- Email: [email protected]
Is there a linkage between bioaccumulation and the effects of alkylphenols on male breams (Abramis brama)?
ARTICLE IN PRESS
Environmental Research 98 (2005) 55–63 www.elsevier.com/locate/envres
Is there a linkage between bioaccumulation and the effects of alkylphenols on male breams (Abramis brama)? Roland Kleina,, Martina Bartela, Xiaohua Hea, Josef Mu¨llerb, Markus Quacka a
Universita¨t Trier, FB VI—Biogeographie, Wissenschaftspark Trier-Petrisberg, D-54286 Trier, Germany b Fraunhofer IME, Postfach 12 60, D-57377 Schmallenberg, Germany Received 2 February 2004; received in revised form 21 September 2004; accepted 22 September 2004 Available online 15 December 2004
Abstract There was some evidence from a previous study that estrogenic disruptors, like alkylphenols, could effect fish in the small River Saar of Southwestern Germany. Concentrations of 4NP and 4NP1EO found in breams (Abramis brama) in the Saar River were much higher than those found in other sampling sites of the German Environmental Specimen Bank, including those from sampling sites in the Rivers Elbe, Rhine, Mulde, and Saale and in Lake Belau. We studied the relationship between accumulation and effect using vitellogenin (vtg) and a hepatosomatic index (HSI) of estrogenic effects and by measuring concentrations of AP and APE accumulated in breams caught at six sampling sites in the River Saar and one in the River Mosel. To link these results with those of the previous study we standardized our sampling efforts to obtain comparable data. Elevated vtg levels were found in the breams at all sampling sites near to or downstream of sewage plant discharges, whereas low vtg levels corresponded to sampling sites not influenced by municipal waste water. While HSI values did not correspond to the location of sampling sites, there was a weak but statistically significant correlation to vtg concentrations. Concentrations of four AP and APE were much more lower, as in the previous study, and were neither linked with sewage treatment plant discharges nor correlated with vtg levels. In conclusion, a linkage between accumulation and the effects of AP and APE could not be established, but the relationship between elevated vtg concentrations and municipal waste water, which contains other important endocrine disruptors, was clear. r 2004 Elsevier Inc. All rights reserved. Keywords: Accumulation; Estrogenic effect; Standardized sampling; AP; APE; Bream (Abramis brama)
1. Introduction Estrogenic and anti-estrogenic effects of chemical substances in the aquatic environment are described by many studies that have utilized different species as indicators of environmental pollution (e.g., Folmar et al., 1996, 2001a, b; Arukwe et al., 2001a, b; Servos, 1999; Tyler et al., 1998; Harries et al., 1996, 1997; Oehlmann et al., 1995; Sumpter and Jobling, 1995). In Germany there are only a few environmental studies dealing with endocrine disruptors (Karbe et al., 2000; Lehmann et al., 2000; Hecker, 2001). This is true for small rivers passing Corresponding author. Fax: +49 651 201 3851.
E-mail address: [email protected] (R. Klein). 0013-9351/$ - see front matter r 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.envres.2004.09.006
through densely populated areas, although Jobling et al. (1998) demonstrated the risk of estrogenic disruptors for fish species living under such conditions. Chemical substances acting as endocrine disruptors are difficult to identify in the field due to the impacts of compounds and of many natural factors that modify the responses of biological systems. In this context, a study of Bo¨hmer and Wenzel (2001) is of greatest interest. They analyzed alkylphenols in the muscle of breams throughout Germany within the framework of the German Environmental Specimen Bank (ESB). They found very high concentrations in breams from the small River Saar, which runs through a highly industrialized area of southwest Germany. The concentrations ranged up to 70 mg/kg fresh weight (FW) of
ARTICLE IN PRESS 56
R. Klein et al. / Environmental Research 98 (2005) 55–63
4NP and more than 200 mg/kg FW of 4NP1EO, which were two- and fivefold higher, respectively, than the next highest values from the River Saale and the River Rhine (near Weil). Low concentrations of 4NP and 4NP1EO (10 and 20 mg/kg FW, respectively) were found in breams of the River Elbe (from the Czech–German border near Hamburg), the River Mulde, the River Rhine at Koblenz and Bimmen, and lake Belau near Kiel. Although Servos (1999) demonstrated in his review that the bioaccumulation and estrogenic response of alkylphenols and alkylphenol polyethoxylates are well studied in the laboratory, our knowledge of these phenomena in the field is insufficient to assess adequately the risk of incorporated concentrations for organisms themselves and for ecosystems. Besides the lack of comparable data, there is a lack of field studies in which both bioaccumulation and estrogenic responses are investigated. The objective of the present study is to investigate the linkage between bioaccumulation and the estrogenic effects of alkylphenols on male breams (Abramis brama) in the above-mentioned small River Saar to get a better understanding of the risk of alkylphenols as endocrine disruptors. Therefore, vitellogenin (vtg) as a marker of estrogenic response and concentrations of alkylphenols in male breams from the River Saar were analyzed in parallel. For this, breams were sampled from different sites along the River Saar at various distances from sewage treatment plant discharges. Thus, it should be possible to locate the sources of alkylphenols or other endocrine disruptors. Great importance was attached to standardized sampling in order to obtain results comparable to those from the study by Bo¨hmer and Wenzel (2001).
2. Material, methods, and sampling sites 2.1. Fish and treatment Male breams were sampled in principle in accordance with the sampling operating procedure (SOP) for breams of the German ESB (Klein and Paulus, 1996). The following adjustments to this SOP were made to reach the objectives of our study: 15 male breams (8–12 years old) were sampled instead of 20 breams of each sex because only male fish are suitable indicators of endocrine disruptors when vtg is the chosen biomarker of estrogenic response. The sampling was conducted in May and June, during the spawning season, whereas the SOP requires that sampling be conducted in August and September. This is because male breams can be identified easily by external spawning spots in the spring. Fish were caught with a fishing rod fish and then transported in big fish boxes to the laboratory where
they were killed individually with electric current. Immediately afterward blood was withdrawn from the heart using heparinized syringes (15 international unit/ ml) with 0.1 mg aprotinin per 100-ml vol. Blood samples were centrifuged (12 min at 3000 U/min and 2 1C). After centrifugation the plasma samples were stored in the gaseous phase over liquid nitrogen. After the blood was sampled the fish were weighed and measured and muscle tissue was dissected under a clean bench according to the requirements of the SOP. The muscle tissue of each individual was stored in Schott–Duran glass containers over liquid nitrogen. 2.2. Indirect ELISA analyses vtg analyses of plasma samples were performed using indirect enzyme-linked immunosorbent assay (ELISA) and ELISA kit Vtg-103 (Biosense Laboratories AS, Bergen, Norway) (Tyler et al., 1996; Arukwe et al., 2001a; Bartel et al., 2002). Plasma samples were diluted 1:1000, 1:10,000, and 1:1,000,000 using blocking/dilution (1% BSA in PBS). Reliable data ranged from 1 to 1000 mg/ml. Lower and higher values were excluded from further analyses. 2.3. Analyses of alkylphenols Muscle samples were homogenized at FraunfoferIME. Analyses of alkylphenols (AP) and alklyphenolmonoethoxylates (AP1EO) using GC/MS/MS (gas chromatography tandem mass spectrometry) was carried out according to Bo¨hmer and Wenzel (2001). In the muscle tissue of 73 breams the following AP and AP1EO were analyzed: 4-tert-octylphenol (4tOP), 4-nonylphenol (4NP), 4-tert-octylphenolmonoethoxylat (4tOP1EO), and 4-nonylphenolmonoethoxylat (4NP1EO). GC/MS/MS was calibrated using the certified reference material of Dr. Ehrenstorfer GmbH, Augsburg, Germany. 2.4. Statistical analyses All statistics were performed using SPSS software (Version 11.0). Most graphs were created as box- and whisker-plots. Tests of mean value differences were carried out using ANOVA. Correlations were calculated using a nonparametric procedure (Spearman). 2.5. Sampling sites The River Saar is a relatively small stream with a high density of sewage treatment plants (Fig. 1, blue dots) that are briefly characterized in Table 1. Because sewage treatment plants can also discharge AP, AP1EO, and other endocrine disruptors, sewage treatment plants
ARTICLE IN PRESS R. Klein et al. / Environmental Research 98 (2005) 55–63
57
Fig. 1. Map of the sampling sites in the Rivers Saar and Mosel. Blue dots indicate locations of sewage treatment plants.
