Dietary copper exposure and recovery in Nile tilapia, Oreochromis niloticus

Dietary copper exposure and recovery in Nile tilapia, Oreochromis niloticus

Aquatic Toxicology 76 (2006) 111–121 Dietary copper exposure and recovery in Nile tilapia, Oreochromis niloticus Benjamin J. Shaw, Richard D. Handy ∗...

295KB Sizes 60 Downloads 115 Views

Aquatic Toxicology 76 (2006) 111–121

Dietary copper exposure and recovery in Nile tilapia, Oreochromis niloticus Benjamin J. Shaw, Richard D. Handy ∗ School of Biological Sciences, University of Plymouth, Drake Circus, Plymouth PL4 8AA, UK Received 1 July 2005; received in revised form 4 October 2005; accepted 4 October 2005

Abstract There are few reports of dietary copper (Cu) toxicity to warm water species of freshwater fish, and little is known about recovery from dietary Cu exposure. In this study Nile tilapia (Oreochromis niloticus) were fed to satiation on a Cu-loaded diet (2000 mg Cu kg−1 dry weight (dw) feed), or a control diet (3 mg Cu kg−1 dw feed), for 42 days. All fish were then fed the control diet for a further 21 days to assess recovery. Nutritional performance, haematology, histology, and tissue ion content (Cu, Na+ , and K+ ) were measured. No mortalities occurred during the experiment. Dietary copper exposure was confirmed by elevated Cu concentrations in the intestine (30-fold), liver (three-fold) and gills (2.7-fold) of Cu-exposed fish compared to controls after 42 days (ANOVA, P < 0.05). Copper-exposed fish showed a reduction in food intake, and weight gain by day 21 of exposure, compared to controls (ANOVA, P < 0.05) and this persisted throughout the experiment. There were no treatment-dependent effects on food conversion ratio or hepatosomatic index, and all fish showed normal tissue Na+ and K+ , and haematology throughout the experiment. Gill and intestine did not show overt pathology, but fatty change was observed in the liver of Cu-fed fish during exposure. The recovery phase on normal food was characterised by a reduction in intestinal and branchial Cu levels back to control values. However, the liver of the Cu-fed fish showed a further 1.7-fold rise in Cu content and marked hepatic lipidosis (increased intracellular fat stores) post-exposure, suggesting redistribution of Cu to the liver and delayed hepatotoxicity. © 2005 Elsevier B.V. All rights reserved. Keywords: Dietary copper exposure; Growth; Liver morphology; Recovery; Nile tilapia; Oreochromis niloticus

1. Introduction Copper (Cu) is an essential micronutrient for vertebrate animals and has numerous functions in cellular biochemistry including vital roles in cellular ∗ Corresponding author. Tel.: +44 1752 232900; fax: +44 1752 232970. E-mail address: [email protected] (R.D. Handy).

0166-445X/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.aquatox.2005.10.002

respiration, and as a co-factor for over 30 different enzymes (Linder, 1991). Copper is also toxic in excess (Handy, 1996), and therefore Cu uptake across the gut must be carefully regulated (mammals, Pe˜na et al., 1999; Harrison and Dameron, 1999; Arredondo et al., 2000; fish, Handy et al., 2000; Clearwater et al., 2000; Bury et al., 2003; Burke and Handy, 2005). Teleost fishes have a nutritional requirement of about 3–10 mg Cu kg−1 dry weight (dw) feed, depending on

112

B.J. Shaw, R.D. Handy / Aquatic Toxicology 76 (2006) 111–121

the species, feeding regime and life stage (Clearwater et al., 2002). Although the precise Cu requirements of Nile tilapia, Oreochromis niloticus, are uncertain, values for tilapia species in aquaculture are suggested (e.g. 4 mg Cu kg−1 feed for O. niloticus × O. aureus hybrids, Shiau and Ning, 2003). Thresholds for excess dietary Cu toxicity in freshwater fish are between 16 and 730 mg Cu kg−1 dw feed (Clearwater et al., 2002), but most of the available toxicity data are for salmonid fish (Handy, 1996; Clearwater et al., 2002). In salmonids at least, the toxic effects of excess dietary Cu can include reduced growth (review, Clearwater et al., 2002), severe lesions in the gut at very high Cu inclusions (10 g Cu kg−1 food, Handy, 1996), or more subtle changes in intestinal cell proliferation and turnover (Berntssen et al., 1999; Lundebye et al., 1999). Fatty change in the liver (Handy et al., 1999), altered haematology (Knox et al., 1982), and increased costs of locomotion (Campbell et al., 2002) are also implicated in long-term metabolic adjustments to dietary Cu exposure in trout (review, Handy, 2003). However, there are few reports on dietary Cu toxicity in tropical or sub-tropical freshwater fish, except perhaps on African catfish (Handy et al., 2000). Aqueous Cu exposure studies on O. niloticus confirm that Cu can be acutely toxic at around 1–5 ␮g l−1 total Cu (Co˘gun and Kargin, 2004), and aqueous sublethal effects include reduced growth and nutritional impairment (Ali et al., 2003), but to our knowledge chronic dietary Cu toxicity in Nile tilapia has not been investigated. Nile tilapia are one of the most important freshwater finfish in world aquaculture (Likongwe et al., 1996; Barriga-Sosa et al., 2004), and a very important species in global capture fisheries (Balirwa, 1992). Among the numerous regions now inhabited by Nile tilapia, many are under threat from metal pollutants including copper (e.g. Khallaf et al., 1998, 2003; Shakweer, 1998; Adham et al., 1999). Wild Nile tilapia are also known to ingest contaminated lake water during feeding with consequent deleterious effects on gut function (Getachew, 1988). In this study, we aimed to make the first toxicological assessment of excess dietary copper, and subsequent recovery, in Nile tilapia, O. niloticus. We adopt a holistic approach similar to Handy et al. (1999) and explore nutritional performance, tissue electrolyte contents and Cu accumulation, haematology, and histopathology of the intestine, liver and gills.

