB-EA at concentration up to 125 mg/L. It was 85.37% when CHSE-214 cells were treated with MB-EA at 250 mg/L. On the other hand, the relative survival rate was more than 78.72 ± 18.87% when RTG-2 cells were treated with MB-EA at concentrations up to 250 mg/L, which was not significantly different from that of the untreated group. Cell viability decreased significantly with increasing MB-EA concentration. Relative survival rates of CHSE-214 cells and RTG-2 cells treated with MB-EA at 500 and 1000 mg/L were significantly different from those of the untreated group.

Fig. 3. MTT analysis of the viability of two fish cell lines, CHSE-214 and RTG-2, at various concentrations of ethyl acetate extraction from Pseudomonas aeruginosa MB1-3 culture supernatant (MB-EA). *, significantly different from control (p < 0.05).

Effect of MB-EA on hatching of rainbow trout fertilized eggs

The survival rate of firtilized eggs slightly increased in the group treated with MB-EA at 250 mg/L. However, hatching rates of eyed eggs were not significantly different between the control group and the MB-EA 30 min dipping group, but different from MB-EA 24 h bath group for 1 week (Table 2). Deformities of hatched eggs did not appear in any groups for 10 days.

Table 2 Hatching rates of fertilized or eyed rainbow trout (Oncorhynchus mykiss) eggs treated with 250 mg/L of the ethyl acetate fraction of Pseudomonas aeruginosa MB1-3 cultural supernatant (MB-EA) by dipping for 30 min or bathing for 24 h

Group	Fertilized egg, for 1 week, %			Eyed egg
	Mean survival rate*	Mean hatched rate†		Mean hatched rate‡
Control	76.4 ± 18.5	11.1 ± 2.2a		51.3 ± 34.5a
MB-EA 30 min	82.0 ± 17.0	31.3 ± 25.9ab		51.6 ± 27.3a
MB-EA 24 h	91.7 ± 7.5	3.8 ± 2.9b		2.0b

* (Survival egg/stocked fertilized eggs) ×100

^† (Hatched eggs/eyed eggs from fertilized eggs) × 100

^‡ (Hatched eggs/eyed eggs) × 100

^{a, b} Different letters in the same column indicate significant differences within groups at p < 0.05.

Antifungal activity-guided isolation and identification of an active compound from MB-EA

To isolate active compounds from MB-EA, MB-EA was subjected to RP-MPLC, affording seven subfractions (Fr. I–VII) (Fig. 4). As a result of confirming antifungal activities of these seven subfractions against Saprolegnia sp., Fr. IV showed higher antifungal activity than the parent fraction MB-EA (Table 3). Through semi-preparative HPLC of Fr. IV, compound 1 was isolated and identified as 2-heptyl-4-quinolone (Fig. 5) by direct comparison of their spectroscopic data with previously reported data (Rattanachuay et al., 2011; Youn et al., 2018).

Fig. 5. Chemical structure of compound 1 isolated from the ethyl acetate fraction of Pseudomonas aeruginosa MB1-3 culture supernatant (MB-EA).

MIC and MFC of compound 1 against Saprolegnia sp.

Mycelial growth was not detected in the group treated with 200 mg/L of compound 1. Thus, MIC of compound 1 was determined to be 200 mg/L. Subsequently, when Saprolegnia sp. agar block was cultured in a new medium without MB-EA, mycelial growth was not observed from 800 mg/L. On the other hand, the MIC and MFC of formalin used as a standard were both 50 mg/L (Table 4).

Table 4 Determination of minimal inhibitory concentration (MIC) and minimal fungicidal concentration (MFC) based on mycelium growth of Saprolegnia sp. in GY broth containing various concentrations of compound 1 and formalin

		Concentration (mg/L)
		0	3.125	6.25	12.5	25	50	100	200	400	800	1600	3200
Compound 1	MIC	+*	+	+	+	+	+	+	-	-	-	-	-
	MFC	+	+	+	+	+	+	+	+	±	-	-	-
Formalin	MIC	+	+	+	+	+	-	-	-	-	-	-	-
	MFC	+	+	+	+	+	-	-	-	-	-	-	-

* +, Mycelium growth; -, no mycelium growth.

