In vitro activity of chemicals and commercial products against Saprolegnia parasitica and Saprolegnia delica strains

Abstract Oomycetes of the genus Saprolegnia are responsible for severe economic losses in freshwater aquaculture. Following the ban of malachite green in food fish production, the demand for new treatments pushes towards the selection of more safe and environment‐friendly products. In the present work, in vitro activity of ten chemicals and three commercial products was tested on different strains of Saprolegnia, using malachite green as reference compound. The compounds were screened in agar and in water to assess the minimum inhibitory concentration (MIC) and the minimum lethal concentration (MLC), respectively. Two strains of Saprolegnia parasitica and one isolate of Saprolegnia delica were tested in triplicate per each concentration. Among tested chemicals, benzoic acid showed the lowest MIC (100 ppm) followed by acetic acid, iodoacetic acid and copper sulphate (250 ppm). Sodium percarbonate was not effective at any tested concentration. Among commercial products, Virkon™S was effective in inhibiting the growth of the mycelium (MIC = MLC = 1,000 ppm). Actidrox® and Detarox® AP showed MIC = 5,000 and 1,000 ppm, respectively, while MLCs were 10‐fold lower than MICs, possibly due to a higher activity of these products in water. Similarly, a higher effectiveness in water was observed also for iodoacetic acid.

urge the screening of new molecules and products active against Saprolegnia spp.
Treatments with MG were administered either as monocomponent baths or as multicomponent baths (in combination with formaldehyde and other products) (Sudova et al., 2007). However, particular concern is directed towards its use in the production of food fish: The high affinity and persistence of MG and, particularly, its reduced form (leucomalachite green-LMG) in animal tissues (Alborali, Sangiorgi, Leali, Guadagnini, & Sicura, 1997;Bauer, Dangschat, Knoppler, & Neudegger, 1988;Clifton-Hadley & Alderman, 1987;Machova et al., 1996;Plakas, El-Said, Stehly, Gingerich, & Allen, 1996) generates public health issues related to its potential carcinogenicity, teratogenicity and mutagenicity in humans, as suggested by experimental evidence in mammals (Culp et al., 2006;Mayer & Jorgenson, 1983;Werth, 1958). Following a considerable amount of studies demonstrating its toxicity and carcinogenicity in different animal species, the use of MG in the production of fish destined to human consumption is not authorized in the European Union (EU) (European Commission, 2010). However, since residues of MG and LMG have been detected in aquaculture products in monitoring programmes in EU Member States, the European Food Safety Authority (EFSA) established that food contaminated with MG/LMG at or below the reference point for action (RPA) of 2 μg/kg is unlikely to represent a public health concern (EFSA, 2016).
Several other chemicals have been used with different degrees of success for the treatment of Saprolegnia infections.
Among products specifically registered for aquaculture in some European countries, Pyceze® is employed in daily baths as prophylactic/therapeutic measure against fungal and bacterial infections in fish food eggs. The active ingredient of Pyceze® is bronopol (2-brom o-2-nitropropane-1,3-diol), a broad-range biocide used as a preservative in pharmaceutical and cosmetic industry (Bryce, Croshaw, Hall, Holland, & Lessel, 1978;Kumanova, Vassileva, Dobreva, Manova, & Kupenov, 1989;Toler, 1985), which would therefore pose no severe toxicological risks to human health. Nevertheless, its toxicity towards phytoplankton and zooplankton has been demonstrated.
