Comparative In Vitro and In Vivo Effects of Feed Additives on Rainbow Trout Response to Ichthyophthirius multifiliis

Control of the skin parasitic ciliate Ichthyophthirius multifiliis is currently based on laborious chemical and mechanical approaches, thus advocating for alternative control measures. Here, we show that the early development of trophonts (feeding stage residing in the epidermis) in the Rainbow Trout Oncorhynchus mykiss epidermis was inhibited at 5 d postinfection after the administration of feed additives, including garlic Allium sativum, oregano Origanum onites, thyme Thymus vulgaris, coriander Coriandrum sativum, and astaxanthin (a carotenoid derived from the alga Haematococcus pluvialis). However, no inhibition was observed at 8 d postinfection. We investigated whether the early inhibition was caused by (1) the feed additives’ direct effect on the parasite and/or (2) stimulation of the Rainbow Trout immune responses. Garlic exhibited the highest in vitro killing capacity toward theronts (the infective stage of the parasite), followed by oregano, thyme, and astaxanthin, whereas coriander had no in vitro parasiticidal effect. Immune reactions were measured by plasma lysozyme activity in Rainbow Trout after feeding and by recording immune gene expression in trout leukocytes that were exposed to feed additives. Oregano-fed fish showed a significantly (P< 0.05) elevated plasma lysozyme activity. Oregano and thyme—and to a lesser extent, garlic, astaxanthin, and coriander—induced a significant regulation of innate and adaptive immune genes in vitro. The results suggest that the investigated additives inhibit early parasite development directly as well as indirectly (i.e., by elevating the host immune response). The parasitic ciliate Ichthyophthirius multifiliis is the causative agent of white spot disease (also known as ichthyophthiriosis) in freshwater fish worldwide (Buchmann et al. 2001; Matthews 2005; Dickerson 2006; Jørgensen 2016). The infection is characterized by macroscopically visible white spots in the epidermis and may be associated with heavy mortality and morbidity (Hines and Spira 1973; Dickerson 2006; Buchmann 2019). The life cycle of the parasite is direct but comprises several stages. The trophont is the feeding stage, residing in the epidermis of the skin. After a period of growth, it escapes the epidermis and appears as a freely swimming tomont, which subsequently encysts in a gelatinous capsule (the tomocyst stage) in which tomites (daughter cells) are produced by mitotic divisions. The tomites develop into infective theronts after leaving the cyst; they then seek and penetrate the external surfaces of the fish host (Matthews 2005; Dickerson 2006; Buchmann 2019). No vaccines are available for prevention of I. multifiliis infection, and control is currently based on repeated treatments using chemicals and management (Heinecke and Buchmann 2009). Malachite green and formalin are *Corresponding author: heidi.mathiessen@sund.ku.dk Received January 16, 2020; accepted September 9, 2020 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. North American Journal of Aquaculture © 2020 The Authors. North American Journal of Aquaculture published by Wiley Periodicals LLC on behalf of American Fisheries Society ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1002/naaq.10167