Pseudomonas aeruginosa strain MB1-3 and MB-EA showed anti-Saprolegnia activities (Fig. 1). MB-EA showed antifungal activity even after it was subjected to 70°C heat shock. Thus, it was presumed to be a heat-stable non-protein substance. Moghaddam et al. (2012) have reported that P. aeruginosa (PTCC1430) can stop fungal growth at a cell concentration of 10⁷ with MIC of about 10⁴ CFU/ml, which could be used as a biological agent to control saprolegniosis. However, since P. aeruginosa can cause disease in humans, there is a limit of using the bacterium itself as a probiotic. Pseudomonas aeruginosa is known to produce different types of secondary substances depending on the strain, nutrient composition, and culture environment (Leisinger and Margraff, 1979). Its anti-microbial substances have especially attracted attention of researchers. Kerr et al. (1999) have isolated P. aeruginosa with strong in vitro inhibition activities (by cross streak assay) against Candida albicans and Aspergillus fumigatus and confirmed that pyocyanin and 1-hydroxy-phenazine are its major anti-fungal factors. González-Palacios et al. (2020) have reported that two isolates of P. fluorescens (LE89 and LE141) can reduce S. parasitica infection in rainbow trout and that the mode of action of these bacteria is related to the production of siderophores. Siderophores inhibited both mycelial growth and cyst germination of S. parasitica without heat treatment at high levels. However, such inhibitory effects disappeared after heat treatment at 70°C for 1 h or 100°C for 5 min (González-Palacios et al., 2020). Wang and Zhang (2017) have partially purified an antifungal peptide of 58 kDa from cell-free disruption supernatant of P. protegens strain XL03. The peptide inhibited the growth of Saprolegnia hyphae and the germination of spores in vitro with an MIC of 0.0625 mg/ml. The peptide was also thermally stable between 30°C and 60°C similar to MB-EA. Thermal stability of antimicrobial substances is considered to be a very important feature in the process of manufacturing, storing, and using the material.

MB-EA inhibited mycelial growth in a concentration-dependent manner (Fig. 2). MB-EA at concentration higher than 200 mg/L inhibited mycelial growth significantly different from formalin at 50 mg/L and MG at 0.001 mg/L. In particular, 300, 400, and 500 mg/L of MB-EA showed no difference in inhibitory effect by day for 5 days, meanings that a sufficient growth inhibitory effect could be obtained with MB-EA even at a concentration lower than its IC₅₀. Our results with 50 mg/L formalin and 0.001 mg/L MG differed from those of Jee et al. (2007). They reported that there was no mycelial growth for 6 days in the presence of 50 mg/L formalin and that mycelial growth was inhibited by 1 mg/L of MG. However, there was no mycelial growth in the presence of 1 mg/L of MG (data not shown), and similar to the result of control at 50 mg/L of formalin in our experiment (Fig. 2). Compared to results of Jee et al. (2007) that 1 mg/L of MG showed mycelial growth inhibition rate of 38.3% on the 6th day, the concentration of MB-EA in terms of an effect equivalent to 1 mg/L of MG was calculated to be about 197 mg/L. The mycelial inhibitory effect of 50 mg/L of formalin was not different from that of the control after 1 day. This meant that the effect was lost due to volatilization of formaldehyde. Jho and Shin (2018) have reported that the half-life of formaldehyde is between 24 and 168 h (1 to 7 days) in surface water based on the USA Environmental Protection Agency's data (US EPA, 2008). Due to volatility and toxicity of formalin, there are few studies on long-term bathing treatment with formalin. The concentration of formalin recommended by aquatic pharmaceutical manufacturers for disinfection of freshwater fish eggs is 1000–2000 mg/L. A short-term application of 15 min is recommended (NFQS, 2022). Formalin often affects the gills, liver, kidney and spleen of treated fish. Tavares-Dias (2021) has suggested that the appropriate concentration of formalin for a prolonged bath is lower than 25 mg/L and that concentrations higher than 25 mg/L are appropriate for short-term baths while a lower concentration of 12.5 mg/L is recommended for more sensitive fish species or extremely sick fish. Therefore, it is considered that the organism should be treated for less than 3 days at a much lower concentration than 50 mg/L considering the cytotoxicity of formalin in the case of bath treatment.