Formalin, a solution of 37% formaldehyde, has been effectively employed to prevent (Bly, Quiniou, Lawson, & Clem, 1996;Schreier, Rach, & Howe, 1996) and treat (Cline & Post, 1972;Marking, Rach, & Schreier, 1994;Walser & Phelps, 1994) Saprolegnia infection in fish eggs. As a prophylactic measure, it inhibits cyst germination at a concentration of 250 mg/L (Bly et al., 1996). Daily flushes of formalin with 100, 200 and 400 mg/L increased per cent hatch of channel catfish Ictalurus punctatus eggs in comparison with non-treated eggs (Walser & Phelps, 1994). Although formalin is still included in the list of allowed substances for all the food-producing species and is marketed as a biocide for the disinfection of equipment and facilities (European Commission, 2010), it is not listed under the Biocidal Products Regulation (BPR) (European Parliament and Council, 2012), which is required for all products marketed as biocides. Moreover, formalin is currently not approved as a veterinary medicine for the treatment of live fish in most of the EU countries. However, in Spain, one formalin-based product (Aquacen) has a marketing authorization for the control of Philasterides dicentrarchi in the farming of turbot Psetta maxima (Verner-Jeffreys & Taylor, 2015). In the United States, formalin has been approved by Food and Drug Administration (USFDA, 2018) in three commercial formulations (Formalin-F ™ , Formacide-B and Parasite-S®) for egg disinfection in aquaculture. From 1 January 2016, formaldehyde has been classified as a category 1B carcinogen (European Commission, 2014), therefore its use should be subjected to certain restrictions. Furthermore, besides the carcinogenic risk, formaldehyde represents a risk for exposed workers since it is a powerful irritant and allergenic substance.
Several other studies were performed to identify suitable compounds for the treatment or disinfection against Saprolegnia.
Ozone has shown effectiveness comparable to formalin in preventing saprolegniosis in brown trout (Salmo trutta fario) eggs (Forneris et al., 2003). The possible use of ozonized water in fish tanks is, however, controversial due to its strong oxidant properties (Fukunaga, Suzuki, & Takama, 1991) and potential toxicity to the branchial epithelium that could negatively affect respiration and osmoregulation.
The fungicide activity of iodophores was demonstrated for the treatment of eggs in disinfectant baths, allowing to increase the hatching rate (Walser & Phelps, 1994). Despite their effectiveness and suitability for the disinfection of fish eggs, the potential use of iodophores for the treatment of a large number of fish is limited, due to the high concentrations needed (Bruno, van West, & Beakes, 2011) resulting in increased costs and potential toxicity.
Hydrogen peroxide and boric acid are reported in the literature as promising compounds to control Saprolegnia infections.
Ali, Thoen, Evensen, and Skaar (2014) tested boric acid in vitro against Saprolegnia parasitica and Saprolegnia diclina strains and in vivo on eggs and larvae of Atlantic salmon (Salmo salar). In vitro results showed inhibition of the mycelium growth at concentrations above 0.6 g/L, while in vivo tests allowed a high survival rate after continuous (0.2-1.4 g/L) and intermittent (1-4 g/L) exposure of eggs and larvae. These results suggest that boric acid could be safely used in aquaculture although its environmental impact must be carefully investigated.
The aim of this work was to perform an in vitro screening of promising molecules and commercial products against Saprolegnia spp. in order to provide information for the selection of safer and more environmentally friendly alternative treatment of saprolegniosis in aquaculture. The activity of new molecules is compared to the effectiveness of two compounds (malachite green, copper sulphate) used in the past to control saprolegniosis in aquaculture.

| Strains tested
Tests were carried out on three Saprolegnia strains: one reference strain of S. parasitica (CBS 223.65 provided by CSIC-RJB Madrid, Spain) isolated in Holland from northern pike (Esox lucius), one field strain of S. parasitica (ITT 320/15/20) isolated in Italy from brown trout (Salmo trutta fario) and one field strain of Saprolegnia delica (ITT 290/15/15) isolated in Italy from rainbow trout (Oncorhynchus mykiss). Each strain has been tested in triplicate per each concentration.

| Inocula
Saprolegnia spp. strains were maintained with periodic subcultures on glucose-yeast (GY) agar medium (5 g D-(+)-glucose, 1 g yeast extract, 12 g agar in 1L deionized water) supplemented with 6 mg/L of penicillin and 10 mg/L of oxolinic acid (GY + P + Ox) (Alderman & Polglase, 1986) and kept at 18°C. For the in vitro trials, subcultures of the strains employed were incubated at 18°C until growth covered the full diameter of the dish (48-72 hr). Inocula were obtained from the outer 10 mm of the culture, using a sterile 5-mm-diameter glass cannula (protocol I) or cutting a 4 × 4 mm piece with a sterile scalpel.