Table 1 Brief description of sewage treatment plants in the rivers Saar and Mosel, which played an important role in the selection of sampling sites Sewage treatment plant
Year of initiation
Extendible by a capacity of (number of habitants)
Saargemu¨nd Brebach
n/a 1961, upgraded March 2001 1989 1961 1995 1989 1986 n/a n/a
n/a 40,000
Burbach Gersweiler Vo¨lklingen Saarlouis Dillingen Konz Trier
200,000 7000 80,000 93,000 50,000 27,000 311,000
n/a, not available.
were important criteria in sample site selection. Furthermore, the sampling sites of Gu¨dingen and Rehlingen were used as ‘‘fixed points’’ because they are included in the German Environmental Specimen Bank and (especially Gu¨dingen) are the sites at which
high concentrations of AP and AP1EO were found in a previous study connected with the German Environmental Specimen Bank (Bo¨hmer and Wenzel, 2001). Two sampling sites (Kleinblittersdorf and Hanweiler) were located upstream of Gu¨dingen, and Hanweiler was to serve as a reference site in the River Saar. Two further sampling sites (Burbach and Fraulautern) were selected between Gu¨dingen and Rehlingen, both of which were influenced by the discharge of sewage treatment plants. Trier was selected to serve as a reference site for a bigger river characterized by a lower density of sewage treatment plants (Fig. 1). The River Saar is enlarged to serve as a waterway and therefore is characterized by a high number of barrages between Gu¨dingen and the river mouth at Konz near Trier, where it flows into the River Mosel (Fig. 1). This means that the flow rate is low with a high rate of sedimentation, especially on the upstream side of the barrages. Water quality ranges from ‘‘critically loaded’’ (water quality class II–III) at most sites to ‘‘heavily loaded’’ (water quality class III) at the site of Fraulautern (http://www.umwelt.Saarland.de/medien/ inhalt/guete00.jpg). Even though the River Mosel is also used as waterway and has a very high density of barrages, low flow, and a high sedimentation rate, the
ARTICLE IN PRESS R. Klein et al. / Environmental Research 98 (2005) 55–63
58
water quality is assessed as ‘‘low load’’ (water quality class II) in most parts and at the study site of Trier (http://www.wasser.rlp.de/gis/wasser_gg_gis.htm).
3. Results 3.1. Number of fish and biometric characterization The target number of 15 male breams could not be reached at all sampling sites (Fig. 2). At most sampling sites far more females were caught than males. The sex ratio varied from 1:2 to 1:9 (males:females). In Burbach and Fraulautern two and one male breams (respectively) had ovotestes (two to five eggs appending male gonads, seen macroscopically). To verify ELISA some females were studied as well (Fig. 2). Most breams were 8–12 years old and showed no differences in somatic growth (Bartel et al., 2002). The corpulence factors were also comparable, as no significant differences were found (Po0:05; Fig. 3). Values of a hepatosomatic index (HSI), used as an indicator of chemical effects, such as increasing liver size, differ clearly between some sampling sites (Fig. 4). The lowest values were measured in Rehlingen and Trier, which differed significantly (Po0:05) from all other sampling sites except for Kleinblittersdorf. The highest values were found in Burbach, but significant differences were found at the lowest values only for Trier and Rehlingen (Table 2). 3.2. Analyses of alkylphenols All values for octylphenol-monoethoxylate analyses were very low. Most analyses fell below the limit of quantification (87%) or ranged from 0.2 (the limit of quantification) to 0.6 mg/kg FW. Therefore, these values are not presented and discussed in this paper.
Fig. 3. Corpulence factors of male breams at different sites in the Rivers Saar and Mosel.
Fig. 4. Hepatosomatic indices of male breams at different sites in the Rivers Saar and Mosel. Table 2 Homogeneous subgroups (ANOVA) of the HIS of male breams Sampling site
N
Subgroups of a ¼ 0:05 1
Trier Rehlingen Kleinblittersdorf Fraulautem
10 15 16 14
Hanweiler Gu¨dingen Burbach
8 16 15
Significance
Fig. 2. Numbers of breams studied at each sampling site.
1.35 1.39
0.71
2
1.39 1.62
0.07
3
4
1.62 1.67 1.73 1.81
1.73 1.81 1.96
0.17
0.07
Mean values (median) of 4-tert-octylphenol ranged from 0.3 (Trier) to 1.1 mg/kg (Gu¨dingen) FW and did not differ significantly (Fig. 5). The content in the muscle tissue of 4 individuals from Burbach was slightly elevated (48 mg/kg FW), but due to the low values of a
ARTICLE IN PRESS R. Klein et al. / Environmental Research 98 (2005) 55–63
59
Table 3 Homogeneous subgroups (ANOVA) of 4-nonylphenol in the muscle tissue of breams from the Rivers Saar and Mosel Sampling site
N
Subgroups for a ¼ 0:05 1
Trier Rehlingen Gu¨dingen Hanweiler
9 9 10 8
5.02 5.08 7.25 7.77
Kleinblittersdorf Burbach Fraulautem
13 12 10
10.00
Significance
2
3
7.25 7.77 10.00 11.67 20.02
0.06
0.09
1.00
Fig. 5. 4-tert-Octylphenol in muscle tissue of breams from the Rivers Saar and Mosel.
Fig. 6. 4-Nonylphenol in muscle tissue of breams from the Rivers Saar and Mosel.
further 8 fish the mean content was comparable to that of all other sampling sites. In contrast to 4-tert-octylphenol, the values of 4NP differed clearly between some sampling sites (Fig. 6). The content of breams from Fraulautern was significantly different (Po0:05) from the content of breams from all other sampling sites (Table 3). Further significant differences existed between Burbach on the one hand and Trier and Rehlingen on the other, whereas no significance occurred between other sites. The results for 4NP1EO show a less pronounced spatial differentiation (Fig. 7). Although the highest mean value (Fraulautern) is significantly different to all other sites except of Burbach (Po0:05), there is only one further significance between Trier and Kleinblittersdorf (Po0:05). 3.3. Vitellogenin Fig. 8 shows the differences in the concentrations of vtg in the blood plasma of male breams between the
Fig. 7. 4-Nonylphenolmonoethoxylat in muscle tissue of breams from the Rivers Saar and Mosel.