The relative inhibition rate of spore germination of Saprolegnia sp. after treatment with MB-EA increased as the concentration of MB-EA increased (Table 1). Significant inhibition rates of spore germinations were only obtained after treatment with MB-EA at concentrations higher than 200 mg/L. Suppressing germination of fungal spores capable of infecting fish and eggs is important to block the infection route of aquatic fungal diseases transmitting in the form of spores. The IC₅₀ of MB-EA based on the relative inhibition of spore germination was calculated to be 551.6 mg/L. In this study, spores of Saprolegnia sp. did not germinate 100% even in the control group. Therefore, if MB-EA is used for fungal control of rainbow trout eggs, a concentration much lower than the calculated IC₅₀ value is likely to be the appropriate concentration to reduce the spore germination rate to 50%.

Relative survival rates of two fish cell lines after treatment with MB-EA at 500 and 1000 mg/L were significantly different from those of the untreated group. Hatching rates of eyed eggs and deformities in hatched eggs did not show significant difference between the control group and the MB-EA 30 min or 24 h bath group (Table 2). Therefore, for controlling Saprolegnia sp. in fish eggs using MB-EA, treatment at the same concentration as IC₅₀ for inhibiting spore germination is undesirable because of its potential cytotoxicity.

Oladosu et al. (2022) have extracted pyocyanin from P. aeruginosa isolate and verified its antifungal effect on African catfish eggs infected with Saprolegnia sp. They reported that pyocyanin effectively prevented fungal infection in eggs but stopped the development of the embryo. This might be due to the toxic effect of pyocyanin by mediating hydrogen peroxide production. Therefore, among various substances produced by P. aeruginosa isolates, it is important to distinguish substances that are safe for cells and eggs with antifungal effects.

The anti-Saprolegnia compound of MB-EA, 2-heptyl-4-quinolone (HHQ) is a direct biosynthetic precursor of 2-heptyl-3-hydroxy-4-(1H)-quinolone referred to as the “Pseudomonas quinolone signal” (PQS) (Pesci et al., 1999). Faille (2010) has found that the culture supernatant of P. aeruginosa strain PA14 can inhibit the growth of Saprolegnia sp. and that its active compound is HHQ. However, that study only suggested that the anti-Saprolegnia molecule of P. aeruginosa strain PA14 was HHQ using a bioautography method in TLC with a pure compound HHQ. It did not evaluate the cytotoxicity of the antifungal compound or its effect on fish or fish eggs. In contrast, our study confirmed that the antifungal compound found in the culture supernatant of MB1-3 was HHQ through isolation and structural identification by spectroscopic analyses. In addition, we determined the MIC and MFC of HHQ against Saprolegnia sp.

HHQ is an intercellular signaling molecule produced by P. aeruginosa (Diggle et al., 2007). HHQ, PQS, and some 4-quinolone derivatives can modulate interspecies and interkingdom interactions (Reen et al., 2011, 2012, 2015). However, detailed mechanisms remain mostly unknown.

Reen et al. (2011) have found that motility, biofilm formation, and adhesion are repressed in a broad range of bacteria and C. albicans, and that potent bacteriostatic activity is shown in the presence of HHQ. The significant reduction in the ability of C. albicans to form biofilms in the presence of HHQ reveals an interkingdom function. They confirmed that biofilm-specific effects of HHQ were seen only during development, but not in yeast-hyphal switch. This supported our results that IC₅₀ of MB-EA against mycelium growth was lower than that for Saprolegnia sp. spore germination.

However, biological specificity, motility inhibition, and anti-biofilm activity of HHQ were altered in its analogs, such as alteration of the C position of the anthranilate-derived ring, shortening or lengthening of the alkyl chain at C-2, and substitution of C-3 (Reen et al., 2015; Ramos et al., 2020). Reen et al. (2015) have reported that when HHQ is altered at any position, the anti-biofilm activity of the parent molecule is completely lost, suggesting that the anti-biofilm activity of HHQ is very sensitive to structural modifications. Ramos et al. (2020) have compared the inhibition of biofilm formation in various synthetic derivatives of HHQ and revealed that the bacteriostatic activity of HHQ molecules is highly sensitive to modifications of the HHQ molecular structure. Therefore, the anti-biofilm activity of HHQ is considered to be related to the antifungal activity of MB-EA.