| Products tested
Products under test include ten molecules (acetic acid, benzoic acid, boric acid, copper sulphate, iodoacetic acid, lactic acid, oxalic acid, tartaric acid, hydrogen peroxide and sodium percarbonate) belonging to different chemical classes and three commercial products (Actidrox®, De Marco, Italy; Detarox® AP, Perdomini, Italy; and Virkon ™ S, Dupont, UK). Malachite green was used as a reference compound, and copper sulphate was added because it is widely used to control saprolegniosis in aquacultured fish.
Each product was dissolved and/or diluted in sterile deionized water (with the exception of benzoic acid which was dissolved in absolute ethanol and subsequently diluted in water) at concentrations of 0.1; 1; 5; 10; 50; 100; 250; 500; 1,000; 5,000 ppm.

| In vitro tests
Tests were performed following protocols I and II according to Alderman (1982).
For protocol I, different concentrations of the products under test were added to sterilized liquid GY agar at a temperature of 49°C.
Mixtures were then distributed in six-well plates (Ø 35 mm), allowing to test five different concentrations and one negative control.
Following overnight solidification, a 5-mm-diameter well was excised in the centre of the agar using a sterile glass cannula. The well was then filled with a standard 5-mm inoculum, culture surface uppermost.
Plates were incubated at 18°C and checked after 24, 48, 72 hr and F I G U R E 1 Protocol II: Filters supporting mycelium immersed in a solution of the product under test [Colour figure can be viewed at wileyonlinelibrary.com] 6 days, determining the colony diameter of the growing mycelium as average of two axes measured at 90° from each other. The minimum inhibitory concentration (MIC) was defined as the lowest concentration inhibiting the growth of the mycelium after 6 days of incubation.
For protocol II, polycarbonate filter membranes (diameter 25 mm, pore size 5 μm-Whatman International Ltd., UK) were sterilized by autoclaving and then placed on the surface of 92-mm petri dish (six per each petri dish). Filters were used as a support for the inoculum, obtained by placing a 4 × 4 mm piece of agar with growing mycelium.
The inoculum was placed, inverted, at the centre of the filter. The dishes containing the filters were then incubated for 24 hr, until the resulting mycelial growth had almost reached the edge of the filters.
The agar attached to the original inocula was then clipped off using hot forceps tips. The mycelia together with their supporting filters were lifted off the agar surface and placed in sterile 92-mm petri dishes containing different concentrations of the products under test (Figure 1). Sterile deionized water was used as negative control.
Filters were kept in contact with the solution for 1 hr and submitted to periodic agitation in order to achieve a better contact of the mycelium with the product under test. Subsequently, filters were washed twice with sterile water (for 5 and 30 min, respectively) and seeded on fresh dishes containing GY + P + Ox agar medium. Dishes were incubated at 18°C and checked after 24, 48, 72 hr and 6 days, determining the radial growth of the mycelium beyond the filter as average of two axes measured at 90° from each other. The minimum lethal concentration (MLC) was defined as the lowest concentration inhibiting any further growth of the mycelium after 6 days of incubation.

| RE SULTS
Minimum inhibitory concentrations and MLCs obtained for each tested compound and each strain are listed in Table 1

| D ISCUSS I ON
In our study, the activity of malachite green and its suitability as positive reference for the screening of antifungal compounds were confirmed (Bailey & Jeffrey, 1989;Marking et al., 1994). Although copper sulphate has been used in the past in fish culture for the control of parasitic and Saprolegnia infections (Straus et al., 2012;Sun et al., 2014), currently it is not approved for therapeutic use in aquaculture. According to recent research, the activity of this compound may vary according to different stages of Saprolegnia development: Sun et al. (2014) showed how copper sulphate prohibited the release of primary zoospores at concentrations ≥ 1 mg/L and inhibited mycelium growth at concentrations ≥ 0.5 mg/L for 24 hr.
In our study, we evaluated the effect of copper sulphate on hyphal growth, showing that after 6 days of continuous exposure in agar, a complete inhibition was achieved only at 250 ppm (MIC); however, concentrations of 50 and 100 ppm inhibited the aerial mycelium.