Fig. 8. vtg in blood plasma of male breams from the Rivers Saar and Mosel.
sampling sites. The lowest values were found in Trier and were significantly different (Po0:05) from those of all other sites except for Hanweiler. Here vtg concentrations were also low and differed significantly (Po0:05) from all sites, with the exception of Rehlingen. There
ARTICLE IN PRESS 60
R. Klein et al. / Environmental Research 98 (2005) 55–63
Table 4 Correlations (Spearman–Rho) between vtg concentrations, the HIS, and the corpulence factor vtg vtg
Corpulence factor
HSI
1.00
Corpulence factor
HIS
0.23* 0.05 73.00
0.35* 0.00 71.00
1.00
0.10 0.42 71.00
Correlation coefficient Significance (two-sided) N
73.00
Correlation coefficient Significance (two-sided) N
0.23* 0.05 73.00
73.00
Correlation coefficient Significance (two-sided) N
0.35* 0.00 71
0.10 0.42 71
1.00 71
Po0:01 (two-sided).
were no significant differences between the sites from Kleinblittersdorf to Rehlingen.
Table 5 Correlations (Spearman–Rho) between alkylphenols and vtg concentrations in male breams
3.4. Correlations To screen the impact of elevated vtg concentrations on male breams, vtg values were correlated with the HSI and the corpulence factor. Values of vtg under and over the limit of quantification were excluded. A nonparametric test was used because the vtg concentrations were not Gaussian distributed. Table 4 shows weak but highly significant correlations between vtg concentrations and the HSI, whereas vtg concentrations and the corpulence factor do not correlate. To verify whether alkylphenols induce vtg in male breams the concentrations of both variables were also correlated, excluding 4tOP because the values were low and did not differ greatly. Table 5 clearly shows no correlation between alkylphenols and vtg concentrations, but relatively strong and highly significant correlations between 4NP and 4NP1EO.
4. Discussion Corpulence factors serve as indicators of comparability between the breams from different sampling sites because they show the relationship between the length and weight of more or less coeval fish (8–12 years old). This means that the breams of all sampling sites were more or less well fed. Therefore, one can assume that the fish samples were comparable, meeting an important prerequisite for comparable data. The HSI is considered an indicator of estrogenic effects (Harries et al., 1997; Karbe et al., 2000; Karels et al., 2001). Folmar et al. (2001b) discussed the accumulation of vtg in the livers of male fish, which are probably not able to excrete the produced vtg via the kidneys in an adequate way. Consequently, the accumulated vtg
vtg vtg
4NP
4NP1EO
Spearman–Rho Po0:01 (two-sided) N
1.00 47
Spearman–Rho Po0:01 (two-sided) N
0.14 0.36 47
Spearman–Rho Po0:01 (two-sided) N
0.20 0.18 47
4NP
4NP1EO
0.14 0.36 47
0.20 0.18 47
1.00
0.81
47 0.81 47
47 1.00 47
results in a hypertrophy of liver cells and of the liver itself. One should note that chemical substances other than endocrine disruptors and lipid accumulation could also cause liver hypertrophy (Karels et al., 2001). With the exception of Rehlingen and Trier, the HSI of male beams from all sampling sites was higher than 1.5. Compared to the long time series of the German Environmental Specimen Bank, values 41.5 may be considered to be elevated; breams from the sampling site of Gu¨dingen had, in most years, the highest HSIs of between 1.8 and 2.4 (Bartel et al., 2002). We stress that this study was conducted in the spring, whereas investigations of the German ESB are conducted in summer, outside the spawning season. Therefore, absolute amounts should be compared with care. Overall, one can conclude that at least the HSI values from Gu¨dingen and Burbach indicate liver hypertrophy. Nutritional status did not seem to impact the differences in liver size between the sampling sites because there was no correlation between the corpulence factor and the HSI. In contrast, a low but highly significant correlation of the HSI and vtg could suggest a certain influence of vtg on liver hypertrophy. Nevertheless, vtg does not
ARTICLE IN PRESS R. Klein et al. / Environmental Research 98 (2005) 55–63
cause exclusively high values of the HSI, which can be seen by the fact that in Hanweiler and Rehlingen vtg and the HSI were running in the opposite direction. In conclusion, the HSI cannot be considered a reliable indicator of endocrine effects. This corresponds to earlier findings in field studies. Whereas in laboratory tests the induction of biomarkers like vtg or Zona radiata proteins by endocrine disruptors ran parallel with liver hypertrophy, this correlation was less obvious in field studies (Allen et al., 1999; Folmar et al., 2001b; Karels et al., 2001). The influence of chemical substances not acting as endocrine disruptors and the impact of natural factors (climate, hydrology, nutrients, etc.) contribute to the complex factors that determine the HSI. vtg concentrations measured in the blood plasma of male breams indicate clearly that fish in the River Saar are exposed to estrogen disruptors, with the exception of those from the site of Hanweiler. They are very often significantly higher than those of fish from Hanweiler and Trier. Until now there have been few data available on vtg concentrations in male breams or in Cyprinids to allow a comprehensive assessment of measured values. Compared with two studies done by Lehmann et al. (2000) and Karbe et al. (2000) with breams from the largest German rivers, the Rhein and the Elbe, the vtg concentrations in this study were extremely high. Even the vtg concentrations in breams from Hanweiler and Trier should be considered evidence of the exposure of fish to estrogenic disruptors. In contrast, following results from Tyler et al. (1996), who found vtg concentrations of 900–1000 mg/L in female bream, an interpretation of our results should lead to different conclusions. This shows that it is a moot question whether the measured concentrations of different studies may be compared with each other because of a lack of standardization and harmonization of field and laboratory methods. Therefore, the comparison of vtg analyses of other species of cyprinids should be assessed very carefully as well (Purdom et al., 1994; Harries et al., 1995; Folmar et al., 1996; Karels et al., 2001; RodgersGray et al., 2001). Despite this one can doubtlessly assert that the elevated vtg concentrations in male breams of the River Saar are associated with the discharge of sewage treatment plants, which are assumed to be the most important source of estrogenic disruptors. This corresponds with the findings of other studies in England, Finland, and the USA on rainbow trout (Oncorhynchus mykiss), common carp (Cyprinus carpio), roach (Rutilus rutilus), walleye species (Stizostedion vitreum), and flounder (Platichthys flesus) (Harries et al., 1996, 1997; Folmar et al., 1996; Allen et al., 1999; Folmar et al., 2001a; Karels et al., 2001). In contrast, Nichols et al. (1999) could not affirm this correlation in a study of bullhead minnows (Pimephales promelas).
