Due to the toxic effects many heavy metals display even at very low concentrations, heavy metal pollution poses a serious threat to human health and the environment (Suárez-Serrano et al. 2010; Agrawal et al. 2011). Consequently, it is important to gain an understanding of the presence and distribution of heavy metals in the environment, especially in aquatic ecosystems where heavy metal pollution can be particularly dangerous. Bioindicator organisms are well-suited for monitoring heavy metal pollution and its impact on the environment. Prior studies indicate that crayfish can accumulate heavy metals with rates depending on the external concentration of the metal (Guner 2007; Soedarini et al. 2012). Due to this potential, crayfish have been used as bioindicator species in many studies (Lopez et al. 2004; Hothem et al. 2007; Hagen and Sneddon 2009). Through their position in the food web, crayfish also have the potential for transferring toxins and contaminants to other organisms of higher trophic levels (Wigginton and Birge 2007).

To improve our capacity to accurately detect and monitor heavy metal pollution, it is important to understand all factors that influence the accumulation of metals in crayfish. Many studies have determined the accumulation of metals in crayfish tissues in a dose and/or time dependent manner, without taking into account significant differences in tissue concentration levels of heavy metals or the possibility of selective accumulation of a specific heavy metal in males or females (Kouba et al. 2010). Gender is one of the most important factors bearing a potential effect on heavy metal accumulation. The aim of this study was to understand the differences in heavy metals and metalloid accumulation by crayfish (Astacus leptodactylus) tissues, with special focus on the distribution trends between male and female crayfish. We also sought to examine the interrelationships between metals, and to model the accumulation for male and female crayfish separately.

Materials and Methods

Sixty crayfish specimens (A. leptodactylus), 37 males and 23 females, of varying sizes were collected from Lake Hirfanlı, a dam lake located in Kırşehir, Turkey (39° 16′ 20.28″ N, 33° 31′ 7.68″ E) and transported in Igloo cooler boxes. Crayfish specimens were stored at –20°C in plastic bags until cephalothorax, exoskeleton, gills, hepatopancreas and abdominal muscles were dissected. Samples of those tissues were then prepared for ICP-MS measurements by digestion, following the method of Bernhard (1976). An X-Series II ICP-MS (Thermo Fisher Scientific Advanced Mass Spectrometry, Bremen, Germany), equipped with ID100 Autodiluter and Cetac Asx-260 autosampler accessories, was utilized. Dilutions were made with a 2 % nitric acid matrix in ultrapure water. All standard curves were prepared by using the QCS-27 series of elements (High Purity Standards, South Carolina, USA). Concentrations of the relevant element in the tissue samples were taken into account and a correlation coefficient over 0.99 was obtained for each calibration curve. Measurements of standards were conducted after every 20 samples to ensure consistency, and 10 ppb 209Bi was utilized as an internal standard. Interferences created by IA and IIA metals were removed via Collision Cell Technology (CCT) (Pick et al. 2010). Three runs were conducted for each sample. The dwell time was 10 ms for all elements except Al, for which a dwell time of 50 ms was used due to this element’s low atomic weight. Sampling and washing times were 90 s each. 27Al, 52Cr, 53Cr, 55Mn, 60Ni, 63Cu, 65Cu and 75As were the heavy metals measured. LUTS-1 non-defatted lobster hepatopancreas (National Research Council Canada, Ontario, Canada) was utilized as reference material; sample measurements were adjusted using the recovery rates obtained. Limits of detection were 0.037 ± 0.019, 0.094 ± 0.009, 0.031 ± 0.001, 0.048 ± 0.002, 0.051 ± 0.025, and 0.005 ± 0.025 ppb for Al, Cr, Mn, Ni, Cu and As; respectively. No sample was found to have any metal concentration below detection limits. Recovery rates were 99.0 %, 95.36 %, 96.14 %, 115.6 %, 155.02 % and 15.85 % for Al, Cr, Mn, Ni, Cu and As; respectively. High Cu recovery rates are probably caused by interference by 25Mg and 40Ar. Very low As recovery rates are attributable to the low temperature of sublimation for this element (613 °C), as the extraction method utilized substantially higher temperatures.

