Relationship between biological characteristics of fish and their contamination with trace metals : a case study of perch Perca fluviatilis L . in the Baltic Sea

Biological characteristics of fish are expected to affect the bioaccumulation patterns of trace metals in aquatic organisms. Nonetheless, this topic is relatively unstudied among fish species. In this study we explored how biological characteristics such as sex, maturity, and age of perch affect the contamination of their liver and muscle tissue with mercury, cadmium, copper, and zinc. According to our analyses, the contamination of fish with trace metals was complex and depended on the one hand on the type of metal and on the other hand on the studied fish biological characteristics. In the presence of significant statistical differences the concentration of trace metals was higher in males than in females and in immature than in mature individuals. Mercury was the only trace metal that accumulated with age in the fish. However, no generic relationships between the studied variables were found, and this may hint at the lack of such a relationship.


*
The trace metal loading into aquatic ecosystems increases with elevated industrial and traffic emissions and intensified land use practices.This may lead to severe ecological consequences, e.g.changes in growth, development, reproductive potential, and survival of fish (Sorensen, 1991;Szefer, 2002;Webb et al., 2006).Besides, some trace metals and their compounds are carcinogenic (Janssen et al., 2000;Diaconescu et al., 2008).Thus, the accumulation of trace metals in fish tissues does not affect only fish populations, but also has repercussion on their consumers, including humans.Considering the importance of fish in the human diet, consumption of contaminated fish poses a significant threat to human health (Schmitt et al., 2006;Klavins et al., 2009).Therefore, it is essential to better understand the trophic pathways of trace metals and their bioaccumulation in aquatic food webs.nen et al., 1994, 1998;Simm and Kotta, 2000;Voigt, 2000Voigt, , 2002Voigt, , 2003) ) but also in perch (Voigt, 2001;Szefer, 2002;Szefer et al., 2003;Bignert et al., 2011).Nevertheless, the majority of studies just report values without giving deeper insight into how the accumulation of trace metals in fish tissues is related to fish biological characteristics.
Perch (Perca fluviatilis L.) is a widely distributed fish being native both in temperate fresh-and brackishwater ecosystems in the northern hemisphere except America.Its muscle tissue is lean and contains only about 1% fat (Bignert et al., 2011), which makes perch highly demanded as a diet food on markets.Perch is the most dominating commercial fish species in Estonian coastal fishery, particularly in the Pärnu Bay area (Järv, 1996;Kotta et al., 2008).As it is a relatively sedentary species (Kipling and Le Cren, 1984;Böhling and Lehtonen, 1985;Järv, 2000), the concentration of contaminants in perch reflects the water quality in the ambient environment.Therefore, perch was recommended as an indicator species to assess the biological effects of environmental pollution (HELCOM, 2011).
The present study describes how the contamination of different tissues with cadmium, copper, zinc, and mercury depends on the age, sex, and maturity of perch.Our hypotheses were as follows: (1) the content of trace metals differs among sexes, (2) the contamination of perch increases during the maturation process, and (3) the bioaccumulation of trace metals in perch tissues depends on their age.