61
In these studies the range of discharges influencing vtg concentrations in fish were very different. In our study the longest reach was 6600 m (Kleinblittersdorf). Only Harries et al. (1996, 1997) found significantly elevated vtg concentrations compared with a reference site at a similar distance from a discharge (5000 m); the reaches were shorter in all other studies. There are many factors influencing the range of discharges from sewage treatment plants. In our study the main factors for a great range are that the River Saar is a very small, backed-up river with a very high density of sewage treatment plants. In addition, breams themselves have an impact on the range of discharges. First, they are mobile to a certain degree because their home range covers several kilometers. Second, breams usually utilize sediment as their nutrition habitat, and sediment is known to be a sink of many chemical substances (Jobling and Sumpter, 1993; Servos, 1999). Thus, breams are exposed to endogen disruptors to a greater extent than are many other fish species. Normally, it is difficult to detect chemical substances acting as estrogenic disruptors in field studies because in rivers a complex and highly dynamic mixture of chemicals affects organisms. Therefore, the results of Bo¨hmer and Wenzel (2001) seemed applicable to our study. In a German-wide study they found the highest concentrations of 4NP and 4NP1EO in the muscle of breams from the River Saar at the sampling sites of Gu¨dingen and Rehlingen, which are also included in our study. Up to now, we have known less on the accumulation of these substances in fish. Consequently, it is difficult to assess comprehensively the measured concentrations (Bo¨hmer and Wenzel, 2001). Nevertheless, the study of Bo¨hmer and Wenzel (2001), which analyzed time series of breams from Gu¨dingen and Rehlingen based on stored samples of the German Environmental Specimen Bank, allows us to classify our analysis with a comparison with the previous findings. This shows clearly that the concentrations of NP (nonylphenol) from all sampling sites were conspicuously lower than those at Gu¨dingen and Rehlingen between 1992 and 1998. In Gu¨dingen the concentrations were reduced dramatically. Whereas in 1992 and 1994 the concentrations ranged between 60 and 100 mg/kg FW, and even in 1998 concentrations of 30 mg/kg FW were found, only 7 mg/kg FW was measured 2001. A clear reduction of accumulated NP concentrations since 1992 can also be observed at Rehlingen, a reduction even more distinct between 1998 and 2001 than that seen at Gu¨dingen (30 mg/kg FW in 1998 and 5 mg/kg FW in 2001). The values of NP1EO decreased more drastically in Gu¨dingen than those of NP. In 2001 the concentrations were reduced by a factor of eight, compared to 1992 and 1994, when more than 240 mg/kg FW was measured. In Rehlingen a decrease could also be observed, but it was
ARTICLE IN PRESS 62
R. Klein et al. / Environmental Research 98 (2005) 55–63
neither as clear as the decrease in Gu¨dingen nor as dramatic as that for NP: the concentration went from 70 mg/kg FW in 1992, to about 30 mg/kg FW in 1998, and to 25 mg/kg FW in 2001. The described decrease is contradictory to those of most other chemical substances determined on a routine basis since 1992 in the muscle of breams by the German ESB, which either have remained on the same level or increased since the beginning of the investigation (Wagner et al., 2004). A decrease has only been observed for Se, y-HCH, and HCB in Gu¨dingen and y-HCH in Rehlingen. Compared to sampling sites in the Rivers Elbe, Rhein, Donau, Mulde, and Saale, the River Saar is characterized by the highest levels of PCB (especially PCB 52 and 101) and—despite the decrease—of y-HCH. Additionally, the site of Rehlingen shows the highest levels of Tl and As, since an increase in 1995. The more or less drastic decrease of AP and APE could mean that these chemical substances did not cause the induction of vtg in the liver of male breams. This assumption is confirmed by two factors: there was more or less no clear differentiation in concentrations of 4NP and 4NP1EO between the sampling sites and we found no correlation between these substances and vtg concentrations. The physiological relationship between the concentrations of NP and NP1EO in blood plasma and the concentration of vtg in the muscle is indirect because the concentration in muscle tissue can be used as an indicator of concentrations in liver. Based on this, it is probable that the above-mentioned assumption is true. In this context, it is important that octylphenol, as the most potent estrogenic disruptor of the analyzed AP and APE (Servos, 1999), showed no spatial differentiation and very low values. This is further evidence that AP and APE did not impact highly the induction of vtg in the liver of male breams. Consequently, it remains ambiguous which chemical substance in the effluent caused the elevated vtg concentrations in male breams in the River Saar between Kleinblittersdorf and Rehlingen. Furthermore, it is difficult to prove to what extent chemical substances determined by the German ESB (chlorinated hydrocarbons and some elements) contribute to the production of vtg because these data are available only for Gu¨dingen and Rehlingen, both of which show elevated vtg concentrations. Besides AP and APE, natural estrogens (17X-estradiol und esterone) and synthetic estrogens (17a-ethynyl-estradiol) are the main components of estrogen disruptors in municipal waste water (Rodgers-Gray et al., 2001). Therefore, we cannot determine to what extent they each affect the vtg concentrations found. Overall, one can conclude that vtg is a reliable indicator of exposure to estrogenic disruptors and of a first estrogen effect in organisms but not the overall relevance of estrogenic disruptors for biota.
Acknowledgments This study was funded by the German Environmental Agency (UBA) in Berlin. We thank Ms. Petra Willems and Mr. Bernd Fontaine from the University of Trier for ELISA analyses and fishing, respectively.
References Allen, Y., Matthiessen, P., Scott, A.P., Haworth, S., Feist, S., Thain, J.E., 1999. The extent of oestrogenic contamination in the UK estuarine and marine environments—further surveys of flounder. Sci. Total Environ. 233, 5–20. Arukwe, A., Kullmann, S.W., Hinton, D.E., 2001a. Differential biomarker gene and protein expressions in nonylphenol and estradiol-17X treated juvenile rainbow trout (Oncorhynchus mykiss). Comp. Biochem. Physiol. 129, 1–10. Arukwe, A., Yadetie, F., Male, R., Goksoyr, A., 2001b. In vivo modulation of nonylphenol-induced zonagenesis and vitellogenesis by the antiestrogen, 3,30 4,40 -tetrachlorobiphenyl (PCB-77) in juvenile fish. Environ. Toxicol. Pharmacol. 10, 5–15. Bartel, M., Klein, R., Nentwich, K., Paulus, M., Wagner, G., 2002. Beitra¨ge zur Probenahme und Probenbeschreibung fu¨r die Umweltprobenbank des Bundes—Jahresbericht 2001. Univ. Trier, Trier, Germany, unpublished, 145pp. Bo¨hmer, W., Wenzel, A., 2001. Teilbericht II: Alkylphenole und Alkylphenolethoxylate. In: Umweltbundesamt (Hrsg.), Organische Zinnverbindungen, Alkylphenole und Bisphenol A in marinen und limnischen Biota der Umweltprobenbank. UBA Texte 06/01, pp. 63–122. Folmar, L.C., Denslow, N.D., Rao, V., Chow, M., Crain, D.A., Enblom, J., Marcino, J., Guilette Jr., L.J., 1996. Vitellogenin induction and reduced serum testosterone concentrations in feral male carp (Cyprinus carpio) captured near a major metropolitan sewage treatment plant. Environ. Health Perspect. 104/10, 1096–1101. Folmar, L.C., Denslow, N.D., Kroll, K., Orlando, E.F., Enblom, J., Marcino, J., Metcalfe, C., Guillette Jr., L.J., 2001a. Altered serum sex steroids and vitellogenin induction in walleye (Stizostedion vitreum) collected near a metropolitan sewage treatment plant. Arch. Environ. Contam. Toxicol. 40, 392–398. Folmar, L.C., Gardner, G.R., Schreibman, M.P., Magliulo-Cepriano, L., Mills, L.J., Zaroogian, G., Gutjahr-Gobell, R., Haebler, R., Horowitz, D.B., Denslow, N.D., 2001b. Vitellogenin-induced pathology in male Summer Flounder (Paralichthys dentatus). Aquat. Toxicol. 51, 431–441. Harries, J., Jobling, S., Matthiessen, P., Sheahan, D.A., Sumpter, J.P., 1995. Severity and extent of contamination of some English rivers and drinking water reservoirs by estrogenic substances. Report to UK Department of the Environment, Ref. PECD 7/7/384. Harries, J.E., Sheahan, D.A., Jobling, S., Matthiessen, P., Neall, P., Routledge, E.J., Rycroft, R., Sumpter, J.P., Tylor, T., 1996. A survey of estrogenic activity in United Kingdom inland waters. Environ. Toxicol. Chem. 15/11, 1993–2002. Harries, J.E., Sheahan, D.A., Jobling, S., Matthiessen, P., Neall, P., Sumpter, J.P., Tylor, T., Zaman, N., 1997. Estrogenic activity in five United Kingdom rivers detected by measurement of vitellogenesis in caged male trout. Environ. Toxicol. Chem. 16/3, 534–542. Hecker, M., 2001. Natu¨rliche Variabilita¨t endokriner Funktionen und ihre Modulation durch anthropogene Einflu¨sse—Untersuchungen am Beispiel des Brassen (Abramis brama [L.]) in der Elbe und in einem Referenzgewa¨sser. Berichte aus dem Zentrum fu¨r Meeresund Klimaforschung Reihe E, Nr 16, 149.