The parasitic ciliate Ichthyophthirius multifiliis is the causative agent of white spot disease (also known as ichthyophthiriosis) in freshwater fish worldwide Matthews 2005;Dickerson 2006;Jørgensen 2016). The infection is characterized by macroscopically visible white spots in the epidermis and may be associated with heavy mortality and morbidity (Hines and Spira 1973;Dickerson 2006;Buchmann 2019). The life cycle of the parasite is direct but comprises several stages. The trophont is the feeding stage, residing in the epidermis of the skin. After a period of growth, it escapes the epidermis and appears as a freely swimming tomont, which subsequently encysts in a gelatinous capsule (the tomocyst stage) in which tomites (daughter cells) are produced by mitotic divisions. The tomites develop into infective theronts after leaving the cyst; they then seek and penetrate the external surfaces of the fish host (Matthews 2005;Dickerson 2006;Buchmann 2019).
No vaccines are available for prevention of I. multifiliis infection, and control is currently based on repeated treatments using chemicals and management (Heinecke and Buchmann 2009). Malachite green and formalin are effective compounds, but due to their toxicity and carcinogenicity, farmers are replacing these compounds with less environmentally damaging chemicals, such as hydrogen peroxide (Rach et al. 2000;Rintamäki-Kinnunen et al. 2005), sodium percarbonate (Buchmann et al. 2003;Heinecke and Buchmann 2009), and peracetic acid (Rintamäki-Kinnunen et al. 2005;Meinelt et al. 2009;Straus and Meinelt 2009;Bruzio and Buchmann 2010). Replacement of formalin and malachite green by the alternative compounds has challenged fish farm operations. Several of the compounds mentioned can partly control I. multifiliis infections but are less successful than malachite green treatment (Lahnsteiner and Weismann 2007). Furthermore, treatments using chemicals such as formalin may damage the fish epidermis, leaving the fish susceptible to other pathogens (Buchmann et al. 2004).
Alternatives to this laborious practice may be found in functional feeding, as various feed additives have been shown to decrease the susceptibility of Rainbow Trout Oncorhynchus mykiss to I. multifiliis infection (Xueqin et al. 2012). Garlic Allium sativum as a feed additive was reported to increase growth, immunity, appetite, and the response to bacterial and fungal pathogens (Shakya and Labh 2014), including Aeromonas hydrophila (Mahfouz et al. 2009;Nya and Austin 2009) and monogenean infestations (Inoue et al. 2016). Oregano Origanum onites in feed has been shown to confer some resistance toward Lactococcus garvieae (Diler et al. 2017) and infections by the parasites Ichthyobodo salmonis and Trichodina truttae (Mizuno et al. 2018). Thyme Thymus vulgaris as a feed additive also conferred some resistance toward Streptococcus iniae infection (Yilmaz et al. 2013). Although the mechanisms have not been fully characterized, astaxanthin (a carotenoid derived from the alga Haematococcus pluvialis) was suggested to have immunestimulating effects (Thompson et al. 1995;Amar et al. 2004;Moghaddam et al. 2015;Liu et al. 2016). Astaxanthin feeding was also associated with reduced mortality due to Edwardsiella tarda infection (Kim et al. 2012).
The objective of this study was to evaluate the effect of feed additives, including garlic, oregano, thyme, and coriander Coriandrum sativum extracts and astaxanthin, on the susceptibility of Rainbow Trout to I. multifiliis. Fish susceptibility was assessed by an exposure to I. multifiliis infection following feeding, and the mechanisms behind any potential effect were further assessed by conducting in vitro assays. A previous study pointed to a direct effect on the parasite. A garlic extract killed theronts within 900 min at 62.5 mg/L and within 180 min at 312.5 mg/L, and the extract killed tomocysts at 570 mg/L (Buchmann et al. 2003). Compounds like those mentioned above may also have an immune-stimulating effect on the host (Xueqin et al. 2012;Diler et al. 2017;Mizuno et al. 2018). This was tested by the in vitro assays.
We conducted three experiments: (1) an in vivo feeding study that evaluated Rainbow Trout susceptibility to I. multifiliis infection, (2) an in vitro immobilization study that investigated the direct effects of feed additives on the parasite, and (3) an in vitro leucocyte stimulation study that assessed the immune-stimulating properties of the feed additives.

Fish
Rainbow Trout of the Rakkeby strain (Jutland, Denmark) were brought to the University of Copenhagen after being hatched and reared under pathogen-free conditions at the Bornholm Salmon Hatchery (Xueqin et al. 2012). Smaller fish were used for the in vivo feeding trial and larger fish were used for the in vitro leukocyte trial, as the latter study demanded large amounts of head kidney cells. All fish tanks were kept in the same room with a temperature of 16.5°C during acclimatization (14 d) and during subsequent feeding and infection.
In total, 98 fish (7 treatments × 2 replicates × 7 fish = 98 fish) were used for the in vivo study; mean body weight was 31.4 g (SD = 6.6), and mean total body length was 13 cm (SD = 2.5). For the in vitro leukocyte exposure trial, 20 fish with a mean body weight of 48 g (SD = 5.5) and a mean body length of 15.75 cm (SD = 1.25) were used for isolation of leukocytes for measurement of immune gene expression via quantitative PCR (qPCR).