SPSS 17 (IBM, Portsmouth, UK) was used for all statistical analysis. Prior to one-way ANOVA and independent t test analysis, all results were subjected to the Kolmogorov–Smirnov Test to observe the normality of data distribution. A logarithmic transformation was applied for non-parametric data and all results were subjected to a second Kolmogorov–Smirnov Test to pinpoint non-parametric data after transformation. For determining the homogeneity of variance, Levene’s test was carried out. Tukey’s test was utilized for data with homogeneous variance, and Tamhane’s test was applied for data with heterogeneous variance. For non-parametric data, a Mann–Whitney U Test was carried out following a Kruskal–Wallis Test (Barrento et al. 2008). In this study, hierarchical clustering methods based on a Euclidean distance measure and Ward’s hierarchical agglomerative clustering technique were used (Lopez et al. 2004). The data were standardized following a z-score method. Multiple linear regression analysis was used to predict the concentrations of heavy metals in each of the four tissues tested. The analysis was applied to all heavy metals in all tissues; however, only models displaying R2 > 0.85 were considered in this paper. Normality graphs were inspected to observe the parameter distribution. A stepwise method was chosen during analysis.

Results and Discussion

The highest concentrations of elements were generally obtained in gill tissue for both genders (Tables 1a, b, 2a, b; Fig. 1a, b), likely because of the direct contact of the gills with the outside environment (Kurun et al. 2010). The active role that gills play in the regulation of ionic balance might be another reason (Alcorlo et al. 2006). For Al, in both male and female specimens, accumulation differences between every tissue were statistically significant, except the difference between the exoskeleton and abdominal muscle. In males, the order of accumulation of Al was gills > exoskeleton ≈ abdominal muscle > hepatopancreas. In females, the order of Al accumulation was gills > exoskeleton = abdominal muscle > hepatopancreas. Gills were previously reported to play an important role in Al intake, and accumulation of this metal was also observed in hepatopancreas and muscles to a lesser degree (Alexopoulos et al. 2003; Kurun et al. 2010). While those observations are supported in this study, we have also observed a relatively high Al accumulation in the exoskeleton, comparable to the muscle accumulation rates. The high exoskeletal concentration of Al observed may be explained by the fact that the specimens were collected in autumn. Due to the relatively lower ambient temperatures, the metabolic rates of crayfish are slower and moulting is expected to be less frequent. As such, greater amounts of Al may have accumulated on the exoskeleton between moults. Al is also a common sediment constituent and its widespread presence may further account for the high exoskeletal Al content. Increased exposure times to heavy metals have previously been reported to cause higher accumulation rates in the exoskeleton (Guner 2007).

Table 1 Means of comparison for males
Full size table
Table 2 Means of comparison for females
Full size table
Fig. 1
figure 1

a Metals and metalloid trends for male specimens. b Metals and metalloid trends for female specimens

Full size image

Cr is an important trace element for many organisms, but has toxic and mutagenic effects in higher concentrations (Srinath et al. 2002). Cr+6 in particular is very toxic, and generally not utilized in biological systems (Vinod et al. 2010). In male specimens, 52Cr accumulations were significantly different for each tissue. While similar trends in accumulation were observed in females, no significant difference was present between the hepatopancreas and muscle accumulation rates. In both sexes, gills and the exoskeleton were the prime sites of chromium accumulation, a result previously reported in laboratory studies (Bollinger et al. 1997). In males, the order of accumulation of Cr was gills > exoskeleton > hepatopancreas > abdominal muscle. In females, the order was gills > exoskeleton > hepatopancreas ≈ abdominal muscle. Statistically significant differences were observed between the 53Cr accumulations of each tissue of male specimens, while the female cohort was similar except for the lack of a significant difference between the accumulation rates of the exoskeleton and abdominal muscles. It is curious that the hepatopancreas accumulation was greater than exoskeletal accumulation for 53Cr, while the opposite was true for 52Cr. This trend might be caused by fractionation during absorption of chromium, especially if the exoskeleton can selectively absorb 52Cr over 53Cr. In males, the order of 53Cr accumulation was gills > hepatopancreas > exoskeleton > abdominal muscle. In females, 53Cr was accumulated in the following order: gills > hepatopancreas > exoskeleton ≈ abdominal muscle.