ARTICLE IN PRESS R. Klein et al. / Environmental Research 98 (2005) 55–63 Jobling, S., Nolan, M., Tyler, C.R., Brighty, G., Sumpter, J.P., 1998. Widespread sexual disruption in wild fish. Environ. Sci. Technol. 32, 2498. Jobling, S., Sumpter, J.P., 1993. Detergent components in sewage effluent are weakly oestrogenic to fish: an in vitro study using rainbow trout (Oncorhynchus mykiss) hepatocytes. Aquat. Toxicol. 27, 361–372. Karbe, L., Westendorf, J., Beck, M.-P., Gutendorf, B., Hecker, M., 2000. O¨kotoxikologische Relevanz endokrin wirksamer Stoffe in der aquatischen Umwelt—Untersuchungen am Modell Elbe. AbschluXbericht FE-Vorhaben 29765001/02, Univ. Hamburg, p. 113. Karels, A., Markkula, E., Oikari, A., 2001. Reproductive, biochemical, physiological, and population responses in perch (Perca fluviatilis L.) and roach (Rutilus rutilus L.) downstream of two elemental chlorine-free pulp and paper mills. Environ. Toxicol. Chem. 20/7, 1517–1527. Klein, R., Paulus, M., 1996. Richtlinie zur Probenahme und Probenbearbeitung, Transport und Lagerung—Limnische Probenarten—Brassen (Abramis brama). In: Umweltbundesamt (Hrsg.), Umweltprobenbank des Bundes—Verfahrensrichtlinien: Teil 6. Erich Schmidt Verlag, Berlin, pp. 1–26. Lehmann, J., Paris, F., Stu¨renberg, F.-J., Blu¨m, V., 2000. O¨kotoxikologische Untersuchungen an freilebenden Brassen, LO¨BFJahresbericht 1999, pp. 127–131. Nichols, K.M., Miles-Richardson, S.R., Snyder, E.M., Criesy, J.P., 1999. Effects of exposure to municipal wastewater in situ on the reproductive physiology of the fathead minnow (Pimephales promelas). Environ. Toxicol. Chem. 18, 2001–2012.
63
Oehlmann, J., Schulte-Oehlmann, U., Stroben, E., Bauer, B., Bettin, C., Fioroni, P., 1995. Androgene Effekte zinnorganischer Verbindungen bei Mollusken. In: Umweltbundesamt (Ed.), Umweltchemikalien mit endokriner Wirkung, UBA-Texte 65/95, Berlin, pp. 77–85. Purdom, C.E., Hardiman, P.A., Bye, V.J., Eno, N.C., Tyler, C.R., Sumpter, J.P., 1994. Estrogenic effects of effluents from sewage treatment works. Chem. Ecol. 8, 275–285. Rodgers-Gray, T.P., Jobling, S., Kelly, C., Morris, S., Brighty, G., Waldock, M.J., Sumpter, J.P., Tyler, C.R., 2001. Exposure of juvenile roach (Rutilus rutilus) to treated sewage effluent induces dose-dependent and persistent disruption in gonadal duct development. Environ. Sci. Technol. 35/3, 462–470. Servos, M.R., 1999. Review of the aquatic toxicity, estrogenic responses and bioaccumulation of alklylphenols and alkylphenol polyethoxylates. Water Qual. Res. J. Can. 34/1, 123–177. Sumpter, J.P., Jobling, S., 1995. Vitellogenesis as a biomarker for estrogenic contamination of the aquatic environment. Environ. Health Perspect. 103, 173–178. Tyler, C.R., van der Eerden, B., Joblin, S., Panter, G., Sumpter, J.P., 1996. Measurement of vitellogenin, a biomarker for exposure to oestrogenic chemicals, in a wide variety of cyprinid fish. Comp. Biochem. Physiol. B 166, 418–426. Tyler, C.R., Jobling, S., Sumpter, J.P., 1998. Endocrine disruption in wildlife: a critical review of the evidence. Crit. Rev. Toxicol. 28/4, 319–361. Wagner, G., Quack, M., Paulus, M., Nentwich, K., Klein, R., Bartel., M., 2004. Beitra¨ge zur Probenahme und Probenbeschreibung fu¨r die Umweltprobenbank des Bundes–Jahresbericht 2003. Univ. Trier, Trier, Germany, unpublished, 152pp.
Environmental Research 98 (2005) 55–63 www.elsevier.com/locate/envres
Is there a linkage between bioaccumulation and the effects of alkylphenols on male breams (Abramis brama)? Roland Kleina,, Martina Bartela, Xiaohua Hea, Josef Mu¨llerb, Markus Quacka a
Universita¨t Trier, FB VI—Biogeographie, Wissenschaftspark Trier-Petrisberg, D-54286 Trier, Germany b Fraunhofer IME, Postfach 12 60, D-57377 Schmallenberg, Germany Received 2 February 2004; received in revised form 21 September 2004; accepted 22 September 2004 Available online 15 December 2004
Abstract There was some evidence from a previous study that estrogenic disruptors, like alkylphenols, could effect fish in the small River Saar of Southwestern Germany. Concentrations of 4NP and 4NP1EO found in breams (Abramis brama) in the Saar River were much higher than those found in other sampling sites of the German Environmental Specimen Bank, including those from sampling sites in the Rivers Elbe, Rhine, Mulde, and Saale and in Lake Belau. We studied the relationship between accumulation and effect using vitellogenin (vtg) and a hepatosomatic index (HSI) of estrogenic effects and by measuring concentrations of AP and APE accumulated in breams caught at six sampling sites in the River Saar and one in the River Mosel. To link these results with those of the previous study we standardized our sampling efforts to obtain comparable data. Elevated vtg levels were found in the breams at all sampling sites near to or downstream of sewage plant discharges, whereas low vtg levels corresponded to sampling sites not influenced by municipal waste water. While HSI values did not correspond to the location of sampling sites, there was a weak but statistically significant correlation to vtg concentrations. Concentrations of four AP and APE were much more lower, as in the previous study, and were neither linked with sewage treatment plant discharges nor correlated with vtg levels. In conclusion, a linkage between accumulation and the effects of AP and APE could not be established, but the relationship between elevated vtg concentrations and municipal waste water, which contains other important endocrine disruptors, was clear. r 2004 Elsevier Inc. All rights reserved. Keywords: Accumulation; Estrogenic effect; Standardized sampling; AP; APE; Bream (Abramis brama)
1. Introduction Estrogenic and anti-estrogenic effects of chemical substances in the aquatic environment are described by many studies that have utilized different species as indicators of environmental pollution (e.g., Folmar et al., 1996, 2001a, b; Arukwe et al., 2001a, b; Servos, 1999; Tyler et al., 1998; Harries et al., 1996, 1997; Oehlmann et al., 1995; Sumpter and Jobling, 1995). In Germany there are only a few environmental studies dealing with endocrine disruptors (Karbe et al., 2000; Lehmann et al., 2000; Hecker, 2001). This is true for small rivers passing Corresponding author. Fax: +49 651 201 3851.