Preparation of Additives
The herbs used in the experiment were dried oregano (Santa Maria, Brøndby, Denmark), thyme (Santa Maria), and coriander (Urtekram, Mariager, Denmark), while the garlic (ØGO') used was fresh. Oregano and thyme batches were ground using a mortar to a maximum particle size of 200 µm, while garlic was pressed by a garlic presser to release fluid. The astaxanthin used was purchased from Sigma-Aldrich (Catalog Number [CN] SML0982, Darmstadt, Germany).

Parasites for Infection
Trophonts of I. multifiliis originating from a Danish trout farm in Jutland were harvested from infected Rainbow Trout that were euthanized by an overdose (300 mg/ L) of anaesthetic (tricaine methanesulfonate ; CN A5040; Sigma-Aldrich). The fish were left in clean water for approximately 4 h, allowing trophonts to leave the epidermis. The trophonts transformed into tomonts and, via the tomocyst stage, into infective theronts (Heinecke and Buchmann 2009).

In Vivo Experiment
In vivo feeding trial.-The experiment investigating the effect of feed additives was conducted using seven groups in duplicate (Figure 1). The fish were acclimatized in a total of 14 fish tanks, each with seven fish, in a volume of 100 L (aerated municipal water at 16.5 ± 1.5°C) and containing internal Eheim Pickup 160 biofilters (Eldorado, Haderslev, Denmark) for 14 d under a 12-h light : 12-h dark cycle; a daily feeding regime corresponding to 2% of body weight at day 0 (EFICO Enviro 920 Advance; Bio-Mar, Brande, Denmark) was administered for 7 d. Treatment groups were subjected to seven feeding regimes as follows: (1) garlic added at 60 g/kg feed; (2) oregano at 60 g/kg feed; (3) thyme at 60 g/kg feed; (4) coriander at 60 g/ kg feed; (5) astaxanthin at 100 mg/kg feed; (6) astaxanthin at 200 mg/kg feed; and (7) control, with no feed additive. The concentration (60 g/kg) used for the herbal feed additives was determined based on a previous experiment regarding the fish's willingness to eat; however, the two concentrations of astaxanthin were chosen based on existing literature. To differentiate the treatment groups, fish were tagged by fin clipping according to the feed additive. On day 7, blood samples (from 7 treatments × 2 replicates × 3 fish·replicate −1 ·treatment −1 = 42 fish) were collected by caudal vein puncture using heparinized syringes. The blood samples were centrifuged at 3,000 × g for 10 min at 4°C; plasma was extracted and stored at −80°C (Alishahi and Buchmann 2006) until lysozyme analysis.
The remaining 56 fish were exposed to infection with I. multifiliis at approximately 2,100 parasites/fish (58,800 trophonts/tank) in two 100-L tanks (7 treatments × 4 fish/treatment = 28 fish/tank). Each of the two tanks contained four fish from each of the seven feeding regimes. At 5 and 8 d postinfection, the fish were anaesthetized and subjected to parasitological examination of the dorsal, caudal, and pelvic fins. The fish were monitored in the tanks until release of the parasite.
Feed preparation for in vivo trials.-One milliliter of rapeseed oil was used to bind the feed additives to the dried feed pellets (EFICO Enviro 920 Advance, 3 mm). The same oil was also applied to the control feed but without applying feed additives. The astaxanthin was mixed with the oil before adding it to the feed pellets for proper distribution. The herbal feed additives were added to the feed before the oil was applied.
Lysozyme activity measurement.-Lysozyme in plasma was measured by a spectrophotometric assay using Micrococcus lysodeikticus (CN M3770; Sigma-Aldrich) as substrate (Skov et al. 2018). A 10-µL volume of plasma was added to each well in a 96-well microplate in duplicate. A 190-µL quantity of a suspension containing M. lysodeikticus (0.2 mg/mL) in sodium phosphate buffer was added to the well. The plate was spun at 574 × g for 10 s, and the optical density was measured at 450 nm every 15 s for 30 min (Epoch Spectrophotometer; Holm & Halby, Brøndby, Denmark).