Mn yielded one of the highest accumulation rates in every tissue, likely because of its status as an essential element (Baden and Eriksson 2006). However, previous studies indicate that Mn accumulation in high concentrations may also have toxic effects (Baden and Eriksson 2006; Becquer et al. 2010). No significant difference was observed between the Mn accumulations in the hepatopancreas-exoskeleton and gills-muscle of male specimens. Other tissues had statistically significant differences. In the female cohort, differences between the Mn accumulation rates of abdominal muscle and both hepatopancreas and the exoskeleton were not significant. Mn accumulation was reported to be the most prevalent in gills or exoskeleton (Kurun et al. 2010; Naghshbandi et al. 2007). However, the greatest accumulation amounts were observed in the hepatopancreas for this study. In males, Mn was accumulated in the order: exoskeleton ≈ hepatopancreas > gills ≈ abdominal muscle. In females, differences between tissue accumulation amounts were not pronounced enough to draw statistically significant conclusions, though hepatopancreas ≈ exoskeleton > abdominal muscle > gills was the general trend.

It is still not well known whether Ni is essential for crayfish (Khan and Nugegoda 2003; Yilmaz and Yilmaz 2007). However, it constitutes a toxic environmental hazard at high concentrations, retarding growth and preventing reproductive activity (Khan and Nugegoda 2003). In male specimens, significant differences were observed in Ni accumulation for each tissue, except between the exoskeleton and hepatopancreas. In the female cohort, every tissue accumulated significantly more Ni than the abdominal muscle, but no other statistically significant difference was observed. Previous studies support our observations on Ni accumulation preferences by tissue (Khan and Nugegoda 2003). In males, Ni accumulation followed the order: gills > exoskeleton ≈ hepatopancreas > abdominal muscle. In females, the general order of Ni accumulation was gills: exoskeleton ≈ hepatopancreas ≈ gills > abdominal muscle, although the gills—abdominal muscle connection does not have statistical support.

63Cu accumulation yielded no significant differences across the tissues of female specimens, while only the differences between the hepatopancreas and other tissues were significant for the male cohort. In males, the order of accumulation of 63Cu was: gills ≈ exoskeleton ≈ abdominal muscle > hepatopancreas. In females, no significant difference was observed between the 63Cu accumulation levels of each tissue. For the male cohort, hepatopancreatic accumulation of 65Cu was significantly lower compared to gills and the exoskeleton, while no statistically meaningful difference was observed between abdominal muscles and the hepatopancreas. In addition, gill tissue accumulated significantly more 65Cu than the abdominal muscles. Tissue accumulations of this metal exhibited no significant differences in females. In males, 65Cu accumulation levels did not display large enough differences to yield a complete, statistically significant accumulation order, but gills > exoskeleton > abdominal muscle > hepatopancreas was the generally observed result. In females, no significant difference was observed between the 65Cu accumulation levels in any tissue. Differences in copper absorption between the genders are notable, as copper is a component of the carrier protein hemocyanin and therefore constitutes an essential element for crustaceans. The differences in copper accumulation between male and female specimens may be due to the time period at which the specimens were collected. As female specimens were ovigerous during the collection period, the eggs could have been supplied with copper from the female’s tissues, thus explaining the depletion of copper in the latter and the lack of significant differences detected between tissues due to the low concentrations of copper. In the male cohort, copper accumulated primarily in the gills and the exoskeleton. Similar results indicating the gills as prime sites of copper absorption exist in the literature (Kurun et al. 2010), although the hepatopancreas was also reported as a site of absorption (Alcorlo et al. 2006; Naghshbandi et al. 2007). Also for female cohorts, accumulation rates between 63Cu and 65Cu were not significantly different in gills, hepatopancreas and muscle tissue, while for male cohorts significant differences were only lacking in gill tissue.

The toxicity of As depends primarily on the element’s valance state, with inorganic As (As3+ and As5+) displaying the greatest toxicity (Batista et al. 2011; Li et al. 2011). Our results indicated that the male cohort had no significant differences in the accumulation of this metalloid between the exoskeleton and the hepatopancreas, as well as between the gills and the hepatopancreas. Other accumulation differences were statistically significant. Abdominal muscle accumulated substantially less As compared to the other tissues in females, but no other significant results were apparent. The hepatopancreas was previously indicated as the major site of As accumulation (Mary Bitner Anderson et al. 1997a; Alcorlo et al. 2006). In males, As accumulation differences were not significant. The observed general order was hepatopancreas > exoskeleton = gills > abdominal muscle. In the female cohort, the accumulation results likewise could not be used to produce a statistically meaningful accumulation order, with hepatopancreas > gills = exoskeleton > abdominal muscle emerging as the general trend. As reported in a number of previous works, even in the same species of crustacean, the same heavy metals can preferentially accumulate in different tissues. The primary cause of this phenomenon is thought to be the concentration of the metal in question and the time period in which the animals were exposed to the metal (M. B. Anderson et al. 1997b; Bollinger et al. 1997). The effect of each element on the accumulation of other elements might be another reason for this situation.