MATERIAL AND METHODS
The study was conducted in Pärnu Bay, located in the northeastern part of the Gulf of Riga, the Baltic Sea.The bay is shallow (4-5 m) and semi-exposed.It receives yearly 1.6 km 3 of fresh water from the Pärnu River; this amount corresponds approximately to the volume of the bay.Both agricultural and municipal pollution loads to Pärnu Bay are high and the bay is considered as one of the most polluted basins in the Baltic Sea range in terms of hazardous substances, including trace metals and numerous persistent organic pollutants such as polychlorinated biphenyls, DDT, dioxins, furans, etc. (HELCOM, 2001, 2010).
Perch were collected from commercial trap nets with a mesh size of 16−36 mm (bar length) at three stations in the northeastern part of Pärnu Bay adjacent to the mouth of the Pärnu River (Fig. 1, Appendix).The sampling was performed during October and November 2006 and in March 2007 in order to cover the main maturation periods of perch.Altogether 344 perch were analysed.Fish total length (TL, cm), weight (TW, g), sex, and maturity stage to a routine six-point macroscopic histological maturity scale (Anon., 2007) were estimated.The sampled fish were aged using the opercular bones according to Tesch (1971).
The analyses were performed at the Laboratory of Chemical Analyses, Tallinn University of Technology.The laboratory has accreditation No. L116 since 2003 by the Estonian Accreditation Centre for the analyses of the studied elements.The laboratory participates annually (and achieves satisfactory results) in an interlaboratory comparison programme organized by QUASIMEME.The quality assurance for each series of analyses was provided by the parallel analysis of reference materials.
For the determination of trace metals fish were dissected and representative tissues, such as the liver and muscle, were weighed and two replicate sub-samples of each sample were then prepared (1-2 g of wet weight).Samples (n = 79) were processed in an Automatic Microwave Digestion System (AntonPaar Multiwave 3000) using concentrated HNO 3 Suprapur1 "Merck".
The concentration of zinc (Zn) and copper (Cu) were determined by the flame technique of Atomic Absorption Spectrophotometry (AAS) method (Spectra AA 220F; Varian, Australia).The concentration of cadmium (Cd) was determined using a flameless technique and the concentration of mercury (Hg) by cold vapour AAS.
To check for contamination, blanks containing bidistilled water were used after every five samples.The limits of detection for the analysis of Cu and Zn were 1.0 and 0.7 mg g -1 and of Cd and Hg 0.13 and 0.3 µg g -1 , respectively.
The general linear model (GLM) was used to seek tissue-specific (levels: liver, muscle) statistical relationship between trace metal concentrations, fish age, and maturity as well as to assess how the metal concentrations vary among sexes.For each metal and tissue two models were built: one model included sex and maturity as independent factors and metal concentrations as dependent variables; the other model included only mature individuals, and sex was as an independent factor, age a covariate, and metal concentrations dependent variables.Linear regression analysis was used to visualize relationships between co-variants and dependent variables.

RESULTS
The studied perch ranged between 9 and 5 cm in total length, 7 and 506 g in weight, and 1 and 8 years in age.Both females and males contained immature (maturity stage II) and mature individuals (maturity stage III and IV).As a rule, the immature female perch were smaller and mature female larger than males of the same age.The average biological parameters are reported in Table 1.
The first set of GLM analyses on the effects of sex and maturity on the concentration of trace metals in fish showed that largely half of the studied metals were independent of fish sex and maturity.However, sex, maturity, and their interaction explained the variability of Cd in both liver and muscle tissues.Besides, only sex explained the variability of Cu in muscle tissue (Table 2).In the presence of significant statistical differences the concentration of trace metals was higher in males than in females.As for Cd, such differences were due to immature individuals as there was no statistical difference between sexes among mature individuals (Figs 2 and 3).
The other set of GLM analyses on the effects of sex and age on the concentration of trace metals in fish showed that the accumulation of trace metals into fish tissues was independent of fish age.As the only exception the concentration of Hg in muscle tissue increased with fish age (Table 3, Fig. 4).

