Assessing welfare in fish: the answer is in the water!

Fish as second class vertebrates: Both the public and the research community have somewhat lower moral standards for scientific procedures involving fish than mammals. This is probably due to the difficulty in empathising with the evolutionarily more distant fish. The perceptions that fish do not experience feelings and emotions and do not merit quality care need to be challenged. There is a widely recognised lack of detailed manuals and guidance for care and housing of experimental fish. Currently it is unclear if fish would benefit from environmental enrichment during holding and experimentation and the majority are held in bare aquaria. Conditions are deemed acceptable if fish just survive, grow and reproduce.
Happy fish: We believe that there is a difference between survive and thrive and we therefore wish to pursue a novel idea; the development and validation of measures of 'good' welfare (as opposed to indicators of 'poor' welfare) in fish. We believe that measurements of biochemical brain messengers (brain monoamines, BMAs) have the potential to provide such a means. Serotonin, for example, contributes to feelings of well-being in humans, is widely acknowledged as our mood hormone, and is often referred to as the "happy molecule''. Dopamine also plays important roles in behaviour, cognition, voluntary movement, motivation, punishment and reward, sleep, mood, attention, working memory and learning. Adrenaline participates in the fight-or-flight response and melatonin has been shown to be effective in treating one form of depression and seasonal affective disorder. The strong relationships between BMAs and 'mood' are reflected by the fact that these molecules are the standard target for drugs to treat human depression. Research over the past 30 years has established great similarities between fish and mammalian BMA systems . Dopamine and serotonin levels (and their oxidized amine metabolites) have previously been measured in the brains of fish, with responses to stress being well-documented.
Look in the water: Most objective methods for assessing welfare in fish (i.e. not utilising behavioural observations alone) suffer from a great drawback: the need to sample invasively (i.e. by drawing blood) or even lethally (brain tissue for measuring BMAs levels). This makes them highly unsuitable for routine assessment and monitoring. Our team at Cefas has pioneered the non-invasive measurement of cortisol, a well-documented stress hormone, as well as various sex steroids and melatonin in fish by measuring them in the water rather than the blood. We also have preliminary data suggesting that serotonin is also excreted in the water in measurable concentrations. Hence our extensive experience to date suggests that many important molecules that act within the brain and the body are excreted, secreted or leak out of fish into the water (through the gills, skin, urine and faeces) and can be measured non-invasively in the water.
What is a good aquarium? Our experimental plan includes three widely used laboratory fish species (zebrafish, trout and stickleback) and three different holding conditions (standard-as per manuals and common practices; optimum- standard + appropriate environmental enrichment based on the behaviour and ecology of the species e.g. substratum, water current, shelter etc; sub-standard- standard + mild stressors such as netting, chasing, exposure to noise).
The science bit: Water samples will be collected and extracts will go through a number of analytical tests to identify a marker of welfare and develop a method that allows routine measurements to be made. In all experiments, the water chemistry profile will be linked to both behaviour and the levels of cortisol, a measure of stress.

The stress response in fish is very similar to that in other vertebrates and can be described in three stages. The primary response involves the activation of the hypothalamus-sympathetic-chromaffin cell axis which produces cathecolamines (adrenaline and noradrenaline); the hypothalamic-pituitary-interrenal tissue (HPI) axis which produces mainly cortisol in teleost fish. The secondary response is the physiological and behavioural adjustment to stress conditions. It includes the activation of a number of metabolic pathways which induce changes in blood chemistry and haematology, respiration, acid-base disturbances with ion losses at the gills, cellular response and immune function As a direct consequence of elevated circulating levels of catecholamines and cortisol, a wide range of secondary changes are evoked including the changes in rates of turnover of brain neurotransmitters such as dopamine and serotonin. If exposure to the stressor is prolonged tertiary responses occur, which are changes in the whole organism and have an eventual impact on the population via immuno-suppression and inhibition of reproduction. We are interested in the consequences of primary and secondary responses, particularly changes in brain monoamine (BMAs) levels. We hypothesise that these change with feelings and moods in fish, and can be developed into a welfare marker in fish. Since our ultimate aim is to produce a tool that can be used routinely, we will focus on the tank water as the substrate for the marker. In addition to targeted analysis of BMAs and their metabolites in the water, we will employ two novel approaches for collecting and analysing the water chemistry; the concentration of bio-molecules present in the water by polar organic chemical integrative samplers (POCIS) and the analysis of the resulting water extracts by comparative metabolomics.

This research project will shape and influence a wide variety of national and international beneficiaries including scientific, veterinary, government [international, national, local and devolved government], consumers, stakeholders and the general public. All these groups will be involved throughout the project life cycle and participate in developing appropriate strategies and decision making tools to inform, manage and encourage the adoption of project outcomes. In particular, by involving a comprehensive range of experts in the consortium, the project will serve the needs of policy makers through the creation of new knowledge and expertise for early adoption of the project deliverables by informing the decision making processes and thereby facilitating policy decisions. The team will engage with stakeholders in developing an understanding of the project approach to the relevance of ethical and legislative issues through the use of surveys, focus groups, interviews and direct observations. The public opinion on the research will be sought by polls of selected and representative groups with the specific aim of determining what works, what does not work, and the reasons for each outcome. Partners will collaborate with relevant national and international government agencies, principally to ensure an understanding of the key drivers. It is feasible that such an approach may facilitate over time a change in the legislative structure of fish testing. The importance and recognition of these impacts will be embedded throughout the project from data collection to interpretation/modelling to the various outputs, which will be communicated to stakeholders and the general public. Thus, apart from the publication of scholarly articles, and specialist conference papers, the team will organise annual workshops and meetings [involving members of the advisory group] to engage with, and inform stakeholders and policy makers, to exchange information, and to learn from each other. The participants will work with institutional press offices to publicise key outcomes to the wider public in newspapers, popular magazines, radio and television. The benefits of this research programme will have a direct impact on improving well being and economic and sustainable goals. The outcomes developed during the research will have a lasting impact and legacy on animal testing globally. Critically, this project will have a sound link to all stakeholders and will deliver beyond the value of its constitutive parts.