Also for iodoacetic acid, an inhibition of aerial mycelium on agar was observed at concentrations lower than MIC. The mechanisms that determine this would require further investigation, but chemically induced morphological hyphae changes with inhibition of aerial mycelium of Saprolegnia were already hypothesized in a previous work (Kaminskyj & Heath, 1992).
Sodium percarbonate is an environmentally safe compound that has been successfully tested in vitro against developmental stages of Ichthyophthirius multifiliis (Buchmann, Jensen, & Kruse, 2003;Heinecke & Buchmann, 2009); dosages of 12.5 mg/L for 180 min and 62.5 mg/L for 90 min were effective in killing I. multifiliis theronts (Buchmann et al., 2003), while tomont stage appears considerably more tolerant to the chemical. In our study, it was not possible to identify a MIC and MLC values of sodium percarbonate for Saprolegnia mycelium. However, a possible higher activity of this compound against other developmental stages of Saprolegnia (i.e., zoospores) cannot be excluded.
The compounds tested in the present study, with few exceptions (hydrogen peroxide, malachite green, Virkon ™ S), performed differently in protocols I and II (Table 1). As a general rule, the MLC was higher than the MIC, and for some of the tested molecules (boric and tartaric acid, sodium percarbonate) it was not possible to determine a MLC at tested concentrations. In water trials, Saprolegnia strains were kept in contact with the tested compounds for 1 hr only, while in agar such contact was continuous. Our results would therefore suggest that for some compounds, a more prolonged bath would possibly be required in order to achieve a lethal effect at lower concentrations.
However, one molecule (iodoacetic acid) and two commercial products (Actidrox®, Detarox®AP) resulted considerably more effective in water.
For Actidrox®, the different performances observed in the two protocols are possibly linked to the characteristics of the product itself, which releases peracetic acid (effective against a wide range of microorganisms) when solubilized in water. A similar behaviour was observed for Detarox®AP, an acidic sanitizer with oxidant properties, formulated with stabilized peracetic acid and hydrogen peroxide, that is widely used in the food industry for the disinfection of production equipment. Also for Detarox®AP, active compounds showed greater effectiveness in the presence of water. Both peracetic acid and hydrogen peroxide show a wide-range antimicrobial activity (Baldry, 1983;Jussila, Makkonen, & Kokko, 2011;Kitis, 2004) and low environmental impact, and are considered suitable alternative sanitizers (Pedersen, Meinelt, & Straus, 2013). Particularly, results of in vitro assessments (Jaafar, Kuhn, Chettri, & Buchmann, 2013;Jussila et al., 2011;Picón-Camacho, Marcos-Lopez, Beljean, Debeaume, & Shinn, 2012;Picón-Camacho, Marcos-Lopez, Bron, & Shinn, 2012) highlight a promising role of peracetic acid-based products for the control of parasitic and "oomycotic" infections (i.e., white spot disease, crayfish plague) in aquaculture. In water, Detarox®AP is degraded quickly, leaving residues of acetic acid and its salts; this would suggest a possible low environmental impact of the product but represent a challenge in controlling the effectiveness of the product in the farm.
Peracetic acid decay can be significantly affected by organic matter content (Pedersen et al., 2013) and possibly influenced by other water properties (hardness, ion composition). Similarly, information available in the literature (Barnes, Gabel, Durben, Hightower, & Berger, 2004) suggests that possible differences in the activity of hydrogen peroxide may occur, depending on physical and chemical properties of the water.
