Tolerance of Atlantic salmon to novel feed mycotoxins- Implications on fish performance and nutrient interactions

The use of new feed ingredients will introduce new challenges to the farmed salmon such as contaminants and anti-nutrients previously not associated with the farming of marine carnivorous fish such as Atlantic salmon. Screening of currently used Norwegian salmon plant-based feeds as well as novel insect feed ingredients disclosed the presence of toxins produced by fungi (so called mycotoxins). These newly identified mycotoxins include beauvericin (BEA) and enniatins (ENN). In contrary to more studied mycotoxins, these new emerging compounds are fat soluble. Presently no upper limits are set for these mycotoxins in animal feeds. The AquaMyc project aimed to investigate the impact of new feed contaminants (emerging mycotoxins) on Atlantic salmon health and growth. In vitro and in vivo studies were applied in a system biology framework to identify novel impacts of these mycotoxins on fish health and mechanism of toxicity. An acute exposure trial and a subchronic (3 month) Atlantic salmon feeding trial were carried out to investigate how much of the mycotoxins EnnB and Bea salmon can tolerate before negative effects occur and identify toxic effects. These experiments showed that although the two mycotoxins show roughly the same toxicity in cell studies, they had different effects on growth following a feeding trial. The group that had been fed Bea became thinner (reduced fitness factor) despite eating more than the control group. This indicates that high levels of Bea in the feed affect feed utilization and nutrient absorption. Poor feed utilization can be due to both increased energy turnover (need for more energy) and poor nutrient uptake. This was linked to the observed increased intestinal leakage, also at low concentrations, which can result in reduced nutrient absorption. Strongly reduced levels of vitamin E (an important antioxidant) were also observed in the same animals. Although no significant effects on the intestines were observed as a result of EnnB in the feed, EnnB had a marked effect on length growth, already at the lowest doses. This is seen in connection with an increased incidence of skeletal deformities in the group that had the highest level of EnnB in the feed. In addition, the initial in vitro studies on bone cells showed reduced mineralization as a result of EnnB. It is known that reduced mineralization of the skeleton can give rise to skeletal damage. Liver damage was also observed in the ENNB groups. In addition to growth, we observed a reduction in blood percentage already after 1.5 months in salmon fed with EnnB and Bea. This is a clear sign that something is wrong and can be serious if the low blood levels are prolonged. Surprisingly, this effect disappeared at the end of the trial (3 months) in the Bea group but not in the EnnB group. It may therefore appear that the fish are more sensitive to EnnB than Bea. Other in vivo and in vitro studies carried out in the project show that the reduction in red blood cells is probably associated with disruption of the production of heme The cell studies indicated the main mechanism behind the observed toxicity of EnnB and Bea to be mitochondrial disruption and disruption of cellular respiration. This was linked to a novel form of apoptosis, ferroptosis, driven by increased iron uptake followed by lipid related oxidative stress. In vitro experiments also showed that the substances did not interfere with or prevent the cellular immune response against viruses and bacteria, but could themselves initiate inflammatory signals. This can make the salmon more vulnerable to inflammation-related disorders, but we saw no clear indications of inflammation in the gut after 3 months of feeding. We have also developed a PCR-based assay for gene damage. Although the assay showed that high doses of Ennb and Bea could cause gene damage, we believe that this is related to dying cells and not genotoxicity at low exposure levels One of the main aims of the project was to calculate safe levels of the two mycotoxins in feed for salmon. Our data indicate that the levels of EnnB found in some samples of salmon feed and feed ingredients may pose a risk to fish health. Based on current data, Bea is less likely to pose a risk. The project also shows that these substances do not accumulate in the edible part (muscle) or other organs of the salmon. Enniatin and beauvericin in fish feed will therefore most likely not pose any danger to consumers of salmon. The aforementioned experiments examined the effect of single toxins, but a summary of levels of ENNB and BEA in salmon feed made as part of the project shows that they rarely occur alone. If there are high levels of EnnB there may also be high levels of other mycotoxins and other forms of enniatins. Due to structural similarities between ENNB, BEA and other enniatins, it is possible that these work in the same way and will lead to an accumulated effect on blood, liver and growth.