In Vitro Immobilization Experiment
Preparation of solutions for in vitro immobilization.-Stock solutions of the aqueous extract from the herbal feed additives were applied in a concentration of 1 g herb per 7 mL of distilled water. Distilled water was used as a control. Astaxanthin was prepared with a solvent (1 g of astaxanthin dissolved in 1 mL of dimethyl sulfoxide [DMSO]) according to the manufacturer's recommendation. A stock solution was prepared using 250 µL from the preparation mixed with 750 µL of distilled water. The control for astaxanthin was 250 µL of DMSO mixed with 750 µL of distilled water. Dilution series (twofold) based on the stock solutions mentioned above were prepared for in vitro exposure of theronts.
Immobilization assay.-The in vitro antiparasitic capacity of the feed additives was evaluated by an immobilization assay performed on I. multifiliis theronts. Dilution series (twofold) based on the aforementioned stock solutions were prepared, and a 50-µL volume of the dilutions was added to 50 µL of water containing a suspension of the parasite in a 24-well glass plate. Each well contained 5-10 theronts, and their motility was recorded every 15 min for a total of 60 min under the dissection microscope with subillumination (Leica). Theronts with a total loss of motility were considered inactivated.

In Vitro Leukocyte Stimulation Experiment
Cell isolation.-The cell isolation was performed according to Chettri et al. (2011). In brief, head kidneys from 20 Rainbow Trout were removed, homogenized, and subsequently centrifuged (400 × g for 30 min at 4°C) on Percoll (51%; CN GE17-0891-01; Sigma-Aldrich). Leukocytes were harvested, and the concentration of live cells was determined using the trypan blue exclusion method (1:1 ratio) and adjusted to 1.0 × 10 7 cells/mL using Fast-Read 102 slides (CN 630-1893; VWR, Søborg, Denmark). Subsequently, 100 µL of cell suspension were added to each well in a flat-bottom, 96-well microtiter plate (CN 655180; Fisher Scientific, Waltham, Massachusetts, USA) and incubated for 2 h at 15°C. Incubation was followed by removal of nonattached cells by washing, after which the cells were left to acclimatize overnight at 15°C.
Stimulation of cells.-The Rainbow Trout leukocytes were stimulated with a high (70 µg/mL) or a low (35 µg/ mL) feed additive dosage for 1 h; the feed additives were then replaced with 100 µL of lysis buffer containing 2-mercaptoethanol (CN M3148; Sigma-Aldrich; Chettri et al.

2010, 2011). A lipopolysaccharide preparation from
Escherichia coli 0111:B4 (100 + 400 µg/mL; CN L2630; Sigma-Aldrich) was applied as a positive control since it has a well-documented ability to induce a response in this in vitro leukocyte stimulation assay (Chettri et al. 2011).
Real-time quantitative PCR.-Total RNA from leukocytes was extracted using the GenElute Mammalian Total RNA Miniprep Kit (CN RTN350; Sigma-Aldrich) and treated with DNAse (CN AMPD1; Sigma-Aldrich) after extraction to remove potential DNA contamination. The RNA concentration was measured spectrophotometrically using a NanoDrop spectrophotometer (Saveen & Werner ApS, Jyllinge, Denmark), and the samples were kept at −80°C. The RNA was subjected to reverse transcription in a BioRad T100 Thermal Cycler (Thermo Fisher Scientific, Roskilde, Denmark) by means of TaqMan Reverse Transcription Reagents (CN N8080127; Thermo Fisher Scientific) using random hexamers in a 40-µL reaction volume containing a maximum of 1,000 ng of RNA. The reactions occurred under the following conditions: 10 min at 25°C, 60 min at 37°C, and 5 min at 95°C. Complementary DNA was diluted four times by adding 120 µL of RNase-free water (CN 10977-035 RNase; Thermo Fisher Scientific) and was subsequently stored at 5°C. To perform real-time qPCR assays, an AriaMx Real Time PCR System (AH Diagnostics, Tilst, Denmark) was used. Real-time qPCR was conducted using the Brilliant III Ultra-Fast Master Mix Kit (CN 600880; AH Diagnostics), 2.5 µL of complementary DNA template, 0.8-µM primers, and 0.4-µM probes in a 12.5-µL reaction. The final concentration of MgCl 2 was 5.5 mM. Primers and probes are listed in Table  S1 (available in the Supplement in the online version of this article). The PCR conditions were set to one cycle for 10 min at 94°C followed by 40 cycles at 94°C for 10 s and 60°C for 15 s. Each run included wells without template and wells with template but without reverse transcriptase. The genes chosen in this study comprised genes that regulate both innate and adaptive immunity.
Genes encoding an innate immune response were proinflammatory cytokines (interleukin [IL] 1 beta [IL-1β], FIGURE 1. Design of the in vivo trial with Rainbow Trout. On day 0, 98 fish were divided into seven duplicated groups (seven feeding regimes in two tanks with seven fish each). The seven duplicates were tagged by individual fin clipping. On day 7, three fish from each tank (3 × 7 × 2 = 42) were used for blood sampling. The remaining fish were pooled into two tanks (each tank infected with 58,800 Ichthyophthirius multifiliis trophonts), after which the fish were fed without additives. At 5 and 8 d postinfection (dpi), the fish were subjected to parasitological examination for I. multifiliis.

