Chronic dietary mercury exposure causes oxidative stress, brain lesions, and altered behaviour in Atlantic salmon (Salmo salar) parr

Abstract

Atlantic salmon (Salmo salar L.) parr were fed for 4 months on fish meal based diets supplemented with mercuric chloride (0, 10, or 100 mg Hg kg⁻¹ DW) or methylmercury chloride (0, 5, or 10 mg Hg kg⁻¹ DW) to assess the effects of inorganic (Hg) and organic dietary mercury on brain lipid peroxidation and neurotoxicity. Lipid peroxidative products, endogenous anti oxidant enzymes, brain histopathology, and overall behaviour were measured. Methylmercury accumulated significantly in the brain of fish fed 5 or 10 mg kg⁻¹ by the end of the experiment, and inorganic mercury accumulated significantly in the brain only at 100 mg kg⁻¹ exposure levels. No mortality or growth reduction was observed in any of the exposure groups. Fish fed 5 mg kg⁻¹ methylmercury had a significant increase (2-fold) in the antioxidant enzyme super oxide dismutase (SOD) in the brain. At dietary levels of 10 mg kg⁻¹ methylmercury, a significant increase (7-fold) was observed in lipid peroxidative products (thiobarbituric acid reactive substances, TBARS) and a subsequently decrease (1.5-fold) in anti oxidant enzyme activity (SOD and glutathione peroxidase, GSH-Px). Fish fed 10 mg kg⁻¹ methylmercury also had pathological damage (vacoulation and necrosis), significantly reduced neural enzyme activity (5-fold reduced monoamine oxidase, MAO, activity), and reduced overall post-feeding activity behaviour. Pathological injury started in the brain stem and became more widespread in other areas of the brain at higher exposure levels. Fish fed 100 mg Hg kg⁻¹ inorganic mercury had significant reduced neural MAO activity and pathological changes (astrocyte proliferation) in the brain, however, neural SOD and GSH-Px enzyme activity, lipid peroxidative products (TBARS), and post feeding behaviour did not differ from controls. Compared with other organs, the brain is particular susceptible for dietary methylmercury induced lipid peroxidative stress at relative low exposure concentrations. Doses of dietary methylmercury in the range of 5 mg kg⁻¹ induces protective redox defences in the brain as seen from the induction of anti-oxidant enzyme SOD activity. However, above a threshold of 10 mg kg⁻¹ methylmercury these defences are overcome and lipid peroxidative injury (TBARS) as well as severe pathological damage and adverse behaviour become apparent.

Introduction

Both inorganic and organic mercury are well known to cause damage to the central nervous system (CNS) of teleost fish. The organic, methylated, mercury compound readily passes the blood–brain barrier and is considered by far the most neurotoxic form (review, Baatrup, 1991). Methylmercury easily accumulates along the aquatic food chain and dietary exposure is a major route of uptake in fish (Harris and Bodaly, 1998). Inorganic mercury may be methylated in the gut lumen prior to absorption (Rudd et al., 1980), or absorbed directly to produce toxic effects (Handy and Penrice, 1993). Whereas neurotoxic effects of waterborne mercury of toxic effects are well researched (review, Baatrup, 1991), studies on neurotoxic effects after oral exposures remain limited in fish (Wobeser, 1975b, Skak and Baatrup, 1993). The integrated responses at different levels of biological organisation of sublethal neurotoxicity in chronic (>30 days, Veith et al., 1979) dietary mercury exposed fish have not been previously investigated.

Earlier described integrated responses of mercury induced pathogenesis of the nervous system in mammals include increased membrane permeability and neuronal protein production, causing disruption of metabolic functions and structures, leading to a loss of enzyme functions and eventually cell death (review, Chang, 1977). Recently, lipid peroxidative stress has been suggested as an additional mechanism by which mercury exerts initial neurotoxic effects in mammals (Yee and Choi, 1996). Mercury readily deposits in mitochondria, and selective disruption of the mitochondrial electron transport chain has been suggested as a specific mechanisms by which methylmercury induce the formation of free radicals and lipid peroxidative stress (Yee and Choi, 1996, Zaman and Pardini, 1996, Konigsberg et al., 2001). Oxidation of membrane lipids can lead to the loss of cellular or organelle membrane integrity which can eventually result in cell death and pathological injury (Girotti, 1998). Salmonid tissues are characterised by high concentrations of polyunsaturated fatty acids (PUFA) compared with most mammalian tissue, and fish may, therefore, be particularly susceptible to mercury enhanced lipid peroxidative cellular damage (Winston and DiGiulio, 1991). Endogenous anti-oxidant enzymes such as super oxide dismutase (SOD) and glutathione peroxidase (GSH-Px) are involved in the protection against oxidative stress and lipid peroxidation (Pokorný, 1987). Induction of these anti oxidant enzymes indicates an adaptive onset of the redox defence system, whereas inhibition is thought to contribute to oxidative stress in mouse brain following mercury intoxication (Yee and Choi, 1994, Hussain et al., 1997). In fish, reports on mercury induced biochemical disorders of the nervous system are limited to inhibition of key enzymes for neurological function such as Na⁺/K⁺-ATPase (Verma et al., 1983), monoamine oxidase (MAO) (Ram and Sathyanesan, 1985), and acethylcholine esterase (Gill et al., 1990). Although lipid peroxidative products were found in the brain of cat-fish (Heteropneustes fossilis) exposed to aqueous inorganic mercury (Bano and Hasan, 1989), lipid peroxidation as a fundamental process in brain injury has not yet been explored before in dietary mercury exposed juvenile fish.

In fish, most studies on mercury induced regional brain pathology and behaviour changes are restricted to waterborne exposure, mostly reporting structural damage to the olfactory organs and disturbed sensory behaviour (Sutterlin and Sutterlin, 1971, Hara et al., 1976, Rehnberg and Schreck, 1986, Baatrup et al., 1990; review, Baatrup, 1991, Ribeiro et al., 1995). The rapid absorption of aqueous mercury in the central nerve system via the olfactory pathway can not easily be related to food borne exposure in fish, hence behaviour changes other than sensory disturbance might be expected during dietary mercury exposures. Embryonic exposures to aqueous mercury causes latent effects on the feeding behaviour and predator avoidance of hatchlings (Weis and Weis, 1994, Fjeld et al., 1998), which are related to reduced locomotor capacity (Zhou and Weis, 1998). Behaviour studies on oral dietary mercury exposures are lacking in fish.

The aim of the study was to make a total effect assessment of the neurotoxic effects of inorganic and organic mercury by the oral route of uptake using integrated biochemical, pathological and behavioural parameters. Furthermore, the study aims to assess lipid oxidative stress as a possible mechanism by which organic and inorganic dietary mercury exert their damage to fish brain at sublethal exposures. The dosing in the present study was set to reveal the toxic mechanisms involved for organic and inorganic mercury and not to compare dose response of the two mercury forms.