E-mail address: [email protected] (R. Klein). 0013-9351/$ - see front matter r 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.envres.2004.09.006
through densely populated areas, although Jobling et al. (1998) demonstrated the risk of estrogenic disruptors for fish species living under such conditions. Chemical substances acting as endocrine disruptors are difficult to identify in the field due to the impacts of compounds and of many natural factors that modify the responses of biological systems. In this context, a study of Bo¨hmer and Wenzel (2001) is of greatest interest. They analyzed alkylphenols in the muscle of breams throughout Germany within the framework of the German Environmental Specimen Bank (ESB). They found very high concentrations in breams from the small River Saar, which runs through a highly industrialized area of southwest Germany. The concentrations ranged up to 70 mg/kg fresh weight (FW) of
ARTICLE IN PRESS 56
R. Klein et al. / Environmental Research 98 (2005) 55–63
4NP and more than 200 mg/kg FW of 4NP1EO, which were two- and fivefold higher, respectively, than the next highest values from the River Saale and the River Rhine (near Weil). Low concentrations of 4NP and 4NP1EO (10 and 20 mg/kg FW, respectively) were found in breams of the River Elbe (from the Czech–German border near Hamburg), the River Mulde, the River Rhine at Koblenz and Bimmen, and lake Belau near Kiel. Although Servos (1999) demonstrated in his review that the bioaccumulation and estrogenic response of alkylphenols and alkylphenol polyethoxylates are well studied in the laboratory, our knowledge of these phenomena in the field is insufficient to assess adequately the risk of incorporated concentrations for organisms themselves and for ecosystems. Besides the lack of comparable data, there is a lack of field studies in which both bioaccumulation and estrogenic responses are investigated. The objective of the present study is to investigate the linkage between bioaccumulation and the estrogenic effects of alkylphenols on male breams (Abramis brama) in the above-mentioned small River Saar to get a better understanding of the risk of alkylphenols as endocrine disruptors. Therefore, vitellogenin (vtg) as a marker of estrogenic response and concentrations of alkylphenols in male breams from the River Saar were analyzed in parallel. For this, breams were sampled from different sites along the River Saar at various distances from sewage treatment plant discharges. Thus, it should be possible to locate the sources of alkylphenols or other endocrine disruptors. Great importance was attached to standardized sampling in order to obtain results comparable to those from the study by Bo¨hmer and Wenzel (2001).
2. Material, methods, and sampling sites 2.1. Fish and treatment Male breams were sampled in principle in accordance with the sampling operating procedure (SOP) for breams of the German ESB (Klein and Paulus, 1996). The following adjustments to this SOP were made to reach the objectives of our study: 15 male breams (8–12 years old) were sampled instead of 20 breams of each sex because only male fish are suitable indicators of endocrine disruptors when vtg is the chosen biomarker of estrogenic response. The sampling was conducted in May and June, during the spawning season, whereas the SOP requires that sampling be conducted in August and September. This is because male breams can be identified easily by external spawning spots in the spring. Fish were caught with a fishing rod fish and then transported in big fish boxes to the laboratory where
they were killed individually with electric current. Immediately afterward blood was withdrawn from the heart using heparinized syringes (15 international unit/ ml) with 0.1 mg aprotinin per 100-ml vol. Blood samples were centrifuged (12 min at 3000 U/min and 2 1C). After centrifugation the plasma samples were stored in the gaseous phase over liquid nitrogen. After the blood was sampled the fish were weighed and measured and muscle tissue was dissected under a clean bench according to the requirements of the SOP. The muscle tissue of each individual was stored in Schott–Duran glass containers over liquid nitrogen. 2.2. Indirect ELISA analyses vtg analyses of plasma samples were performed using indirect enzyme-linked immunosorbent assay (ELISA) and ELISA kit Vtg-103 (Biosense Laboratories AS, Bergen, Norway) (Tyler et al., 1996; Arukwe et al., 2001a; Bartel et al., 2002). Plasma samples were diluted 1:1000, 1:10,000, and 1:1,000,000 using blocking/dilution (1% BSA in PBS). Reliable data ranged from 1 to 1000 mg/ml. Lower and higher values were excluded from further analyses. 2.3. Analyses of alkylphenols Muscle samples were homogenized at FraunfoferIME. Analyses of alkylphenols (AP) and alklyphenolmonoethoxylates (AP1EO) using GC/MS/MS (gas chromatography tandem mass spectrometry) was carried out according to Bo¨hmer and Wenzel (2001). In the muscle tissue of 73 breams the following AP and AP1EO were analyzed: 4-tert-octylphenol (4tOP), 4-nonylphenol (4NP), 4-tert-octylphenolmonoethoxylat (4tOP1EO), and 4-nonylphenolmonoethoxylat (4NP1EO). GC/MS/MS was calibrated using the certified reference material of Dr. Ehrenstorfer GmbH, Augsburg, Germany. 2.4. Statistical analyses All statistics were performed using SPSS software (Version 11.0). Most graphs were created as box- and whisker-plots. Tests of mean value differences were carried out using ANOVA. Correlations were calculated using a nonparametric procedure (Spearman). 2.5. Sampling sites The River Saar is a relatively small stream with a high density of sewage treatment plants (Fig. 1, blue dots) that are briefly characterized in Table 1. Because sewage treatment plants can also discharge AP, AP1EO, and other endocrine disruptors, sewage treatment plants
ARTICLE IN PRESS R. Klein et al. / Environmental Research 98 (2005) 55–63
57
Fig. 1. Map of the sampling sites in the Rivers Saar and Mosel. Blue dots indicate locations of sewage treatment plants.
Table 1 Brief description of sewage treatment plants in the rivers Saar and Mosel, which played an important role in the selection of sampling sites Sewage treatment plant
Year of initiation
Extendible by a capacity of (number of habitants)
Saargemu¨nd Brebach
n/a 1961, upgraded March 2001 1989 1961 1995 1989 1986 n/a n/a
n/a 40,000
Burbach Gersweiler Vo¨lklingen Saarlouis Dillingen Konz Trier
200,000 7000 80,000 93,000 50,000 27,000 311,000
n/a, not available.
were important criteria in sample site selection. Furthermore, the sampling sites of Gu¨dingen and Rehlingen were used as ‘‘fixed points’’ because they are included in the German Environmental Specimen Bank and (especially Gu¨dingen) are the sites at which
high concentrations of AP and AP1EO were found in a previous study connected with the German Environmental Specimen Bank (Bo¨hmer and Wenzel, 2001). Two sampling sites (Kleinblittersdorf and Hanweiler) were located upstream of Gu¨dingen, and Hanweiler was to serve as a reference site in the River Saar. Two further sampling sites (Burbach and Fraulautern) were selected between Gu¨dingen and Rehlingen, both of which were influenced by the discharge of sewage treatment plants. Trier was selected to serve as a reference site for a bigger river characterized by a lower density of sewage treatment plants (Fig. 1). The River Saar is enlarged to serve as a waterway and therefore is characterized by a high number of barrages between Gu¨dingen and the river mouth at Konz near Trier, where it flows into the River Mosel (Fig. 1). This means that the flow rate is low with a high rate of sedimentation, especially on the upstream side of the barrages. Water quality ranges from ‘‘critically loaded’’ (water quality class II–III) at most sites to ‘‘heavily loaded’’ (water quality class III) at the site of Fraulautern (http://www.umwelt.Saarland.de/medien/ inhalt/guete00.jpg). Even though the River Mosel is also used as waterway and has a very high density of barrages, low flow, and a high sedimentation rate, the
ARTICLE IN PRESS R. Klein et al. / Environmental Research 98 (2005) 55–63
58
water quality is assessed as ‘‘low load’’ (water quality class II) in most parts and at the study site of Trier (http://www.wasser.rlp.de/gis/wasser_gg_gis.htm).