Hydrogen peroxide alone is regarded as one of the most promising antibacterial (Wagner, Oplinger, Arndt, Forest, & Bartley, 2010), antiparasitic (Grave, Horsberg, Lunestad, & Litleskare, 2004;Picón-Camacho, Marcos-Lopez, Bron et al., 2012) and antifungal (Marking et al., 1994) compounds to be used in fish culture. MIC and MLC values obtained in the present study for this compound (5,000 ppm) would possibly limit its applicability in the field due to its toxicity on eggs at concentration > 1,000 μl/L (Gaikowski, Rach, Olson, Ramsay, & Wolgamood, 1998;Gaikowski, Rach, & Ramsay, 1999); however, after 24 hr at 500 ppm hyphal growth was considerably slowed down. Daily administration of the compound would therefore allow to effectively control the infection at a concentration lower than the observed MIC/MLC. These results confirm previous in vivo experiments conducted on eggs of rainbow trout (Marking et al., 1994;Schreier et al., 1996) and of chinook salmon Oncorhynchus tshawytscha (Waterstrat & Marking, 1995) that showed how concentrations of hydrogen peroxide ranging from 500 to 1,000 ppm are effective in controlling S. parasitica and S. ferax infection. Rach, Gaikowski, Howe, and Schreier (1998) and Rach et al. (2005) demonstrated that the toxicity of hydrogen peroxide to fish eggs varies according to different species, but was always above 1,000 μl/L; their results are in accordance with previous studies (Marking et al., 1994;Schreier et al., 1996) in documenting increased hatching rates in fish eggs treated with 1,000 ppm hydrogen peroxide.
Iodoacetic acid performed differently for S. parasitica and S. delica in the two protocols, being more effective at inhibiting S. delica in agar, but more lethal to S. parasitica in water. This product is  is included among the disinfection by-products (DBPs) that could be produced in raw water after disinfection process. Furthermore, studies are in progress to evaluate whether subtoxic doses in water could represent a carcinogenic risk for humans (Marsà, Cortés, Hernándeza, & Marcosa, 2018).
Benzoic acid is an antifungal compound naturally produced in fruit to fight fungal infections (Brown & Swinburne, 1971); this molecule and its derivatives have long been used as antimicrobial preservatives in the food industry and as antifungal agents in topical preparations for the treatment of human infections (Rowe, Sheskey, & Quinn, 2009) (Hoskonen, Heikkinen, Eskelinen, & Pirhonen, 2015).
With respect to protocol I, boric acid considerably slowed down the mycelium growth at 500 ppm but complete inhibition was Virkon ™ S is a mixture of peroxygens, surfactants, organic acids and inorganic salts. Used in the farming industry for the disinfection of equipment and facilities, the product is described as effective against a wide range of viruses, bacteria and fungi (Virkon ™ S product container label). In aquaculture, Virkon ™ S is used at a concentration of 1% w/v for the disinfection of ponds and farm equipment (Sudova et al., 2007). Particularly, in salmonids, the experimental exposure to 1% Virkon ™ S for 15 min was effective in controlling Gyrodactylus salaris infection (Koski, Anttila, & Kuusela, 2016). However, to the best of our knowledge, the effectiveness of this product against Saprolegnia has never been tested. Our results suggest a possible use of Virkon ™ S at concentrations lower than 1% w/v for the control of saprolegniosis.
For some compounds (lactic acid, oxalic acid, tartaric acid), remarkably different behaviours were observed between the two Saprolegnia species here tested (S. delica and S. parasitica) during in vitro trials, in which higher concentrations were needed to inhibit S.
delica. This oomycete species is widely distributed in natural freshwater systems (Sarowar, Van Den Berg, McLaggan, Young, & Van West, 2014), while S. parasitica is considered more primarily pathogenic (Van West, 2006). Information about the selective activity of antifungal compounds towards different Saprolegnia species would allow for the identification of compounds that inhibit the growth of the pathogenic S. parasitica but result harmless or less harmful to the many saprophytic species of Saprolegnia naturally occurring in freshwater ecosystems, thus helping to preserve the structure and diversity of natural oomycete communities.
In conclusion, the in vitro tests performed here show that, with the exception of sodium percarbonate, the compounds/products tested are effective against Saprolegnia spp., although at different concentrations. Among these, benzoic acid and iodoacetic acid showed the lowest MIC/MLC, respectively; however, acetic acid and peracetic acid-based products, particularly in combination with hydrogen peroxide, represent promising candidates for controlling saprolegniosis in aquaculture, due to their effectiveness associated with low environmental impact.
In order to assess the possible field application of these most promising compounds, further tests will be necessary to evaluate their efficacy on different developmental stages of Saprolegnia (i.e., zoospores), their possible cytotoxic effects, and ultimately their safety and efficacy in vivo.