Effect of repeated exposure to AQUI‐S® on the viability and growth of Neoparamoeba perurans

Abstract There have been recent efforts amongst immunologists to develop approaches for following individual fish during challenges with viral and bacterial pathogens. This study contributes to assessing the feasibility of using such approaches to study amoebic gill disease (AGD). Neoparamoeba perurans, agent of AGD, has been responsible for widespread economic and fish loss in salmonid aquaculture. With the emergence of AGD in Europe, research into infection dynamics and host response has increased. This study investigated the effect of repeat exposure to anaesthesia, a necessary requirement when following disease progression in individual fish, on N. perurans. In vitro cultures of N. perurans were exposed every 4 days over a 28‐day period to AQUI‐S® (isoeugenol), a popular anaesthetic choice for AGD challenges, at a concentration and duration required to sedate post‐smolt salmonids. Population growth was measured by sequential counts of amoeba over the period, while viability of non‐attached amoeba in the culture was assessed with a vital stain. AQUI‐S® was found to be a suitable choice for in vivo ectoparasitic challenges with N. perurans during which repetitive anaesthesia is required for analysis of disease progression.


| INTRODUCTION
With the continued expansion of the global aquaculture industry (FAO, 2016), and related research on fish, there is a need for refinement in experimental approaches, including analyses of in vivo immune responses. A limited number of studies have been undertaken in developing methodologies such as individual monitoring (Collet et al., 2015;Urquhart et al., 2016). Benefits of such approaches include a reduction in the number of animals required for challenge experiments and higher quality data output, with reduced infection and response variability (Collet et al., 2015). While previous individual monitoring of fish following disease challenges has focused upon viral or bacterial pathogens, attention must also turn to parasite studies in view of serious parasite issues currently affecting aquaculture, for example sea lice and amoebic gill disease (AGD) affecting salmonid farming (Aaen, Helgesen, Bakke, Kaur, & Horsberg, 2015;Oldham, Rodger, & Nowak, 2016). To assess the suitability of this methodology for ectoparasites, Neoparamoeba perurans, the amoeboid aetiological agent of AGD, was selected as a model due to its recent emergence as a serious pathogenic threat to salmon aquaculture across northern Europe.
The first occasion of AGD as an epizootic was observed in an Atlantic salmon, Salmo salar, and rainbow trout, Oncorhynchus mykiss, sea farm located in east Tasmania, during the summer of 1984-85 (Munday, 1986). It was suggested that the aetiological agent of AGD could be classified as the normally free-living Neoparamoeba sp.
The recognized method of obtaining pathogenic samples of N. perurans is to collect specimens from the gills of infected fish at the point of lethal sampling (Morrison, Crosbie, & Nowak, 2004), which involves at least one exposure to fish anaesthetic. Recent work from Shijie, Adams, Nowak, and Crosbie (2016) has demonstrated that a single exposure to anaesthetics containing eugenol did not inhibit population growth or attachment abilities of cultured N. perurans.
To develop a non-lethal sampling approach requires repeated anaesthesia of fish, which in turn, for an ectoparasitic disease model such as AGD, also results in repeated anaesthesia of the pathogen. Therefore, the first step in developing a non-lethal challenge model for AGD is to examine the effect of repeat exposure of N. perurans to anaesthesia. AQUI-S â is a gel-like anaesthetic that was first developed in New Zealand in 1996. Inspired by the anaesthetic capabilities of clove oil (eugenol), AQUI-S â contains as active ingredient 50% isoeugenol (not present in natural clove oil) and 50% emulsifier polysorbate 80 (Javahery & Moradlu, 2012). It is the only registered foodgrade anaesthetic with zero withdrawal time in Australia, Chile, Costa Rica, Honduras, Korea, and New Zealand; (AQUI-S â , 2015). As of 2014, AQUI-S â has also been approved in Norway for sedation and anaesthesia of Atlantic salmon and rainbow trout prior to and during handling events, and in live fish transport (Kolarevic & Terjesen, 2014). AQUI-S â was therefore selected due to the popularity of use in countries most severely affected with AGD, alongside recent findings of no short-term impacts upon attachment or viability of N. perurans after single exposure (Shijie et al., 2016). This is the first paper to report upon the repeated exposure of N. perurans to fish anaesthetics and to describe any adverse effects found on this aquaculture ectoparasite.