Calculations and Statistical Analysis
Data were analyzed using Microsoft Office Excel (Microsoft, Redmond, Washington, USA) and GraphPad Prism 7 and 8 (GraphPad Software, San Diego, California, USA). Data from the susceptibility trial followed a normal distribution, which allowed application of a oneway ANOVA with Dunnett's multiple comparison test to compare the treatments with the control. Furthermore, a two-way ANOVA was conducted to investigate whether there were differences between the treatments or between the two time points (significance was assessed at P < 0.05). Lysozyme activity in plasma was also distributed according to a normal distribution; in this case, the data were analyzed using a Student's t-test (P < 0.05). The Mantel-Cox test and Gehan-Breslow-Wilcoxon test were performed to evaluate the significant difference between parasite immobilization curves (P < 0.05). Relative immune gene expression was analyzed using the simplified 2 −ΔΔCq method, where Cq is the cycle quantification value (Livak and Schmittgen 2001), as all of the qPCR assays had efficiencies within 100 ± 5% (Schmittgen and Livak 2008). Student's t-test was applied for evaluating the effect of the feed additives on immune gene expression; only cases having both a P-value <0.05 and a fold change of at least 2 were considered significant. In some cases, too few Cq values were obtained in one of the groups compared to use the quantitative 2 −ΔΔCq method. In these cases, the gene expression regulation was assessed by a qualitative approach using the presence/absence of Cq values and the nonparametric Mann-Whitney test.
Ethics and legislation.-The experiments were performed under Animal Experimental Inspectorate License 2013-15-2934-00794 and in accordance with the general welfare guidelines at the University of Copenhagen.

In Vivo Experiment
The total number of parasites and mean intensity recorded were higher at 8 d postinfection compared to 5 d postinfection (Figure 2). The highest number of parasites was recorded in the control fish (both 5 and 8 d postinfection), and the lowest number was found in the groups receiving astaxanthin at 100 mg/kg feed. The parasite escaped the epidermis at 14 d postinfection. At 5 d postinfection, all of the feed additives resulted in a significantly lower parasite burden compared to the control fish. No significant difference between the different feed additives was found (P = 0.5579). At 8 d postinfection, the infection intensities did not differ significantly between fish that received feed additives and the control group. There was a significant difference (P < 0.0001) between the infection levels at 5 and 8 d postinfection.
The plasma lysozyme activity in fish receiving oregano was significantly (P = 0.0045, P = 0.0406) elevated compared to that in the control fish for both laboratory experiments performed. Plasma lysozyme activity in fish receiving the other feed additives did not differ from that of controls (Table 1).