Recovery from dietary Cu exposure has rarely been investigated in fish (e.g. trout, Handy, 1992), and the recovery phase of this experiment therefore adds to the sparse literature on the reversibility of dietary Cu toxicity to fish.

2. Material and methods 2.1. Experimental design Nile tilapia weighing 85.2 ± 2.6 g (mean ± S.E., n = 80 fish), were obtained from laboratory stocks bred at the University of Plymouth, UK, and placed into a recirculation system consisting of four 150 l experimental aquaria (n = 20 fish per tank) with flowing, filtered (pre-filter plus 1000 l biofilter), aerated, and dechlorinated Plymouth tap water at 25 ± 1 ◦ C (see below for water quality, also used in laboratory stock tanks). All fish were fed a control diet with no added Cu, to satiation for 14 days in order to acclimate to experimental conditions. Thereafter fish in two tanks were fed a Cu-supplemented diet for 42 days, whilst fish in the other two tanks remained on the control diet (see below for measured Cu contents of the diets). This was followed by a 21-day recovery period with all tanks fed the control (no added Cu) diet. Photoperiod for the experiment was 12 h light:12 h dark. Throughout the experiment fish were fed to satiation once a day, at approximately 11:00 h. Care was taken to ensure that no uneaten food remained in the tanks during feeding, and Cu did not leach from the feed. Copper concentrations in water samples collected 10 min before and after feeding remained low throughout the experiment (<0.96 ␮mol l−1 , see below). Growth and nutritional performance were monitored throughout the experiment, and fish were randomly sampled from each tank every 21 days for haematology, tissue ion analysis, and histology (see below). Fish were not fed the day before sampling times in order to empty the gut and to facilitate dissection. 2.2. Diet formulation The control diet was based on a commercial feed (Tetra Pond Floating Food Sticks, Tetra, Germany) with a proximate composition of (% of dry diet; from manufacturer’s guidelines): protein 28.0; moisture 7.0; lipid

B.J. Shaw, R.D. Handy / Aquatic Toxicology 76 (2006) 111–121

3.5; ash 7.0; fibre 2.0 (trace elements not reported in the manufacturer’s formulation, see below for measured Cu). The Cu-supplemented diet was formulated by gelatine coating of the commercial feed with copper sulphate. In order to achieve a nominal Cu concentration of 2000 mg Cu kg−1 dw feed, 2.358 g of CuSO4 ·5H2 O (AnalaR grade, BDH, Poole, UK) was dissolved in 35 ml of deionised water with 1.2 g of bovine gelatine (Sigma Diagnostics, Poole, UK) to bond the Cu to the food sticks. The gelatine solution was gradually sprayed onto 300 g of the commercial diet in a Hobart food mixer to ensure even mixing of the food. The gelatine coat dried within minutes, and the Cu diet was stored into air tight containers and frozen at −20 ◦ C to prevent lipid peroxidation. The control diet was similarly treated, except that no Cu was added to the gelatine. The Cu content of diets were confirmed by inductively coupled plasma emission spectrophotometry (ICP-AES; Varian Liberty 200) and was 3.22 ± 0.33 and 1968 ± 5.6 mg Cu kg−1 dw feed, respectively (mean ± S.E., n = 5 and 20 samples) in the control and Cu-supplemented diets.

113

2.4. Haematology Fish were collected (3–4 fish per tank, randomly sampled, or 6–8 fish/treatment at each time point) at 21-day intervals throughout the experiment for haematology (Handy et al., 1999). Briefly, fish were anaesthetised with unbuffered MS222 and whole blood was collected via the caudal vein into heparinised syringes, then fish were weighed and length (total length) recorded. Haematocrit value (HCT) and haemoglobin (Hb) concentration (Sigma Diagnostics kit no. 525-A) were determined immediately. In addition, at the end of the exposure period (day 42) blood was collected and serum separated (1250 rpm for 90 s, Sanyo MSE Micro Centaur) for blood glucose determination (Sigma Diagnostics, glucose (HK) 20 kit, procedure no. 16-UV). The glucose concentration in serum samples was measured in triplicate at 340 nm (Dynex, MRX microplate reader). No blood samples were taken during the recovery phase in view of results obtained during the Cu exposure. 2.5. Histology and tissue ion analysis

2.3. Growth and nutritional performance Growth and nutritional performance were measured according to Handy et al. (1999) with minor modifications. Briefly, food intake was calculated daily for each tank by weighing food containers before and after feeding. All fish were individually weighed at the start of the experiment and at each 21 days sampling interval (a few hours prior to feeding). Individual fish weight was used because the periodic sacrifice of fish during the experiment prevented nutritional parameters being calculated from cumulative tank biomass. Specific growth rate (SGR (% day−1 ) = (loge W2 − loge W1 )/(t2 − t1 ) × 100; at time intervals t1 and t2 , where W1 and W2 are the fish weights at t1 and t2 , respectively) and food conversion ratio (FCR = feed intake (g)/weight gain (g)) were calculated from mean gain in body weight for each treatment for (i) the Cu exposure phase (days 0–42), (ii) recovery phase (days 43–63), and (iii) for entire experiment (Cu exposure and recovery combined; days 0–63). Condition factor (K (%) = weight (g)/length3 (cm) × 100) and hepatosomatic index for each fish (HSI (%) = liver weight (g)/body weight (g) × 100) were similarly determined.