3. Results 3.1. Number of fish and biometric characterization The target number of 15 male breams could not be reached at all sampling sites (Fig. 2). At most sampling sites far more females were caught than males. The sex ratio varied from 1:2 to 1:9 (males:females). In Burbach and Fraulautern two and one male breams (respectively) had ovotestes (two to five eggs appending male gonads, seen macroscopically). To verify ELISA some females were studied as well (Fig. 2). Most breams were 8–12 years old and showed no differences in somatic growth (Bartel et al., 2002). The corpulence factors were also comparable, as no significant differences were found (Po0:05; Fig. 3). Values of a hepatosomatic index (HSI), used as an indicator of chemical effects, such as increasing liver size, differ clearly between some sampling sites (Fig. 4). The lowest values were measured in Rehlingen and Trier, which differed significantly (Po0:05) from all other sampling sites except for Kleinblittersdorf. The highest values were found in Burbach, but significant differences were found at the lowest values only for Trier and Rehlingen (Table 2). 3.2. Analyses of alkylphenols All values for octylphenol-monoethoxylate analyses were very low. Most analyses fell below the limit of quantification (87%) or ranged from 0.2 (the limit of quantification) to 0.6 mg/kg FW. Therefore, these values are not presented and discussed in this paper.
Fig. 3. Corpulence factors of male breams at different sites in the Rivers Saar and Mosel.
Fig. 4. Hepatosomatic indices of male breams at different sites in the Rivers Saar and Mosel. Table 2 Homogeneous subgroups (ANOVA) of the HIS of male breams Sampling site
N
Subgroups of a ¼ 0:05 1
Trier Rehlingen Kleinblittersdorf Fraulautem
10 15 16 14
Hanweiler Gu¨dingen Burbach
8 16 15
Significance
Fig. 2. Numbers of breams studied at each sampling site.
1.35 1.39
0.71
2
1.39 1.62
0.07
3
4
1.62 1.67 1.73 1.81
1.73 1.81 1.96
0.17
0.07
Mean values (median) of 4-tert-octylphenol ranged from 0.3 (Trier) to 1.1 mg/kg (Gu¨dingen) FW and did not differ significantly (Fig. 5). The content in the muscle tissue of 4 individuals from Burbach was slightly elevated (48 mg/kg FW), but due to the low values of a
ARTICLE IN PRESS R. Klein et al. / Environmental Research 98 (2005) 55–63
59
Table 3 Homogeneous subgroups (ANOVA) of 4-nonylphenol in the muscle tissue of breams from the Rivers Saar and Mosel Sampling site
N
Subgroups for a ¼ 0:05 1
Trier Rehlingen Gu¨dingen Hanweiler
9 9 10 8
5.02 5.08 7.25 7.77
Kleinblittersdorf Burbach Fraulautem
13 12 10
10.00
Significance
2
3
7.25 7.77 10.00 11.67 20.02
0.06
0.09
1.00
Fig. 5. 4-tert-Octylphenol in muscle tissue of breams from the Rivers Saar and Mosel.
Fig. 6. 4-Nonylphenol in muscle tissue of breams from the Rivers Saar and Mosel.
further 8 fish the mean content was comparable to that of all other sampling sites. In contrast to 4-tert-octylphenol, the values of 4NP differed clearly between some sampling sites (Fig. 6). The content of breams from Fraulautern was significantly different (Po0:05) from the content of breams from all other sampling sites (Table 3). Further significant differences existed between Burbach on the one hand and Trier and Rehlingen on the other, whereas no significance occurred between other sites. The results for 4NP1EO show a less pronounced spatial differentiation (Fig. 7). Although the highest mean value (Fraulautern) is significantly different to all other sites except of Burbach (Po0:05), there is only one further significance between Trier and Kleinblittersdorf (Po0:05). 3.3. Vitellogenin Fig. 8 shows the differences in the concentrations of vtg in the blood plasma of male breams between the
Fig. 7. 4-Nonylphenolmonoethoxylat in muscle tissue of breams from the Rivers Saar and Mosel.
Fig. 8. vtg in blood plasma of male breams from the Rivers Saar and Mosel.
sampling sites. The lowest values were found in Trier and were significantly different (Po0:05) from those of all other sites except for Hanweiler. Here vtg concentrations were also low and differed significantly (Po0:05) from all sites, with the exception of Rehlingen. There
ARTICLE IN PRESS 60
R. Klein et al. / Environmental Research 98 (2005) 55–63
Table 4 Correlations (Spearman–Rho) between vtg concentrations, the HIS, and the corpulence factor vtg vtg
Corpulence factor
HSI
1.00
Corpulence factor
HIS
0.23* 0.05 73.00
0.35* 0.00 71.00
1.00
0.10 0.42 71.00
Correlation coefficient Significance (two-sided) N
73.00
Correlation coefficient Significance (two-sided) N
0.23* 0.05 73.00
73.00
Correlation coefficient Significance (two-sided) N
0.35* 0.00 71
0.10 0.42 71
1.00 71
Po0:01 (two-sided).
were no significant differences between the sites from Kleinblittersdorf to Rehlingen.
Table 5 Correlations (Spearman–Rho) between alkylphenols and vtg concentrations in male breams
3.4. Correlations To screen the impact of elevated vtg concentrations on male breams, vtg values were correlated with the HSI and the corpulence factor. Values of vtg under and over the limit of quantification were excluded. A nonparametric test was used because the vtg concentrations were not Gaussian distributed. Table 4 shows weak but highly significant correlations between vtg concentrations and the HSI, whereas vtg concentrations and the corpulence factor do not correlate. To verify whether alkylphenols induce vtg in male breams the concentrations of both variables were also correlated, excluding 4tOP because the values were low and did not differ greatly. Table 5 clearly shows no correlation between alkylphenols and vtg concentrations, but relatively strong and highly significant correlations between 4NP and 4NP1EO.