| Anaesthetic exposures
Six replicate flasks were used for each treatment: AQUI-S â (isoeugenol) (AQUI-S New Zealand Ltd.) and control (35 ppt 0.22-lm-filtered sea water). Anaesthetic treatment flasks were exposed to the same concentrations and durations required to anaesthetize postsmolt salmonids to Stage 4 anaesthesia, AQUI-S â at 17 mg/L for 20 min. Due to the small amounts of anaesthetic required, at each time point a stock solution was freshly prepared. An appropriate volume was dissolved in 35 ppt 0.22-lm-filtered sea water, pipetted into the 7 ml seawater overlay to obtain the required concentration, and the flasks agitated to ensure an even distribution of the anaesthetic across the culture.
After the predetermined exposure duration, the overlay containing the anaesthetic and the floating-form amoeba was transferred to 15-ml tubes, and the attached amoeba remaining in the flasks were rinsed once with filtered sea water and 6.9 ml of filtered sea water was provided to restore the overlay. Sea water used in the rinse was discarded. The original overlay containing anaesthetic and floatingform amoeba was centrifuged at 10739g for 10 min-it should be noted that this was additional contact time for the suspended amoeba population with AQUI-S â -the supernatant removed and the amoebae present in the pellet transferred to a 1.5-ml Eppendorf tube containing 1 ml filtered sea water. The amoeba suspensions were centrifuged at 113379g for 1 min followed by the removal of the supernatant and resuspension of amoeba in 1 ml filtered sea water. Amoebae were washed a further time as above then returned to their respective flasks after being resuspended in 100 ll filtered sea water, returning the total overlay volume to 7 ml. Preliminary work utilizing the vital stain Neutral Red (Sigma-Aldrich, N7005) ascertained that the speed of centrifugation and transfer had no negative effect upon the morphology or viability of the amoeba (albeit not amoebae exposed to anaesthetic) and that the speed and duration of centrifugation were sufficient to pellet the suspended amoeba from the suspension (data not shown). This process was also carried out for all control flasks at each time point. All flasks were returned to 13°C until the next scheduled exposure. Flasks were treated with anaesthetic every 4 days for a 28-day period.

| Suspended amoebae
For the viability assessment of amoebae in suspension, a 200 ll aliquot of the seawater overlay was removed from each flask prior to each anaesthetic exposure time point; the seawater overlay of each flask was gently agitated for approximately 5 s and the flask rotated to an upright position so that the overlay pooled into the bottom left corner of the flask to ensure the aliquot obtained was representative of the total overlay. This aliquot was then transferred to a 1.5-ml Eppendorf tube containing 4 ll of the vital stain Neutral Red.
The tubes were kept at 13°C for 40 min to allow the amoebae to take up the stain. The cell suspensions were the centrifuged down for 1 min at 113379g and the supernatant removed. The amoeba pellets were next resuspended in 100 ll filtered sea water and 10 ll of this suspension transferred to a well of a flat-bottomed 96-well plate containing 90 ll sea water. Amoebae were left to settle in the wells for 40 min and then were assessed for their viability with an inverted microscope at 920 magnification. Viable amoebae had a diverse morphology as well as obvious dye inclusions, while nonamoebae had no visible dye inclusions and a spherical morphology ( Figure 1). Viable and non-viable amoebae were counted.

| MS-222 and metomidate flasks
The study also sought to assess the population growth and viability of N. perurans following repeated doses of powder-based fish anaesthetics metomidate (12.5 mg/L) (AquaCalm TM Western Chemical Inc.) and MS-222 (80 mg/L) (Sigma-Aldrich). These treatments were carried out following the same methodology as detailed above for the AQUI-S â flasks. Due to the short exposure duration required for in vivo sedation, three and five minutes, respectively, the additional ten-minute exposure of the suspended amoeba during the 10739g centrifugation of the anaesthetic-containing overlay and wash stage renders the total exposure time for these suspended populations 94 and 93 longer than required for a nonlethal sampling procedure, and thus, these results should be interpreted with caution.

| Statistical analysis
Population growth data were analysed with the statistical software package R (R Core Team, 2016). Total population count was con-

| AQUI-S â non-viable population analysis
No statistical difference was seen in non-viable amoebae percentage, in relation to the total amoebae population, in the AQUI-S â treatment when compared to the control, with the exception of day 20 were no significant differences found at Day 0, with respect to attached amoebae numbers, between any of the treatments and control ( Figure 3). All amoebae were left to adhere overnight before treatment at Day 0, at the same temperature, in the same incubator in the agar base and seawater overlay derived from the same stocks. Therefore, it is not possible to suggest a lack of adherence by the amoeba due to culture differences in these treatments.