After blood sampling fish and terminal anaesthesia in accordance with ethical approval, tissues were carefully dissected for histology and trace metal analysis (Handy et al., 1999 with minor modifications). Briefly, the second gill arches from both opercular cavities, posterior intestine, and then liver were harvested. Half of each tissue collected was fixed (10% buffered formaldehyde-containing saline) for histology, and the other half used for tissue ion analysis (Handy et al., 1999). Gills for histology were decalcified with Rapid Decalcifier (CellPath Plc, UK). These, along with liver and intestine samples, were processed for routine wax histology (8 ␮m thick sections, stained with Mallory’s Trichrome) according to Handy et al. (1999), and photographed (Olympus Vanox microscope and Nikon E990 Digital Camera). Tissues for trace metal analysis were oven dried to a constant weight, digested in 5 ml of concentrated nitric acid, then diluted to 20 ml with deionised water and analysed by inductively coupled plasma atomic emission spectrophotometry for Cu, K+ and Na+ (ICP-AES, Varian Liberty 200) according to Handy et al. (2000). Percentage tissue moisture content was calculated from wet and dry tissue weights.

114

B.J. Shaw, R.D. Handy / Aquatic Toxicology 76 (2006) 111–121

2.6. Water quality Water samples were collected twice weekly throughout the experiment, 10 min prior to and 10 min post-feeding and analysed according to Handy et al. (1999). No statistical differences in water quality between treatments, or between pre- and post-feeding water samples occurred (P > 0.05, t-tests), so data were pooled to give a mean for all measurements in all tanks during the experiment (mean ± S.E., n = 144, in mmol l−1 ): Ca2+ , 0.29 ± 0.001; Cl− , 1.58 ± 0.20; Cu, <0.0009; K+ , 0.41 ± 0.01; Mg2+ , 1.36 ± 0.12; Na+ , 1.29 ± 0.24; total ammonia, 0.13 ± 0.008; nitrate, 0.34 ± 0.007; nitrite, 0.0004 ± 0.0001; dissolved oxygen at the tank outflow, 84 ± 3%; pH, 7.9 ± 0.09; temperature 24.5 ± 0.21 ◦ C. Ammonia, nitrate and nitrite were measured using a multiparameter ion specific meter (Hanna C103). Hand held meters were used for dissolved oxygen (Hanna HI 9142) and pH (Hanna HI 9025). 2.7. Statistical analysis All data were analysed using StatGraphics Plus version 5.1, in a similar way to Handy et al. (1999). No tank effects were observed within treatments, so data were pooled by treatment for statistical analysis. After initial descriptive statistics, and a variance check (Bartlett’s test), data were analysed by analysis of variance (ANOVA) for dose and time-effects. Where only one time point was involved, control versus treatment effects were also analysed using the two-tailed student’s t-test. All data are presented as mean ± standard error (S.E.) and a rejection level of P = 0.05 was used for all statistical analysis.

3. Results 3.1. Copper accumulation No mortalities were observed during the experiment. Dietary copper exposure was confirmed by elevated Cu concentrations in the intestine (30-fold), liver (three-fold) and gills (2.7-fold) of Cu-exposed fish compared to controls after 42 days (ANOVA, P < 0.05, Table 1). Aqueous Cu concentrations remained low (<0.96 ␮mol l−1 ) throughout the experiment, and there

was no temporal increase in the Cu content of control fish at any time compared to the initial fish at the start of the experiment (ANOVAs for time-effects within controls, P > 0.05). During the recovery phase when all fish were fed the control diet, Cu concentrations in the intestines and gills of the Cu-exposed fish decreased so that they were not statistically different from their respective controls at the end of the experiment (ANOVA, P > 0.05). However, hepatic Cu concentrations continued to rise in the Cu-exposed group during the recovery phase (Table 1). This post-exposure rise in hepatic Cu levels represented a further 1.7-fold increase in Cu content over that in the Cu-exposed fish at day 42 (ANOVA, P < 0.05, Table 1), and a four-fold increase in liver Cu content in the exposed fish compared to initial or control fish over the entire experiment. 3.2. Tissue sodium, potassium and moisture content There were no statistically significant treatmentdependent differences in Na+ and K+ levels, or tissue moisture content in any organ between treatments. There were no time-effects within treatment (ANOVAs, P > 0.05), except some variability in the mean Na+ values for the gill data compared to that in the initial fish but this did not follow any particular time sequence. 3.3. Growth and nutritional performance Fish from both treatments gained weight during the experiment, but body weight in the Cu-exposed fish was significantly lower than in the controls at the first sampling point (day 21), and remained less than the controls throughout the experiment (Fig. 1). Cumulative food intake was also lower in the Cu-fed fish (Fig. 1). This was reflected in SGR (% of weight gain day−1 ) of Cuexposed fish. They had an SGR of 1.2 compared to 1.58 in controls during the exposure phase (Table 3). In the post-exposure phase the SGR of both treatments were similar, but this did not compensate for the initially low SGR in the Cu-exposed fish, and over the entire experiment the mean SGR of Cu-exposed fish was less than the control values (Table 3). There were no statistically significant differences in condition factor or HSI, although these tended to be higher in the Cu-fed fish by the end of the experiment (Table 3). Daily ration on a %