Outcome: • In the AquaMyc project we have established the levels of enniatin B (ENNB) and beauvericin (BEA) which are safe for salmon. • Using the knowledge gained from the project we were able to do a risk assessment based on the current level of ENNB and BEA in salmon feed and give advice on the risk for salmon health. • In this project we established that the two mycotoxins did not accumulate in muscle and only trace amounts were found in salmon. From this we concluded that ENNB and BEA levels in feed does not constitute a risk for the end consumer. • With the help of the Aquamyc project, we have established evidence for novel mechanisms of which ENNB and BEA exert damage towards the cell (ferroptosis and disruption of iron/heme homeostasis). • The following toxic endpoints have been established for enniatin and beauvericin exposure: o Reduction in red blood cells, (BEA, ENNB) o Liver damage, (ENNB) o Intestinal leakage (BEA) o Reduction in feed conversion (BEA) o Reduced mineralization (ENNB/BEA) o Stimulation of inflammatory cascade (TNFa) o Growth (BEA/ENNB) • A sigmoid dose response at lower doses were identified. • An assay/setup for measuring intestinal leakage in salmon by modifying a method used on mice was established. • A qPCR-based method has been developed for measuring DNA damage in salmon. • The project has generated 3 master students and one PhD student (waiting for the opponents report) • In the project we have 5 scientific publications of which two are published and three are under review. Four additional manuscripts are under preparation. Impact • We anticipate that the established levels will be used by the food authorities for setting guidance values for ENNB and BEA in salmon feed. Guidance values have been set for most other more common mycotoxins. • Feed producers has until now had a focus on the legislative regulated mycotoxins (DON, OTA, Aphlatoxin, ZEA EtC) this has also been the focus for the global feed surveillance (BIOMIN). Our results have made feed producers aware of the importance of controlling the levels of ENNB in particular in the feed and feed commodities. • The novel mechanisms of toxicity have helped to move the understanding of ENNB and BEA toxicity forward, not only for salmon, but also for mammals. Previous works have mostly focused on the mitochondrial perturbation. We show that low level exposures can exert different adverse effects through different mechanisms in comparison to cytotoxic dose levels. • The setup for measuring intestinal leakage ex vivo in salmon will be used in future studies. • Our novel findings regarding mode of action involving disturbance of heme biosynthesis with reduced red blood cell levels as adverse outcome has moved the research front forward. • The developed Salmon DNA damage assay will be used in other studies.

The use of new feed ingredients will also introduce new challenges to the farmed salmon such as contaminants and anti-nutrients, not early associated with the farming of marine carnivorous fish such as Atlantic salmon. Screening of currently used Norwegian salmon plant-based feeds as well as novel insect feed ingredients showed the presence new mycotoxins such as beauvericin (BEA) and enniatins (ENN). In contrary to more studied mycotoxins, these new emerging compounds are fat soluble and are recently been reported to be present in farmed marine fish. At present, there is knowledge gap on the implication of these emerging mycotoxins on salmon health and welfare, or their potential transfer from feed to fillet. No upper limits are set for these mycotoxins in animal feeds. From in vitro and mammalian and poultry studies it is know that emerging mycotoxins has a potential for generating oxidative stress, osteotoxicity, immunotoxicity and can interfere with cholesterol and vitamin A metabolism. The current proposal aims to investigating the impact of new feed contaminants (emerging mycotoxins) on vitamin requirement, as well as lipid metabolism, growth, health and immune-responses in Atlantic salmon. In addition to these targeted endpoints, system biology approaches applying the salmon genome will be used to gain fundamental knowledge on the interaction between mycotoxins and nutrients. We will use both state of the art salmon in vitro assays and conduct a feeding experiment in order to investigate the tolerance of Atlantic salmon for newly identified fat soluble mycotoxins in novel sustainable feeds.