In Vitro Immobilization Experiment
All herbal feed additives except coriander successfully immobilized theronts. The garlic preparation was the most effective, as we found a significantly (P < 0.0001) lower mobility of the parasite in all concentrations until a 1:1,024 dilution (P = 0.1195). Oregano resulted in a significantly (P < 0.0001 and P = 0.0012) higher immobilization until a dilution of 1:256 (P = 0.7032). Thyme also affected the parasite, but we merely found a significant (P < 0.0001) difference between thyme and the control until a 1:32 dilution (P = 0.9892). We noted a higher theront immobilization in astaxanthin compared to the corresponding DMSO control group, but only an astaxanthin dilution of 1:8 (P = 0.0013) significantly increased immobilization of the parasite (see Table S2 for exact data).
When leukocytes were exposed to a low dosage of astaxanthin, it resulted in a significant downregulation of the SAA gene (P = 0.0039). Exposure to a high dose of astaxanthin resulted in upregulation of the gene encoding IgT (P = 0.023) and a downregulation of the gene encoding precerebellin (P = 0.04). Exposure to a FIGURE 2. Intensity of the Ichthyophthirius multifiliis infection in Rainbow Trout that received feed additives (dpi = d postinfection). The intensity was calculated as the mean number of parasites per fish (all fish infected). Error bars indicate the SE. Stars indicate significant differences (P < 0.05) in comparison with the control at the given time point (dpi). low concentration of lipopolysaccharide led to a significant upregulation of the genes encoding IL-8 (P = 0.0125), TNF-α (P = 0.004), and IL-4/13A (P = 0.045). The high concentration strongly stimulated the expression of immune genes encoding IL-1β (P = 0.01), IL-6 (P = 0.04), IL-8 (P = 0.000997), IL-10 (P = 0.001), TNFα (P = 0.001), and lysozyme (P = 0.02).

DISCUSSION
Control of infectious diseases in fish aquaculture should optimally rely on prophylactic principles, but in many cases the treatment of infected fish is necessary. Availability of effective chemotherapeutants is limited, and their usage raises toxicity and environmental concerns. Sustainable control agents, including herbal extracts from naturally occurring plants, may be considered relevant in this regard. A series of studies has indicated that these plant products have some effect on both bacterial and parasitic infections (Mahfouz et al. 2009;Nya and Austin 2009;Zheng et al. 2009;Amar et al. 2012;Kim et al. 2012;Yilmaz et al. 2013;Breyer et al. 2015;Inoue et al. 2016;Diler et al. 2017;Jørgensen 2017;Mizuno et al. 2018;Skov et al. 2018).

In Vivo Experiment
In the present study, we conducted an in vivo feeding trial using extracts from garlic, oregano, thyme, and coriander along with astaxanthin. This resulted in a reduced parasite FIGURE 3. Regulation of immune genes in Rainbow Trout leukocytes exposed in vitro to three different feed additives (astaxanthin, coriander, and garlic; high dose = 70 µg/mL; low dose = 35 µg/mL). Stars indicate significant differences (P < 0.05) from the control. Abbreviations are defined in Methods.

EFFECT OF FEED ADDITIVES ON RESPONSE TO ICHTHYOPTHIRIUS MULTIFILIIS
burden at 5 d postinfection, suggesting that the garlic, oregano, thyme, and coriander extracts and astaxanthin potentially inhibit early development of I. multifiliis trophonts in Rainbow Trout skin, making them harder to detect under the microscope. The effect of the feed additives did not last and was of short duration at best, as no significant difference between treatments was found at 8 d postinfection. Theoretically, the observations could be explained by a direct antiparasitic effect, an immune-stimulating effect, or both. The results indicate an acute and short-lived effect.
Aqueous extracts of garlic, oregano, and thyme along with astaxanthin showed a direct parasiticidal effect on theronts, which could explain our observations, at least partly. In addition, Rainbow Trout that were fed oregano showed a higher lysozyme activity in plasma, which indicated an effect on the immune system of the fish. In the in vitro leukocyte stimulation experiment, we observed a marked effect of most feed additives, which suggests that activated immune reactions on the fish's body surface may inhibit the early development of trophonts after theront invasion. In this context, it is noteworthy that a corresponding delay of Ichthyobodo and Trichodina infections in salmonids was achieved by a similar administration of oregano (Mizuno et al. 2018). At 5 d postinfection in our study, the parasites were rather small and difficult to discern, which may have led to an underestimation of their numbers. If the size of the parasites was even smaller after hosts were fed the additives, the parasites could go FIGURE 4. Regulation of immune genes in Rainbow Trout leukocytes exposed in vitro to three different feed additives (lipopolysaccharide [LPS], oregano, and thyme; high dose = 70 µg/mL; low dose = 35 µg/mL). Stars indicate significant differences (P < 0.05) from the control. Abbreviations are defined in Methods. 8 unnoticed. The suggested slower development was supported by an extended trophont maturation period in the skin of the host, as the parasite did not escape the epidermis until 14 d postinfection. In control fish, the parasites escaped the epidermis after 7-11 d. The fish were fed for a total of 7 d, although other published studies have chosen feeding periods spanning from 14 to 90 d (Mahfouz et al. 2009;Nya and Austin 2009;Zheng et al. 2009;Yilmaz et al. 2013;Breyer et al. 2015;Inoue et al. 2016;Diler et al. 2017;Mizuno et al. 2018). In this way, we could measure the direct effect on the immune system before any adaptation was induced. Components from fish feed accumulate over time (Montero et al. 2005), and a reduced effect of feed additives has been observed over time (Breyer et al. 2015;Mizuno et al. 2018).