B.J. Shaw, R.D. Handy / Aquatic Toxicology 76 (2006) 111–121

115

Table 1 Total Cu concentrations in gill, liver and intestine of Nile tilapia fed a normal diet (control, 3 mg Cu kg−1 food) or a diet containing elevated copper (2000 mg Cu kg−1 food) for 42 days, followed by a recovery period on normal food for a further 21 days Experimental phase

Sampling day

Treatment

Time-zero Cu exposure Cu exposure Recovery Recovery

0 42 42 63 63

Initial fisha Control Elevated Cu Control Elevated Cu

Tissue Gills

Liver

0.07 ± 0.09 (8) 0.07 ± 0.02 (6) 0.17 ± 0.02 (6)* 0.05 ± 0.00 (6) 0.05 ± 0.00 (6)+

6.63 3.87 11.55 5.50 20.20

Intestine ± ± ± ± ±

1.09 (8) 0.98 (6) 1.69 (6)* 1.08 (6) 1.85 (6)*,+

0.17 ± 0.02 (8) 0.31 ± 0.07 (6) 9.34 ± 1.06 (6)* 0.24 ± 0.03 (6) 0.33 ± 0.10 (6)+

Data are means ± S.E. (n is the number of fish), expressed as ␮mol Cu g−1 dry weight of tissue. a Initial fish were sampled immediately prior to starting the dietary Cu exposure. No time-effects were observed in the control fish compared to the initial fish (ANOVA, P > 0.05). * Significantly different compared to the control within each time point (ANOVA, P < 0.05). + Significantly different at the end of the recovery period compared to Cu-exposed fish at 42 days (ANOVA, P < 0.05).

of body weight basis (% of bw day−1 ) in Cu-exposed fish was lower than that of the controls (significantly different, Table 3), and this is attributed to less cumulative food intake in the Cu-exposed fish (Fig. 1). 3.4. Haematology

Fig. 1. Body weight (A) and cumulative food intake (B) in Nile tilapia fed a normal diet (control, 3 mg Cu kg−1 food, open symbols) or a diet containing elevated copper (2000 mg Cu kg−1 food, black symbols) for 42 days, followed by a recovery period on normal food for a further 21 days. In panel (A) data are means ± S.E., n = 29–36 fish per treatment at each time point. The dashed line indicates the end of the exposure phase, and the return of all fish to normal food. Note the grey symbol in panel A indicates initial fish at time-zero (n = 80 fish). In panel (B) data are means of duplicate tanks for each treatment. (*) Significant difference from the control within each time point (ANOVA, P < 0.05). Different letters indicates a significant difference in time within treatment (ANOVA, P < 0.05).

There were no treatment-dependent changes in either whole blood haemoglobin levels or haematocrit values (t-tests, P > 0.05) after 42 days exposure to excess dietary Cu. Haemoglobin levels were (mean ± S.E., n = 6–8 fish, in g dl−1 ) 5.55 ± 0.40, 6.15 ± 0.44, and 5.83 ± 0.35 for initial fish, and control and Cu-exposed fish at day 42, respectively. Haematocrit (HCT) in initial fish was 24 ± 1.3% (mean ± S.E., n = 8 fish), and although there was a statistically significant increase in HCT over time (t-tests, P < 0.05) in both treatments compared to initial fish, there were no Cu-dependent effects on HCT at the end of the exposure period (HCT% at day 42; controls, 32.8 ± 0.7%; Cuexposed, 31.4 ± 0.9%; mean ± S.E., n = 6 fish). Blood glucose was also measured at the end of the exposure period (day 42), but no differences were observed (t-test, P > 0.05) between treatments (in mmol l−1 ; controls, 3.0 ± 0.5; Cu-exposed, 3.7 ± 0.6; mean ± S.E., n = 6 fish). 3.5. Histology Gill, intestine and livers were examined at the end of the 42 days exposure period, and also during the

116

B.J. Shaw, R.D. Handy / Aquatic Toxicology 76 (2006) 111–121

Fig. 2. Gill, intestine and liver morphology in control and Cu-exposed Nile tilapia. Panels (A) and (B) are gills from control and Cu-exposed fish, and (C) and (D) are transverse sections of intestine from control and Cu-exposed fish, respectively at the end of the exposure phase (day 42). Gills and intestines show normal morphology. Panels (E) and (F) are liver sections from control and Cu-exposed fish, respectively at the end of the exposure phase, and (G) from a Cu-exposed fish after a further 21 days on normal diet (recovery phase). Note, fatty change characterised by lipid deposition in the hepatocytes, which persisted into the recovery phase for the Cu-fed fish. Scale bars indicate size in ␮m. All sections were 8 ␮m thick and stained with Mallory’s trichrome.

B.J. Shaw, R.D. Handy / Aquatic Toxicology 76 (2006) 111–121

recovery phase on normal diet. The gills of both control and Cu-exposed fish showed normal structure, with no evidence of oedema, epithelial lifting, foci of necrosis, aneurisms or peri-venule bleeds in any of the fish examined (Fig. 2A and B). Intestine also showed normal morphology, with no evidence of necrosis, oedema, excessive epithelial sloughing or haemorrhage in any of the fish examined at day 42 (Fig. 2C and D). The normal structure of the gills and gut was also evident in the recovery phase of the experiment (data not shown). However, there were changes in liver morphology at the end of the exposure period (Fig. 2E and F). Three livers from Cu-exposed fish showed rare and small foci of haemorrhage (blood spots), and twothirds (four out of six examined) of the livers from Cu-exposed fish showed fatty change characterised by increased lipid content of the hepatocytes and a concomitant loss of sinusoidal space (Fig. 2F), compared to controls (Fig. 2E). The fatty change in livers of Cuexposed fish continued, and was exacerbated during the recovery phase of the experiment with cells full of lipid (Fig. 2G), while controls remained normal. Foci of necrosis were not observed in any livers, despite lipidosis, although less definition of nuclei in some cells with high lipid contents from Cu-exposed fish at the end of 42 days is suggestive of a few cells in the early stages of nuclear atrophy.

