Efficacy of BioRas? Balance (an enzyme product) to break down hydrogen peroxide following routine treatment applications in aquaculture

Application of water treatments and therapeutants in recirculating aquaculture system (RAS) is complicated by potentially detrimental secondary impacts on critical nitrifying bacterial populations present in biofilters (Noble & Summerfelt, 1996), leading to the accumulation of total ammonia nitrogen (TAN) and/or nitrite nitrogen in the absence of effective oxidation. Hydrogen peroxide (H2O2), while U.S. Food and Drug Administration approved for use in aquaculture under specific conditions, and while considered environmentally benign due to the formation of nontoxic decomposition end products (Block, 1991), is known to negatively impact biofiltration when applied at therapeutic concentrations in RAS (Fredricks, 2015; Schwartz et al., 2000). Considering this, previous research has examined the potential for low-dose H2O2 applications in RAS as a means to avoid biofilter compromise (Bögner et al., 2021; Møller et al., 2010; Pedersen et al., 2012; Pedersen & Pedersen, 2012); however, efficacy against target waterborne pathogens using the low-dose approach has not been adequately assessed. The ability for RAS producers to use therapeutic or disinfection-level H2O2 concentrations in culture tank static bath treatments, followed by decomposition of H2O2 prior to resumption of recirculating water flow through biofilters, would be advantageous. Therefore, the aim of this study was to assess H2O2 decomposition using a novel enzymatic product following static bath applications of H2O2 at both therapeutic- and disinfection-level concentrations. The study findings are presented in this Short Communication. Two single-day experiments were conducted at The Conservation Fund's Freshwater Institute (Shepherdstown, WV, USA) to assess the effectiveness of a new catalase enzyme product, BioRas® Balance, in the decomposition of H2O2 into H2O and O2, applied as a static bath at either therapeutic- or disinfection-level concentrations. The overall goal was to validate that BioRas® Balance breaks down H2O2 in fish culture tank water, such that, in theory, treated tank water would not pose a risk to nitrifying bacteria in RAS biofilters upon resumption of recirculating water flow. The first trial (Trial 1) employed simulated static bath treatments for bacterial gill disease in replicated culture tanks, with the live fish present, followed by the application of BioRas® Balance in a range of concentrations to assess dose–response decomposition of H2O2 over time. Trial 2 was a similar dose–response experiment but employed disinfection-level H2O2 concentration with no live fish present. Trial 1: Diploid Atlantic salmon Salmo salar were received as fertilized eggs and hatched onsite using a temperature-controlled incubation system and then cultured to market-size in freshwater RAS. To begin the study, 216 fish (5 kg mean weight) were stocked randomly into 12 replicated partial recirculating aquaculture systems (PRAS). Each PRAS consisted of a 5.0 m3 (1316 gallon) “Cornell style” dual-drain tank, a gas-conditioning column with counter-current forced-air ventilation and a low head oxygenator with a sump (Figure 1). A 0.4 horsepower pump, located in the sidewall box of each tank, continuously recirculated 379 L/min (100 gpm) of water. Prior to the start of the experiment, salmon were fed to satiation with a commercial salmon diet (8 mm EWOS Dynamic Red) and allowed to acclimate for 3 days. Biomass densities at the start of the experiment were approximately 20 kg/m3. PRAS was operated at a 90% recirculation rate (on a flow basis, i.e., 10% of recirculating water consisted of new, “makeup” water), with a culture tank hydraulic retention time of 15 min, and under these conditions, typical water-quality parameter values for both studies were as follows: temperature (14°C), pH (7.8), alkalinity (277 mg/L), CO2 (4.5 mg/L), TAN (0.2 mg/L) and total suspended solids (1.0 mg/L). Following acclimation, 60-min static bath tank treatments were carried out, simulating therapeutic treatment for bacterial gill disease as specified by FDA-approved guidelines, i.e., a target dose of 100 mg/L H2O2 (35% Perox-Aid; Syndel) applied for approximately 60 min. Hydrogen peroxide was allowed to mix within a tank for 10 min, after which a water sample was taken and analysed to verify H2O2 concentration. For all assessments of H2O2 concentration, a 25 ml sample was collected directly from the tank, after which it was (i) diluted with deionized water, and (ii) a peroxide Vacu-vial (k-5543; CHEMetrics) was used to extract the sample. Each vial came prefilled with an acidic solution and used ferric thiocyanate chemistry principles. In the acidic solution, hydrogen peroxide oxidizes ferrous iron, causing it to react with ammonium thiocyanate to form ferric thiocyanate, a red-orange coloured complex. The depth of the colour is proportional to the concentration of the peroxide. After the Vacu-vial was filled with the sample, it was inverted several times to ensure proper mixing. After 1 min, the Vacu-vial exterior was dried and cleaned with a Kimwipe® and inserted into the preprogrammed (520 nm wavelength) V-2000 photometer (CHEMetrics) that directly provided H2O2 concentration values. Prior to BioRas® Balance addition, initial and final H2O2 concentrations during the static bath treatment (i.e. concentrations at 0 min and 60 min, respectively) were quantified (i) to document that target concentration H2O2 was achieved and (ii) to assess whether decomposition had occurred during the bath treatment phase. Initial and final H2O2 concentrations during static bath treatment were 97.7 ± 3.1 mg/L and 96.0 ± 5.7 mg/L, respectively. BioRas® Balance was then added to the culture tanks to create concentrations of 1.5, 2.5, 5.0 and 10.0 mg/L (target concentrations were selected based on previously unpublished research); each tank was randomly assigned one of these BioRas® Balance concentrations, for a total of three replicates per treatment concentration. Following the addition of the catalase enzyme, water samples were collected and H2O2 concentrations analysed at 10-min intervals until H2O2 concentration within a given tank was considered negligible (i.e. less than 0.5 mg/L). Enzymatic decomposition of H2O2 followed a