DISCUSSION
Our study only partly agreed with the first and second hypotheses as there were no generic relationships between fish sex and maturity and the content of trace metals in the fish tissues.Among the studied metals only Cd and Cu showed some sex dependence with males having higher metal concentrations compared to females.Just to name a few, membrane permeability, the nature of enzyme system and hormones, and the number of available binding sites in male and female fish may account for these differences (Heath, 1987; Fig. 4. Linear regression analysis to visualize the relationships between fish age and content of mercury in muscle tissue.Jørgensen and Pedersen, 1994;Madenijan et al., 2011).Besides, Cu and Zn tend to accumulate to a greater extent than other trace metals, which increases the statistical probability that there are significant differences among sexes (Heath, 1987).In general, the concentrations of the studied trace metals were higher in liver than in muscle conforming to the earlier idea that metals primarily accumulate into tissues that are metabolically more active (Berninger and Pennanen, 1995;Klavins et al., 2009).According to Pihu et al. (2003), the asymptotic growth of female perch is always faster compared with males in both fresh-and brackish-water environments, which means that the females of a given size category are younger than the males of the same length.Thus, provided no variability in fish size, males are expected to have higher metal concentrations than females.Besides, the lower trace metals content in female perch may be a consequence of their stronger immune system (Zeeman and Brindley, 1981;Dautremepuits et al., 2009).
Our study also showed that such sex dependence was stronger in immature than mature individuals with immature males accumulating trace metals at significantly higher rates than immature females.The maturation process requires enhanced energy levels, i.e. increases feeding rates.This results in the elevated metabolic rate and, thus, the raised metal accumulation of fish (Bobori and Economidis, 1996).Although differences in trace metal accumulation among males and females may persist at mature developmental stages, spawning seems to be the mechanism that helps the elimination of toxic substances from the body (Simm and Kotta, 2000;Szefer, 2002;Roots et al., 2004).Consequently, for mature individuals the differences in the concentration of trace metals among sexes decrease down to the level that is not detectable by statistical models.
There exists a large regional variability in the accumulation of trace metals in perch tissues.Earlier studies demonstrated higher metal concentrations either in males (Szefer et al., 2003;Voigt, 2003;Tulonen et al., 2006) or females (Berninger and Pennanen, 1995;Klavins et al., 2009).We believe that the among-study differences are not caused by sex-specific differences in foraging behaviour (Järv et al., 2011).Instead, the spatial differences in fish metabolism may result in the spatially varying sex-specific metal accumulation.
Our study also partly agreed with the third hypothesis as Hg was the only trace metal that accumulated with age in the fish.In general fish, especially predators, have a natural tendency to concentrate Hg in their bodies, often in the form of methylmercury, a highly toxic organic compound of Hg.Since fish are less efficient at depurating than accumulating methylmercury, the concentration of Hg in fish tissues is expected to increase with fish age (Janssen et al., 2000;Tulonen et al., 2006;Klavins et al., 2009;Gewurtz et al., 2011;Lepom et al., 2012).

CONCLUSIONS
To conclude, our analyses identified that the contamination of fish with trace metals is complex and depends on the one hand on the type of metal and on the other hand on fish biological characteristics such as tissue, sex, maturity, and age.However, no generic relationships between the studied variables were found either due to the lack of such relationships or potentially due to confounding environmental variables not identified in this study.Thus, experimental studies are needed to quantify the relationship between metal concentrations in the environment (including fish prey), fish biological characteristics, and the accumulation of trace metals in fish tissues.

ACKNOWLEDGEMENTS
We acknowledge MSc Kristiina Fuch, who performed analyses at the Laboratory of Chemical Analyses, Tallinn University of Technology.Funding for this research was provided by Institutional research funding IUT02-20 of the Estonian Research Council and by the Estonian Science Foundation under grants 7813 and 8254.The study was partly supported by the project 'The status of marine biodiversity and its potential futures in the Estonian coastal sea' 3.2.0802.11-0029 of the Environmental protection and technology programme of the European Regional Fund.

APPENDIX
Biological characteristics of perch and the concentration (mg kg wet weight -1 ) of trace metals in their liver and muscle tissue.F -female, M -male
Fig. 2. GLM analysis on the effect of sex and maturity on the content of cadmium in liver tissue (means and 95% confidence intervals).

Fig. 3 .
Fig. 3. GLM analysis on the effect of sex and maturity on the content of cadmium and copper in muscle tissue (means and 95% confidence intervals).

Table 2 .
Results of the general linear model analyses on the effect of sex and maturity on the content of trace metals in fish tissues.Significant effects and interactions are marked in bold

Table 3 .
Results of the general linear model analyses on the effect of sex and age on the content of trace metals in adult fish muscle tissue.Significant effect is marked in bold