4. Discussion This study is a first report of chronic dietary copper toxicity in Nile tilapia and overall we show that these fish accumulate excess Cu in the liver and intestine, and show a decline in growth and nutritional performance which is associated with liver pathology. Importantly, Nile tilapia do not recover quickly from dietary Cu exposure. Compensatory growth did not occur and the liver showed further increases in Cu content and fatty change during the recovery phase. The latter effects highlight the risk of post-exposure toxicity to the liver following dietary Cu exposure in fish. 4.1. Copper accumulation Copper levels in the tissues of control fish in this study (a few ␮mol g−1 dw or much less, Table 1) are broadly similar to previous reports for tilapia (Shiau

117

and Ning, 2003; Co˘gun and Kargin, 2004), and to rainbow trout held in similar water quality (2 ␮mol g−1 dw or much less, Handy et al., 1999). The Cu accumulation in Nile tilapia also reflected the route of exposure with large increases in the Cu content of the liver and intestine (Table 1) and is consistent with previous studies on temperate species such as rainbow trout (Handy, 1992; Handy et al., 1999; Kamunde et al., 2001; Campbell et al., 2002; Kamunde and Wood, 2003), Atlantic salmon (Berntssen et al., 1999; Lundebye et al., 1999), and even marine teleosts such as grey mullet (Baker et al., 1998). The gills also showed an increase in Cu content during exposure (Table 1), which cannot be explained by aqueous Cu uptake, because waterborne Cu levels remained low throughout the experiment and gill morphology was normal. Food regurgitation was not observed in this study. The small increase in gill Cu content therefore probably reflects systemic Cu in the gill tissue, and distribution of some dietary Cu to the gill from the gut in Nile tilapia. This phenomena has been previously observed with high dietary Cu doses in fish (e.g. rainbow trout, Handy, 1996; Kamunde et al., 2001; grey mullet, Baker et al., 1998). Copper levels in the gill and intestine of Cu-exposed fish returned to control levels during the recovery phase, but hepatic Cu levels continued to increase postexposure (Table 1). This suggests that Nile tilapia can redistribute accumulated Cu for excretion via the liver. There are few studies of post-exposure recovery following dietary Cu exposure in freshwater fish, and most of these are on salmonids. However, Handy (1992) made similar observations in rainbow trout where livers of Cu-fed fish showed a 30% increase in the proportion of whole body Cu held in the organ during a 12-day recovery phase. Several authors have also noted redistribution of newly acquired Cu to the liver of trout during aqueous exposures (e.g. Grosell et al., 2001), and Clearwater et al. (2002) also argues that hepatobiliary excretion of Cu may be a dominant route for dietary Cu excretion for teleost fish. 4.2. Tissue electrolyte content and haematology The absence of treatment-dependent changes in tissue moisture content, total content of Na+ or K+ in tissues (Table 2), and HCT suggest that dietary Cu did not cause any major osmotic disturbances (as in trout, Handy et al., 1999). Haemoglobin and HCT

B.J. Shaw, R.D. Handy / Aquatic Toxicology 76 (2006) 111–121

118

Table 2 Total sodium and potassium concentrations, and moisture content in gill, liver and intestine of Nile tilapia fed a normal diet (control, 3 mg Cu kg−1 food) or a diet containing elevated copper (2000 mg Cu kg−1 food) for 42 days, followed by a recovery period on normal food for a further 21 days Experimental phase

Sampling day

Treatment

Tissue Gills

Liver

Intestine

Sodium Time-zero Cu exposure Cu exposure Recovery Recovery

0 42 42 63 63

Initial fisha Control Elevated Cu Control Elevated Cu

393 256 333 247 264

± ± ± ± ±

28 (8) 16 (6)# 31 (6) 11 (6)# 19 (6)#

120 77 74 87 93

± ± ± ± ±

21 (8) 6 (6) 6 (6) 9 (6) 8 (6)

288 279 283 238 248

± ± ± ± ±

32 (8) 34 (6) 29 (6) 9 (6) 26 (6)

Potassium Time-zero Cu exposure Cu exposure Recovery Recovery

0 42 42 63 63

Initial fisha Control Elevated Cu Control Elevated Cu

212 176 179 152 153

± ± ± ± ±

14 (8) 4 (6) 13 (6) 6 (6)#,+ 8 (6)#

259 223 235 255 266

± ± ± ± ±

40 (8) 13 (6) 13 (6) 10 (6) 17 (6)

222 241 301 259 267

± ± ± ± ±

33 (8) 31 (6) 20 (6) 9 (6) 17 (6)

Moisture (%) Time-zero Cu exposure Cu exposure Recovery Recovery

0 42 42 63 63

Initial fisha Control Elevated Cu Control Elevated Cu

81.5 73.9 81.8 74.7 74.3

± ± ± ± ±

1.0 (8) 3.6 (6) 1.0 (6) 0.6 (6) 0.6 (6)