In Vitro Immobilization Experiment
Garlic, oregano, thyme, and astaxanthin all showed an in vitro capacity for immobilization of I. multifiliis, whereas coriander had no in vitro effect on the parasite. The active compounds in the herbal feed additives have not been clearly defined but are likely to differ from source to source.

In Vitro Leukocyte Stimulation Experiment
Astaxanthin was previously shown to activate immune parameters, such as complement, phagocytic activity (Amar et al. 2004), and neutrophil adherence (Kim et al. 2012), but not lysozyme. Since garlic and thyme components may also stimulate the host immune system (Mahfouz et al. 2009;Nya and Austin 2009;Gulec et al. 2013), we cannot exclude the possibility that the parasite inhibition effect observed at 5 d postinfection could be a result of the immunostimulatory effects of the feed additives.
Our in vitro stimulation of leukocytes showed that different feed additives may affect immune reactions differently. It may be hypothesized that thyme and oregano stimulated the immune genes encoding effector molecules, and depression of genes facilitating an inflammatory response suggests that I. multifiliis may profit from inflammation-associated skin processes but may also be adversely affected by effector molecules. Oregano, garlic, and thyme successfully induced upregulation of several genes involved in the adaptive immune response. Garlic at a low dose led to significant upregulation of the C3 gene, supporting previous observations (Breyer et al. 2015). A high garlic dose increased the expression of genes encoding IL-6 and IL-4/13A, whereas other genes were unregulated. Astaxanthin administered in feed has been demonstrated to increase survival in virus-infected fish and to regulate different immune components, such as complement, phagocytic activity (Amar et al. 2004), and neutrophil adherence (Kim et al. 2012). However, during in vitro stimulation of leukocytes, we did not record an effect on genes encoding C3 and C5-regardless of whether the dosage was low or high. Coriander was associated with an inhibitory effect on trophont development at 5 d postinfection but showed no direct parasiticidal effect or any broad influence on immune gene expression in vitro, suggesting that feed additives may affect other immune mechanisms that were not addressed in the present study.
In conclusion, the use of feed containing extracts of garlic, oregano, thyme, and coriander resulted in a significantly lower detection of parasites at 5 d postinfection after administration of additives was discontinued. No significant effect was seen at 8 d postinfection despite a generally lower infection intensity. This suggests that additives exerted an inhibitory effect on early trophont growth and that the effects of the feed additives are acute or short-lived. Host immune factors may influence the development of the early trophont stage in the epidermis of the fish. It is therefore noteworthy that feed containing oregano was associated with an increased lysozyme activity in Rainbow Trout plasma. Most of the additives had a direct parasiticidal effect when tested in vitro. Garlic showed the highest immobilizing effect on theronts, followed by oregano, thyme, and astaxanthin. Coriander had no direct effect on the parasite. In vitro stimulation of Rainbow Trout leukocytes with the herbal extracts resulted in significant regulation of several immune genes that may have influenced the parasite infection. The inhibition of early trophont development along with stimulation of the immune response may be beneficial for fish farmers and the aquaculture industry. However, further studies should elucidate the antiparasitic mechanisms in the fish and optimize the feed administration before use of feed additives on a wider scale.