HHQ exerts antibacterial activity as well as antifungal effects in different ways from bacteriostatic or bactericidal action of antibiotics. Antibiotics exhibit bacteriostatic or bactericidal action by inhibiting the synthesis of substances necessary for cell growth, such as proteins and cell wall components, or by inhibiting DNA replication by interfering with activities of DNA gyrase and topoisomerase IV (Senerovic et al., 2020). On the other hand, HHQ exhibits bacteriostatic activity by influencing the development of bacterial communities through activities such as inhibition of motility or inhibition of biofilm formation. Most antibacterial agents can develop resistance quickly after use with a short efficacy period, whereas HHQ exhibits bacteriostatic activity against antimicrobial-resistant bacteria, which is advantageous for controlling pathogenic microorganisms.

It was found that 48 h MIC and MFC of compound 1 against Saprolegnia sp. were 200 mg/L and 800 mg/L, respectively (Table 4), indicating a fungistatic activity. Compared to formalin, which showed fungicidal effects, the MIC value of compound 1 was four times higher. However, compound 1 does not produce volatile substances, such as formaldehyde harmful to human. In addition, its antifungal effect can be maintained longer than formalin with a single use.

In conclusion, the antifungal activity of MB-EA is related to the activity of 2-heptyl-4-quinolone. In addition, MB-EA can inhibit both mycelial growth and spore germination of Saprolegnia sp. at concentrations up to 300 mg/L without showing adverse effects on fish eggs. Therefore, MB-EA is expected to be developed as a disinfectant for rainbow trout eggs to replace MG through additional studies on its production efficiency, field applicability, and public health safety.

본 연구에서는 P. aeruginosa MB1-3 배양 상층액의 에틸아세테이트 추출물 (MB-EA)을 이용하여 무지개 송어 알의 saprolegniosis 방제 가능성을 평가하고자 하였다. MB-EA의 Saprolegnia sp. FP9001에 대한 항진균활성을 시험관 배양으로 확인한 결과, 50 mg/L에서는 균사 성장이 18.3%, 500 mg/L에서는 55.9% 감소하였으며, 모든 MB-EA 실험 농도군은 대조군과 비교하여 균사 생장 억제에서 유의한 차이(p < 0.05)를 나타냈다. 다양한 농도의 MB-EA 용액에서 Saprolegnia sp. 포자의 발아 정도를 평가하였다. 포자 발아 억제율은 50 mg/L에서 24.3 ± 3.2%, 1000 mg/L에서 65.0 ± 10.8%였다. 어류 주화세포인 CHSE-214 및 RTG-2에 대한 MB-EA의 독성을 MTT 분석을 통해 알아본 결과, 250 mg/L 농도까지 대조군과 유의한 차이를 보이지 않았다. 또한, 무지개 송어 수정란의 생존율과 발안란의 부화율에 대하여 MB-EA가 미치는 영향도 대조군과 30분 2회 처리구 간에 차이가 없었다. MB-EA의 활성 화합물을 RP-MPLC와 HPLC로 분리 정제하고 MS, ¹H-NMR, ¹³C-NMR을 통해 구조를 확인하였다. 주요 활성 물질은 세균 간 신호전달물질인 2-heptyl-4-quinolone (HHQ)으로 동정되었으며 Saprolegnia sp.에 대한 MIC는 200 mg/L이었다. 따라서 MB-EA는 HHQ에 의해 항진균활성을 나타내며, 무지개 송어 양식장에서 어류 수정란의 saprolegniosis 방제 후보 물질로 개발 가능하다 하겠다.

None.

The authors declare that they have no conflict of interest.

Ethical review and approval was not required for the animal study because the egg stage of fish was exempted from ethical review and approval according to the guidelines of Institutional Animal Care and Use Committee.