74.6 70.0 70.0 72.3 73.3

± ± ± ± ±

2.5 (8) 1.2 (6) 0.8 (6) 1.5 (6) 1.3 (6)

76.8 81.1 78.1 79.3 80.4

± ± ± ± ±

3.8 (8) 1.3 (6) 1.2 (6) 0.7 (6) 1.3 (6)

Data are means ± S.E. (n is the number of fish), expressed as ␮mol g−1 dry weight of tissue for Na+ and K+ . There were no Cu-dependent effects on tissue Na+ or K+ at any time point (ANOVA, P > 0.05). No statistical difference in moisture content was observed between control and Cu-exposed fish, or over time within treatments (ANOVA, P > 0.05). a Initial fish were sampled immediately prior to starting the dietary Cu exposure. # Significantly different from initial fish (ANOVA, P < 0.05). + Significantly different in recovery phase compared to 42 days within the control treatment (ANOVA, P < 0.05).

ranges reported here for Nile tilapia (5.6–6.2 g dl−1 and 24–33%, respectively) are very similar to previous reports for tilapia. Shiau and Ning (2003) give values for haemoglobin and HCT ranging from 5.2 to 6.3 g dl−1 and 26–36%, respectively for Cu diets containing 1–5 mg Cu/kg feed. Altwood et al. (2003) give HCT in the range of 23–30% depending on diet formulation and water temperature. The time-dependent increase in HCT from 24% in initial fish to about 32% after 6 weeks in this study, regardless of Cu exposure and without anaemia, is typical of normal growth effects on haematology in fish (Houston, 1997) and has been reported during dietary Cu studies in rainbow trout (Handy et al., 1999). The absence of elevated blood glucose also supports the notion that fish in our study did not suffer overt osmoregulatory stress. Blood glucose values in this study (3–4 mmol l−1 ) were similar to reports for Nile tilapia also held at 25 ◦ C (3 mmol l−1 , Altwood et al., 2003), and within the normal resting

glucose range suggested for unstressed rainbow trout (2–7 mmol l−1 , Hille, 1982). 4.3. Growth and nutritional performance Nile tilapia fed excess dietary Cu showed reduced growth by day 21 of Cu exposure. This persisted throughout the experiment (Fig. 1) and was reflected by a decrease in SGR in the Cu-fed fish compared to controls (Table 3). The reduction in growth in Nile tilapia, is most likely explained by reduced food intake during Cu exposure (Fig. 1). Poor absorption or assimilation of the major nutrients is unlikely given that FCR, condition factor, and intestinal morphology remained similar for both treatments throughout the experiment (Table 3 and Fig. 2). Several authors have noted reductions in growth rate during dietary Cu exposure in fish (Baker et al., 1998; Clearwater et al., 2002), but others have not (Lanno et al., 1985; Handy et al., 1999; Kamunde

B.J. Shaw, R.D. Handy / Aquatic Toxicology 76 (2006) 111–121

119

Table 3 Growth and nutritional performance of Nile tilapia fed a normal diet (control, 3 mg Cu kg−1 food) or a diet containing elevated copper (2000 mg Cu kg−1 food) for 42 days, followed by a recovery period on normal food for a further 21 days Parameter Specific growth rate (% day−1 ) Condition factor (K, %) Hepatosomatic index (%) Mean ration size (% bw day−1 ) Feed conversion ratio

Experimental phase Treatment

Cu exposure (days 0–42)

Recovery (days 43–63)

Entire experiment (days 0–63)

Control Elevated Cu Control Elevated Cu Control Elevated Cu Control Elevated Cu Control Elevated Cu

1.58 1.20 1.86 ± 0.03 (6) 1.75 ± 0.03 (6) 1.25 ± 0.10 (6)# 1.40 ± 0.13 (6)# 1.36 ± 0.04 (36) 1.23 ± 0.03 (35)* 1.03 (2) 1.07 (2)

1.16 1.15 1.76 ± 0.08 (6) 1.79 ± 0.03 (6) 1.87 ± 0.04 (6)# 2.19 ± 0.33 (6)# 0.98 ± 0.08 (29)+ 0.88 ± 0.07 (29)*,+ 1.04 (2) 1.10 (2)

1.41 1.17 – – – – 1.24 ± 0.05 (63) 1.11 ± 0.04 (63)* 1.01 (2) 1.07 (2)

Data are means ± S.E. (n is the fish per treatment, except for ration size, where n is the number of daily measurements/treatment). Feed conversion ratio (FCR) and specific growth rate (SGR) are averages of duplicate tanks/treatment. Condition factor and hepatosomatic index of initial fish were 1.76 ± 0.07 (8), and 0.83 ± 0.10 (8), respectively (means ± S.E., n). Where no error bars are reported, data were calculated from bulk treatment data over time for n = 2 tanks. * Statistically different from the control within time point (t-tests, P < 0.05). # Statistically different from initial fish (ANOVA, P < 0.05). + Significant different in recovery phase compared to 42 days within treatment (t-tests, P < 0.05).

et al., 2001; Campbell et al., 2002), and Clearwater et al. (2002) argues that toxic effects on growth are best rationalised on an exposure dose basis. In rainbow trout at least, the threshold for dietary Cu toxicity on growth rate is about 664–730 mg Cu kg−1 dw feed, equating to a dosage of 35–45 mg Cu kg−1 bw day−1 (Clearwater et al., 2002). In our study, exposure dose was about 25 mg Cu kg−1 bw day−1 , based on a measured dietary Cu level of 1968 mg Cu kg−1 dw feed and ration of 1.27% bw day−1 . Using daily ingested dose for species comparisons of oral toxicity, and growth as a sub-lethal end point, dietary Cu appears to be more toxic to Nile tilapia than rainbow trout. This may be partly explained by differences in water temperature for the two species, since elevation of ambient temperature tends to increase dietary Cu uptake and tissue Cu accumulation (Clearwater et al., 2000). 4.4. Histopathology The gill and intestine did not show overt pathology (Fig. 2) during dietary Cu exposure and this is consistent with the few previous reports on temperate species such as rainbow trout (Handy et al., 1999; Kamunde et al., 2001). However, there was evidence of morphological change in the liver of tilapia. Two-thirds of the Cu-fed fish showed fatty change in the liver (Fig. 2).