dose–response relationship (Figure 2, top), catalysing a first-order decay reaction as reported previously (Arvin & Pedersen, 2014; Pedersen et al., 2006); first-order decay constants and kinetic reaction formulae for each BioRas® Balance application concentration were calculated (Figure 2, bottom) using least-squares regression of the log-transformed H2O2 concentration data. Trial 2: The second trial was carried out to simulate a typical disinfection event using a higher concentration of H2O2 (i.e., 250 mg/L initial target concentration). Dosing of H2O2 and BioRas® Balance, and H2O2 concentrations assessments, were carried out in an identical manner to the procedures used in Trial 1. The only major differences with Trial 2 were that (i) water continued to recirculate within the PRAS (as opposed to a culture tank static bath) and (ii) no fish were present in the culture tank, as H2O2 concentrations were high enough to likely cause injury and mortality to any fish present. Initial and final H2O2 concentrations during the precatalase, disinfection phase were 261.0 ± 15.3 mg/L and 244.9 ± 14.1 mg/L, respectively. Once again, enzymatic decomposition of H2O2 followed a clear dose–response relationship (Figure 3, top), and decay constants and kinetic reaction formulae for each BioRas® Balance dose concentration were calculated (Figure 3, bottom). In principle, reactant half-life values for first-order reactions are independent of initial concentration, and in the current study, half-life values for the initial H2O2 concentrations of 100 and 250 mg/L were similar at each BioRas® Balance concentration (Figure 4). Experimental conditions with the highest BioRas® Balance concentration of 10 mg/L at both 100 mg/L and 250 mg/L H2O2 initial concentrations resulted in H2O2 degradation reactions so fast that only 4 and 3 later-stage reaction data points were obtained respectively. For such high BioRas® Balance dosing, a better first-order model fitting would need additional time-dependent H2O2 concentration data taken at shorter time points (than 10 min). Overall, the results of these experiments are encouraging for RAS farmers, who may require utilizing H2O2 at concentrations that would normally compromise biofilter nitrification processes. Under the conditions of the studies, BioRas® Balance was effective at quickly reducing H2O2 to safe concentrations, in either poststatic bath therapeutic application or postsystem disinfection scenarios. In that latter case, H2O2 disinfection of large-scale RAS likely requires significant volumes of H2O2 to reach disinfection concentrations, after which the farmer is left with an enormous amount of H2O2-laden water. If discharge to high organic content areas, such as constructed wetlands or wood chip compartments, is not an option, then waiting for H2O2 to decay naturally can potentially take a significant amount of time, depending largely on the concentration of microbial action within biofilms on surfaces within the RAS (Schmidt et al., 2006). Without the addition of some form of catalase, therefore, a farmer must slowly dilute the chemical into their effluent, putting them at risk for discharge infraction and increasing the time it takes to restock production systems. Farmers that use H2O2 at concentrations of 250 mg/L and above to disinfect systems and equipment would benefit from having a catalase enzyme treatment, such as BioRas® Balance, to neutralize residual chemical before releasing it into their effluent stream. Rapid decomposition of H2O2 in the culture tanks administered with 5 and 10 mg/L of BioRas® Balance led to dissolved oxygen levels above 300% saturation. Fish from Trial 1 were exposed to this supersaturated tank water for approximately 2 h until system water flow was resumed. Although no observable signs of stress were noted, future studies will need to investigate the physiological response and long-term effects of this temporary supersaturation exposure. Lower concentrations of BioRas® Balance may be needed to manage the rapid increase in oxygen that results from the breakdown of H2O2. Future research is also required to understand the change in reaction kinetics when variables such as temperature and salinity are altered and confirm that BioRas® Balance-treated water will not disrupt biofiltration upon resumption of recirculation flow, in the case of a culture tank static bath treatment. Finally, although catalase is (i) a well-studied enzyme with no toxicological effects observed, including effects on aquatic species, (ii) already in use for a wide variety of technical and food applications and (iii) readily biodegradable (REACH, 2022), consideration and care should be made with regard to the other formulation ingredients in catalase products to ensure compatibility with fish health and food safety. Special thanks are extended to Megan Murray, Anna Knight, Kayla Fairfield, and Curtis Crouse for their assistance with this project. This research was supported by the USDA Agricultural Research Service under Agreement No. 59-8082-0-001. All experimental protocols and methods were in compliance with the US National Research Council's Guide for the Care and Use of Laboratory Animals, and were approved by The Conservation Fund Freshwater Institute’s Institutional Animal Care and Use Committee prior to study commencement. Any use of trade, firm or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. The Conservation Fund and USDA are equal opportunity employers and providers. The Conservation Fund Freshwater Institute authors do not have any conflicts of interest to declare. Research representative Novozymes A/S coauthors were responsible for enzyme sampling and dosing recommendations for BioRas® Balance; however, overall experiment execution was conducted independently by The Conservation Fund Freshwater Institute, and therefore Novozymes A/S co-authors do not have conflicts of interest to declare. Study design and protocols were developed by Travis May, Christopher Good, and Kevin Mann. Study execution was carried out by Travis May, Natalie Redman, and Christopher Good. Manuscript writing and data analyses were performed by Travis May, Christopher Good, and Brian Vinci. Manuscript review and editing was performed by all listed authors. The data that support the findings of this study are available from the corresponding author upon reasonable request.