No differences were observed between the male and female cohorts for the accumulation rates of the same heavy metal in the exoskeleton, hepatopancreas and the abdominal muscle. However, female crayfish accumulated significantly less Ni and As in their gill tissue compared to the male cohort (it must be noted that female specimens were ovigerous during the collection period in this study). This result is supported by the results of Bondgaard et al. (2000), who reported that Cd exposure during ovarian maturation in Carcinus maenas resulted in diminished Cd uptake of gills (Bondgaard et al. 2000). Females may have mechanisms of detoxification not present in males, such as toxins associated with circulating haemolymph lipoproteins that are incorporated into developing oocytes (Lee 1993). This suggests that the transfer of heavy metals into oocytes may relieve females of some of the effects of toxic metals (Martín-Díaz et al. 2006). This conclusion is supported by a strong correlation we have previously observed between Ni and As accumulations in female specimens, especially in gill tissue (R2 = 0.826) (Tunca et al. 2013).

Cluster analysis yielded three closely related clusters, with Al and Mn relatively distant from the more closely related pairs (Fig. 2a, b). Despite slight differences in their sorption trends, 63Cu and 65Cu were recovered in a very close cluster ((63Cu, 65Cu) = 4.626 for males, 4.008 for females). 52Cr and 53Cr likewise formed a single cluster with relatively low Euclidean distances between the isotopes ((52Cr, 53Cr) = 6.918 for males, 5.267 for females). Ni and As were recovered together, possibly because crayfish utilize the same methods to detoxify those elements ((Ni, As) = 8.517 for males, 6.379 for females). It is notable that Ni and As both displayed a lack of accumulation in female crayfish tissues compared to males. It is apparent that Al has the most unusual accumulation profile among the studied metals, as Al constitutes a cluster by itself and this cluster bears great distances to other metals for both genders (e.g. (Al, 52Cr) = 12.910 for males, 10.014 for females). This may be because of the low toxicity of Al or its status as a trivalent cation. The latter explanation would also account for the presence of Cr as the element clustered closest to Al, since Cr is generally trivalent in organic systems.

Fig. 2
figure 2

a Dendrogram of Cluster Analysis (CA) for male specimens. b Dendrogram of Cluster Analysis (CA) for female specimens

Full size image

Regression analysis was also applied to the elements in all tissues to determine accumulation models. Models with coefficients of determination (R2) > 0.85 were included (Tables 3a, b). Concentrations of Cu and Cr isotopes could be predicted in many cases, possibly because each individual isotope formed a basis for prediction of the other. Close clustering of isotopes, observed for both Cu and Cr, also reflect a similar behaviour.

Table 3 Regression models for males and females
Full size table

In conclusion, differences in the accumulation and distribution of heavy metals (Al, Cr (Cr52, Cr53), Cu (Cu63, Cu65), Mn, Ni) and a metalloid (As) were characterized separately in four tissues (exoskeleton, gills, hepatopancreas and abdominal muscle tissue) of male and female crayfish, and the data were used to develop prediction models for the bioaccumulation of these elements in crayfish. The four tissue types tested yielded significantly different accumulation trends for every element tested, except Al, between male and female cohorts. Further, accumulation rates of Ni and As were noted to be different in gill tissues of male and female cohorts. The accumulations of these heavy metals were significantly less in female gill tissue compared to the gill tissue in males. Likewise, gender was observed to significantly affect the accumulation profile of Cu isotopes. For female cohorts, accumulation rates between 63Cu and 65Cu were not significantly different in gills, hepatopancreas and muscle tissue while for male cohorts, significant differences were observed in all tissues tested, except the gills. Furthermore, this study determined the effect of each element on the accumulation of other elements, with accumulation models presented for those relationships with R2 > 0.85. This information may be of value to future laboratory and field investigations.