This was characterised by increasing lipid deposits in the liver, and a consequent loss of sinusoid space. Increased lipidosis may also explain the slightly higher HSI in Cu-fed fish compared to controls (Table 3). Increased lipid content of the liver can be explained by either increased deposition of lipid in excess of nutritional requirements, or a failure to mobilise lipid stores during dietary Cu toxicity (Handy et al., 1999). The latter is more likely given that the notion of over feeding/excessive energy intake is excluded by the absence of changes in condition factor, decreased cumulative food intake, and reduced SGR during Cu exposure (Fig. 1, Table 3). This is the first report of fatty change in the liver of a warm water species of teleost during dietary copper exposure, but with only one other detailed report of fatty change in the liver of rainbow trout during dietary Cu exposure (Handy et al., 1999), it is impossible to make generalisations about differences between temperate and warm water species. However, both the livers of tilapia and rainbow trout show fatty change, and when food intake and SGR are considered, both species eventually showed responses that are consistent with the inability to mobilise hepatic glycogen stores. Notably, hepatic lipidosis became worse in the recovery phase in this study, when Cu accumulation by the liver of tilapia also continued post-exposure

120

B.J. Shaw, R.D. Handy / Aquatic Toxicology 76 (2006) 111–121

(Fig. 2, Table 1). This post-exposure effect has not been previously reported in fish. The ecological significance of post-exposure effects of dietary Cu on the fat metabolism in the liver are uncertain, but impacts on liver functions such as vitellogenesis and glycogen release cannot be excluded. In addition for tilapia aquaculture, we should perhaps be more cautious with sudden reductions in the Cu content of aquafeeds when changing diets, and monitor fish health more closely after accidental or therapeutic Cu exposures. In conclusion, Nile tilapia show dietary Cu toxicity in terms of hepatic and intestinal Cu accumulation, reduced growth and food intake. On a Cu dose-basis the Nile tilapia is more sensitive to Cu than rainbow trout with regard to growth depression. The continued hepatic Cu accumulation post-exposure, and lipidosis, raises concerns about possible delayed toxic effects of dietary Cu in fish after exposure.

Acknowledgements This study was partly supported by a grant from the Leverhulme Trust to R. Handy. Ben Shaw completed this work during the MRes Applied Fish Biology at the University of Plymouth. Technical support from Michael Hockings was appreciated. References Adham, K.G., Hassan, I.F., Taha, N., Amin, T.H., 1999. Impact of hazardous exposure to metals in the Nile and Delta lakes on the catfish, Clarias lazera. Environ. Monit. Assess. 54, 107–124. Ali, A., Al-Ogaily, S.M., Al-Asgah, N.A., Gropp, J., 2003. Effect of sub-lethal concentrations of copper on the growth performance of Oreochromis niloticus. J. Appl. Ichthyol. 19, 183–188. Altwood, H.L., Tomasso, J.R., Webb, K., Gatlin, D.M., 2003. Low temperature tolerance of Nile tilapia, Oreochromis niloticus: effects of environmental and dietary factors. Aquacult. Res. 34, 241–251. Arredondo, M., Uauy, R., Gonz´alez, M., 2000. Regulation of copper uptake and transport in intestinal cell monolayers by acute and chronic copper exposure. Biochim. Biophys. Acta 1474, 169–176. Baker, R.T.M., Handy, R.D., Davies, S.J., Snook, J.C., 1998. Chronic dietary exposure to copper affects growth, tissue lipid peroxidation, and metal composition of the grey mullet, Chelon labrosus. Mar. Environ. Res. 45, 357–365. Balirwa, J.S., 1992. The evolution of the fishery of Oreochromis niloticus (pisces: Cichlidae) in lake Victoria. Hydrobiologia 232, 85–89.