| MS-222 and metomidate non-viable population analysis
After the second dose of anaesthetics (day 8), the percentage of non-viable amoebae in both total and suspended populations in MS-222 and metomidate flasks were significantly higher (p < .001) when compared to the control, (Figures 4 and 5). In the suspended populations ( Figure 5), there was a sustained increase in percentage of non-viable amoebae throughout the rest of the experiment for both these treatments, in which all timepoints remained significantly different (p < .001) to the control. This trend is also seen for the non- offers no hypothesis as to why this effect may be seen. During in vivo challenge experiments, an artificially elongated duration of attachment, during which parasitic amoeba could theoretically spend more time colonizing the gill substrate (Wiik-Nielsen et al., 2016) than completing the natural emigration to the surrounding sea water, may lead to an increased level of disease progression and therefore an elevated immune response, which may not be comparable to the speed of disease progression found in the field. Nonetheless, with reported loss of virulence seen in cultured N. perurans possibly due to lack of attachment to gills (Bridle, Davenport, Crosbie, Polinski, & Nowak, 2015), increased attachment due to the use of isoeugenolbased anaesthetics may help mitigate this problem, if similar attachment processes are involved.
In this study, the amoebae were classed as "non-viable" primarily due to the lack of uptake of the Neutral Red vital stain (Repetto, Del Peso, & Zurita, 2008), but morphology was also taken into consideration. Amoebae in which no stain was seen all held the same spherical morphology (Figure 1), characteristic of in vitro cultures with a suboptimal subculturing schedule, suggesting this morphology is a response to overcrowding, lack of nutrients or environmental stressors (Lima, Taylor, & Cook, 2017;Wiik-Nielsen et al., 2016). As the cultures in this study were washed regularly at 4-day intervals, and percentage of non-viable amoebae were higher in cultures with lower amoebae numbers (Figures 2 and 4), it is unlikely that the However, when comparing percentages of non-viable amoebae as part of the suspended population, significant differences were found between the control and AQUI-S â populations at days 4, 20 and 28 ( Figure 5), but as discussed above this reflects greater numbers of amoebae remaining attached in the AQUI-S â flasks, resulting in non-viable amoebae forming a higher percentage of total suspended cells.
After a single exposure timepoint MS-222, metomidate-based anaesthetics seem to have a strong inhibitory effect upon both the growth of attached in vitro N. perurans cultures and a detrimental effect on viability of floating-form amoeba after repeated exposure.
As previously highlighted, the suspended amoebae in metomidate and MS-222 flasks were in contact with their anaesthetics for substantially longer than required for in vivo anaesthetization. This increase in exposure time must be taken into consideration when evaluating the outcomes of the attached amoeba growth, as both populations are interdependent (Crosbie et al., 2012). However, it should also be considered that during an individually monitored challenge, fish are sampled with an in-tank anaesthesia methodology (Collet et al., 2015), wherein the suspended amoeba will remain in contact with the anaesthetic while Stage 4 anaesthetized fish are netted out, processed and placed in a smaller recovery tank followed by the initial tank being drained with a flow-through system. Any amoeba which remains in this tank after draining and refilling will have also been exposed to whichever anaesthetic was used for a longer duration that initially required for Stage 4 anaesthesia. Previous studies investigating adherence behaviour of N. perurans have shown high-density colonization of aquarium surfaces, highlighting their potential as areas for attachment and replication (Rolin, Graham, McCarthy, Martin, & Matejusova, 2016) and may therefore act as an additional source of infection over time; however, the impact of amoebae shed from gills in reinfection and disease progression over the challenge, if any, is not known. It could be argued that this prolonged exposure of anaesthetics to the suspended amoeba population may even be more representative of the environmental conditions during non-lethal sampling.
Metomidate is able to block the synthesis of plasma cortisol by inhibiting the mitochondrial cytochrome P 450 -dependent enzymes required to catalyse the glucocorticoid (Small, 2003), an effect which has also been reported in fish treated with MS-222 (Chevion, Stegeman, Peisach, & Blumberg, 1977;Fabacher, 1982). Akinrotimi, Gabriel, and Orokotan (2013) have shown that metomidate also has the dose-dependent ability to impair the activities of plasma enzymes such as transaminases in the African sharptooth catfish, Clarias gariepinus, with the highest level of impairment seen at 12 mg/L. MS-222 has been shown to inhibit the growth of Gram-negative bacteria (Fedewa & Lindell, 2005); however, the concentrations (5,000-200 mg/L) used in this latter study were far higher than those used in vivo. Similar inhibitory effects on N. perurans p450 pathway and transaminases may have played a role in the suppression of population growth, attachment and viability of amoebae in the flasks treated with these anaesthetics (Figures 2, 3, 4 and 5).
Such impacts may not be seen during in vivo challenges due to shorter exposure periods of fish to anaesthetics; therefore, the suitability of MS-222 and metomidate as anaesthetics for non-lethal sampling AGD challenges should be investigated further, with more efficient cleaning of the suspended population, utilizing faster spin times or filtering methods to obtain more appropriate exposure times.
In conclusion, this study illustrates the importance of selecting an appropriate anaesthetic when working with ectoparasites. Isoeugenol-based, specifically AQUI-S â , anaesthetics are suitable for both harvesting and repeated exposure in vivo and in vitro for work with the ectoparasite N. perurans.