2,3,7,8-TCDD Effects on Visual Structure and Function in Swim-Up Rainbow Trout PAULO S. M. CARVALHO* AND DONALD E. TILLITT Columbia Environmental Research Center, United States Geological Survey, 4200 New Haven Road, Columbia, Missouri 65201 and University of MissourisColumbia, Fisheries and Wildlife Department, 302 AnheusersBush, Columbia, Missouri 65211 An understanding of mechanisms of contaminant effects across levels of biological organization is essential in ecotoxicology if we are to generate predictive models for population-level effects. We applied a suite of biochemical, histological, and behavioral end points related to visual structure and function and foraging behavior to evaluate effects of 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) on swim- up rainbow trout. We detected a dose-dependent decrease in densities of retinal ganglion cells (RGC), key retinal neurons that link the eye with the brain. These changes resulted in corresponding deficits in visual/motor function including reductions in visual acuity and in scotopic and photopic thresholds due to TCDD. The loss of RGCs suggests an increase in convergence of synapses from photoreceptors to RGCs as a cellular mechanism for the visual deficits. Dose-dependent increases in immu- nohistochemical detection of CYP1A protein in the vasculature of the brain and eye choroid was proportional with decreased ganglion cell densities in the retina. TCDD- induced AHR-regulated effects on these tissues might be involved in the detected decrease in ganglion cell densities. Prey capture rate decreased after TCDD exposure only at the highest treatment groups evaluated. Collectively, these results show that TCDD causes biochemical and structural changes in the eye and brain of rainbow trout that are associated with behavioral deficits leading to decreased individual fitness. Introduction Most fish depend on vision for the perception of food at the critical transitional stage from endogenous to exogenous feeding during early life stages (1). Proper development of the visual system structure and establishment of functional vision (2) is essential for successful survival, growth, and recruitment, which are critical processes in the maintenance of fish populations. Dioxin-like chemicals are chlorinated contaminants of environmental concern due to their lipophilicity and ten- dency to bioaccumulate in food chains (3). Extensive research has indicated that most, if not all, of their effects are exerted directly or indirectly through the aryl-hydrocarbon (Ah) receptor (4). The effects of the most potent dioxin, 2,3,7,8- tetrachlorodibenzo-p-dioxin (TCDD), has been extensively evaluated in the early life stages of fish, especially in salmonines. A suite of dose-dependent effects of TCDD have been found in different salmonine species during the period from hatch to swim-up including vascular dysfunction, edema around the yolk sac and heart, craniofacial malfor- mations, decreases in growth rate, and increased mortality (5-7). Factors involved in the observed cardiovascular dysfunction have been identified, although the precise physiological and molecular mechanisms are not yet un- derstood (8). Severe cardiovascular dysfunction and associ- ated edema usually happen before swim-up and are probable causes of mortality at higher doses of TCDD. It has been recently demonstrated that dioxin-like contamination in lake trout larvae reached acutely lethal levels and was an important historical factor in the decline of lake trout populations in Lake Ontario (9). However, dioxin-like contaminant con- centrations measured in field-collected salmonines are presently often below acutely lethal concentrations (9). It is therefore important to explore sublethal effects of dioxins in salmonines at environmentally realistic doses, the approach we used in this study. The importance of the evaluation of ecologically relevant end points in aquatic ecotoxicology in conjunction with the potential mechanisms associated with them has been stressed since the 1980s (10, 11). This approach would help in the development of predictive models of relevant population level effects for important fisheries resources. There are some examples in the literature connecting underlying biochemical mechanisms to behavioral changes (12). However, we still have large gaps in knowledge about the integration of mechanisms of contaminant effect from the molecular to the individual level of biological organization. The study of effects of chemical contaminants on the visual system structure and function provides an interesting paradigm to evaluate links between contaminant effects from the sub- organismal to the individual level of biological organization. Models involving individual performance parameters related to vision and foraging aspects can be used to predict early life stage survival through the critical period (13). In this study, we evaluated visual structure and function to deter- mine the effects of TCDD on biochemical, histological, and behavioral parameters in viable swim-up rainbow trout. First, we tested the hypothesis that a dose-response relationship exists between TCDD exposure and the end points of effect related to visual function (acuity, motion detection, and low- light sensitivity). Second, we evaluated dose-related effects on prey capture rate. Third, we histologically examined key tissues associated with vision. Brain, retina, and the eye choroid vasculature were evaluated by immunohistochem- istry for dose-related CYP1A induction, and the linear densities of retinal photoreceptors (rods and cones) and retinal ganglion cells were quantified. These studies were conducted to provide quantitative linkages between bio- chemical, tissue, organ. and individual level responses of rainbow trout to TCDD. Methods Fish and Exposures. Unfertilized eggs from three 3-year-old females of Eagle Lake strain rainbow trout (Oncorhynchus mykiss) were shipped in an insulated box (5 °C) from Ennis National Fish Hatchery, Montana, and warmed to 10 ( 0.5 °C at a rate of 1 °C/h upon arrival. Eggs were fertilized with milt from two males, water-hardened, and placed in per- forated plastic baskets in groups of 30. Well water (10 ( 0.5 °C, pH 8.2-8.4, hardness 270 mg of CaCO3 L -1 ) was * Corresponding author present address: UFSC - CCB - Depto. Bioquı ´mica C.P. 476 - Cidade Universita ´ria Floriano ´ polis - SC - 88040- 900 Brasil. E-mail: pcarvalho@ccb.ufsc.br; phone: (48)-3316561. Environ. Sci. Technol. 2004, 38, 6300-6306 6300 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 23, 2004 10.1021/es034857i CCC: $27.50  2004 American Chemical Society Published on Web 06/11/2004