Barriga-Sosa, I.D.L.A., Jim´enez-Badillo, M.D.L., Ib´an˜ ez, A.L., Arredondo-Figueroa, J.L., 2004. Variability of tilapias (Oreochromis spp.) introduced in Mexico: morphometric, meristic and genetic characters. J. Appl. Ichthyol. 20, 7– 14. Berntssen, M.H.G., Hylland, K., Wendelaar Bonga, S.E., Maage, A., 1999. Toxic levels of dietary copper in Atlantic salmon (Salmo salar L.) parr. Aquat. Toxicol. 46, 87–99. Burke, J., Handy, R.D., 2005. Sodium sensitive and insensitive copper accumulation by isolated intestinal cells of rainbow trout, Oncorhynchus mykiss. J. Exp. Biol. 208, 391– 407. Bury, N.R., Walker, P.A., Glover, C.N., 2003. Nutritive metal uptake in teleost fish—review. J. Exp. Biol. 206, 11–23. Campbell, H.A., Handy, R.D., Sims, D.W., 2002. Increased metabolic cost of swimming and consequent alterations to circadian activity in rainbow trout (Oncorhynchus mykiss) exposed to dietary copper. Can. J. Fish. Aquat. Sci. 59, 768– 777. Clearwater, S.J., Baskin, S.J., Wood, C.M., McDonald, D.G., 2000. Gastrointestinal uptake and distribution of copper in rainbow trout. J. Exp. Biol. 203, 2455–2466. Clearwater, S.J., Farag, A.M., Meyer, J.S., 2002. Bioavailability and toxicity of dietborne copper and zinc to fish. Comp. Biochem. Physiol. 132C, 269–313. Co˘gun, H.Y., Kargin, F., 2004. Effects of pH on the mortality and accumulation of copper in tissues of Oreochromis niloticus. Chemosphere 55, 277–282. Getachew, T., 1988. Digestive efficiency and nutrient composition gradient in the gut of Oreochromis niloticus L. in lake Awasa, Ethiopia. J. Fish Biol. 33, 501–509. Grosell, M.H., McGeer, J.C., Wood, C.M., 2001. Plasma copper clearance and biliary excretion are stimulated in copperacclimated trout. Am. J. Physiol. 280, R796–R806. Handy, R.D., 1992. The assessment of episodic pollution. II. The effects of cadmium and copper enriched diets on tissue contaminant analysis in rainbow trout (Oncorhynchus mykiss). Arch. Environ. Contam. Toxicol. 22, 82–87. Handy, R.D., 1996. Dietary exposure to toxic metals in fish. In: Taylor, E.W. (Ed.), Toxicology of Aquatic Pollution. Cambridge University Press, Cambridge, pp. 29–60. Handy, R.D., 2003. Chronic effects of copper exposure versus endocrine toxicity: two sides of the same toxicological process? Comp. Biochem. Physiol. 135A, 25–38. Handy, R.D., Sims, D.W., Giles, A., Campbell, H.A., Musonda, M.M., 1999. Metabolic trade-off between locomotion and detoxification for maintenance of blood chemistry and growth parameters by rainbow trout (Oncorhynchus mykiss) during chronic dietary exposure to copper. Aquat. Toxicol. 47, 23– 41. Handy, R.D., Musonda, M.M., Phillips, C., Falla, S.J., 2000. Mechanisms of gastrointestinal copper absorption in the African walking catfish: copper dose-effects and a novel anion-dependent pathway in the intestine. J. Exp. Biol. 203, 2365–2377. Harrison, M.D., Dameron, C.T., 1999. Molecular mechanisms of copper metabolism and the role of the Menkes disease protein. J. Biochem. Mol. Toxicol. 13, 93–106.

B.J. Shaw, R.D. Handy / Aquatic Toxicology 76 (2006) 111–121 Hille, S., 1982. A literature review of the blood chemistry of rainbow trout, Salmo gairdneri rich. J. Fish. Biol. 20, 535–569. Houston, A.H., 1997. Review: are the classical hematological variables acceptable indicators of fish health? Trans. Am. Fish. Soc. 126, 879–894. Kamunde, C.N., Grosell, M., Lott, J.N.A., Wood, C.M., 2001. Copper metabolism and gut morphology in rainbow trout (Oncorhynchus mykiss) during chronic sub-lethal dietary copper exposure. Can. J. Fish. Aquat. Sci. 58, 293–305. Kamunde, C., Wood, C.M., 2003. The influence of ration size on copper homeostasis during sub-lethal dietary copper exposure in juvenile rainbow trout, Oncorhynchus mykiss. Aquat. Toxicol. 62, 235–254. Khallaf, E.A., Galal, M., Authman, M., 1998. Assessment of heavy metals pollution and their effects on Oreochromis niloticus in aquatic drainage canals. J. Egypt. Ger. Soc. Zool. 26B, 39–74. Khallaf, E.A., Galal, M., Authman, M., 2003. The biology of Oreochromis niloticus in a polluted canal. Ecotoxicology 12, 405–416. Knox, D., Cowey, C.B., Adron, J.W., 1982. Effects of dietary copper and copper:zinc ratio on rainbow trout Salmo gairdneri. Aquaculture 27, 111–119.

121

Lanno, R.P., Slinger, S.J., Hilton, J.W., 1985. Maximum tolerable and toxicity levels of dietary copper in rainbow trout (Salmo gairdneri Richardson). Aquaculture 49, 257–268. Likongwe, J.S., Stecko, T.D., Stauffer, J.R., Carline, R.F., 1996. Combined effects of water temperature and salinity on growth and feed utilization of juvenile Nile tilapia Oreochromis niloticus (Linnaeus). Aquaculture 146, 37–46. Linder, M.C., 1991. Biochemistry of Copper. Plenum Press, New York. Lundebye, A.K., Berntssen, M.H.G., Wendelaar Bonga, S.E., Maage, A., 1999. Biochemical and physiological responses in Atlantic salmon (Salmo salar) following dietary exposure to copper and cadmium. Mar. Pollut. Bull. 39, 137–144. Pe˜na, M.M.O., Lee, J., Thiele, D.J., 1999. A delicate balance: homeostatic control of copper uptake and distribution. J. Nutr. 129, 1251–1260. Shakweer, L.M., 1998. Concentration levels of some trace metals in Oreochromis niloticus at highly and less polluted areas of Mariut lake. J. Egypt. Soc. Zool. 25A, 125–144. Shiau, S.Y., Ning, Y.C., 2003. Estimation of dietary copper requirements of juvenile tilapia, Oreochromis niloticus × O. aureus. Anim. Sci. 77, 287–292.