4. Discussion Corpulence factors serve as indicators of comparability between the breams from different sampling sites because they show the relationship between the length and weight of more or less coeval fish (8–12 years old). This means that the breams of all sampling sites were more or less well fed. Therefore, one can assume that the fish samples were comparable, meeting an important prerequisite for comparable data. The HSI is considered an indicator of estrogenic effects (Harries et al., 1997; Karbe et al., 2000; Karels et al., 2001). Folmar et al. (2001b) discussed the accumulation of vtg in the livers of male fish, which are probably not able to excrete the produced vtg via the kidneys in an adequate way. Consequently, the accumulated vtg
vtg vtg
4NP
4NP1EO
Spearman–Rho Po0:01 (two-sided) N
1.00 47
Spearman–Rho Po0:01 (two-sided) N
0.14 0.36 47
Spearman–Rho Po0:01 (two-sided) N
0.20 0.18 47
4NP
4NP1EO
0.14 0.36 47
0.20 0.18 47
1.00
0.81
47 0.81 47
47 1.00 47
results in a hypertrophy of liver cells and of the liver itself. One should note that chemical substances other than endocrine disruptors and lipid accumulation could also cause liver hypertrophy (Karels et al., 2001). With the exception of Rehlingen and Trier, the HSI of male beams from all sampling sites was higher than 1.5. Compared to the long time series of the German Environmental Specimen Bank, values 41.5 may be considered to be elevated; breams from the sampling site of Gu¨dingen had, in most years, the highest HSIs of between 1.8 and 2.4 (Bartel et al., 2002). We stress that this study was conducted in the spring, whereas investigations of the German ESB are conducted in summer, outside the spawning season. Therefore, absolute amounts should be compared with care. Overall, one can conclude that at least the HSI values from Gu¨dingen and Burbach indicate liver hypertrophy. Nutritional status did not seem to impact the differences in liver size between the sampling sites because there was no correlation between the corpulence factor and the HSI. In contrast, a low but highly significant correlation of the HSI and vtg could suggest a certain influence of vtg on liver hypertrophy. Nevertheless, vtg does not
ARTICLE IN PRESS R. Klein et al. / Environmental Research 98 (2005) 55–63
cause exclusively high values of the HSI, which can be seen by the fact that in Hanweiler and Rehlingen vtg and the HSI were running in the opposite direction. In conclusion, the HSI cannot be considered a reliable indicator of endocrine effects. This corresponds to earlier findings in field studies. Whereas in laboratory tests the induction of biomarkers like vtg or Zona radiata proteins by endocrine disruptors ran parallel with liver hypertrophy, this correlation was less obvious in field studies (Allen et al., 1999; Folmar et al., 2001b; Karels et al., 2001). The influence of chemical substances not acting as endocrine disruptors and the impact of natural factors (climate, hydrology, nutrients, etc.) contribute to the complex factors that determine the HSI. vtg concentrations measured in the blood plasma of male breams indicate clearly that fish in the River Saar are exposed to estrogen disruptors, with the exception of those from the site of Hanweiler. They are very often significantly higher than those of fish from Hanweiler and Trier. Until now there have been few data available on vtg concentrations in male breams or in Cyprinids to allow a comprehensive assessment of measured values. Compared with two studies done by Lehmann et al. (2000) and Karbe et al. (2000) with breams from the largest German rivers, the Rhein and the Elbe, the vtg concentrations in this study were extremely high. Even the vtg concentrations in breams from Hanweiler and Trier should be considered evidence of the exposure of fish to estrogenic disruptors. In contrast, following results from Tyler et al. (1996), who found vtg concentrations of 900–1000 mg/L in female bream, an interpretation of our results should lead to different conclusions. This shows that it is a moot question whether the measured concentrations of different studies may be compared with each other because of a lack of standardization and harmonization of field and laboratory methods. Therefore, the comparison of vtg analyses of other species of cyprinids should be assessed very carefully as well (Purdom et al., 1994; Harries et al., 1995; Folmar et al., 1996; Karels et al., 2001; RodgersGray et al., 2001). Despite this one can doubtlessly assert that the elevated vtg concentrations in male breams of the River Saar are associated with the discharge of sewage treatment plants, which are assumed to be the most important source of estrogenic disruptors. This corresponds with the findings of other studies in England, Finland, and the USA on rainbow trout (Oncorhynchus mykiss), common carp (Cyprinus carpio), roach (Rutilus rutilus), walleye species (Stizostedion vitreum), and flounder (Platichthys flesus) (Harries et al., 1996, 1997; Folmar et al., 1996; Allen et al., 1999; Folmar et al., 2001a; Karels et al., 2001). In contrast, Nichols et al. (1999) could not affirm this correlation in a study of bullhead minnows (Pimephales promelas).
61
In these studies the range of discharges influencing vtg concentrations in fish were very different. In our study the longest reach was 6600 m (Kleinblittersdorf). Only Harries et al. (1996, 1997) found significantly elevated vtg concentrations compared with a reference site at a similar distance from a discharge (5000 m); the reaches were shorter in all other studies. There are many factors influencing the range of discharges from sewage treatment plants. In our study the main factors for a great range are that the River Saar is a very small, backed-up river with a very high density of sewage treatment plants. In addition, breams themselves have an impact on the range of discharges. First, they are mobile to a certain degree because their home range covers several kilometers. Second, breams usually utilize sediment as their nutrition habitat, and sediment is known to be a sink of many chemical substances (Jobling and Sumpter, 1993; Servos, 1999). Thus, breams are exposed to endogen disruptors to a greater extent than are many other fish species. Normally, it is difficult to detect chemical substances acting as estrogenic disruptors in field studies because in rivers a complex and highly dynamic mixture of chemicals affects organisms. Therefore, the results of Bo¨hmer and Wenzel (2001) seemed applicable to our study. In a German-wide study they found the highest concentrations of 4NP and 4NP1EO in the muscle of breams from the River Saar at the sampling sites of Gu¨dingen and Rehlingen, which are also included in our study. Up to now, we have known less on the accumulation of these substances in fish. Consequently, it is difficult to assess comprehensively the measured concentrations (Bo¨hmer and Wenzel, 2001). Nevertheless, the study of Bo¨hmer and Wenzel (2001), which analyzed time series of breams from Gu¨dingen and Rehlingen based on stored samples of the German Environmental Specimen Bank, allows us to classify our analysis with a comparison with the previous findings. This shows clearly that the concentrations of NP (nonylphenol) from all sampling sites were conspicuously lower than those at Gu¨dingen and Rehlingen between 1992 and 1998. In Gu¨dingen the concentrations were reduced dramatically. Whereas in 1992 and 1994 the concentrations ranged between 60 and 100 mg/kg FW, and even in 1998 concentrations of 30 mg/kg FW were found, only 7 mg/kg FW was measured 2001. A clear reduction of accumulated NP concentrations since 1992 can also be observed at Rehlingen, a reduction even more distinct between 1998 and 2001 than that seen at Gu¨dingen (30 mg/kg FW in 1998 and 5 mg/kg FW in 2001). The values of NP1EO decreased more drastically in Gu¨dingen than those of NP. In 2001 the concentrations were reduced by a factor of eight, compared to 1992 and 1994, when more than 240 mg/kg FW was measured. In Rehlingen a decrease could also be observed, but it was
ARTICLE IN PRESS 62
R. Klein et al. / Environmental Research 98 (2005) 55–63
neither as clear as the decrease in Gu¨dingen nor as dramatic as that for NP: the concentration went from 70 mg/kg FW in 1992, to about 30 mg/kg FW in 1998, and to 25 mg/kg FW in 2001. The described decrease is contradictory to those of most other chemical substances determined on a routine basis since 1992 in the muscle of breams by the German ESB, which either have remained on the same level or increased since the beginning of the investigation (Wagner et al., 2004). A decrease has only been observed for Se, y-HCH, and HCB in Gu¨dingen and y-HCH in Rehlingen. Compared to sampling sites in the Rivers Elbe, Rhein, Donau, Mulde, and Saale, the River Saar is characterized by the highest levels of PCB (especially PCB 52 and 101) and—despite the decrease—of y-HCH. Additionally, the site of Rehlingen shows the highest levels of Tl and As, since an increase in 1995. The more or less drastic decrease of AP and APE could mean that these chemical substances did not cause the induction of vtg in the liver of male breams. This assumption is confirmed by two factors: there was more or less no clear differentiation in concentrations of 4NP and 4NP1EO between the sampling sites and we found no correlation between these substances and vtg concentrations. The physiological relationship between the concentrations of NP and NP1EO in blood plasma and the concentration of vtg in the muscle is indirect because the concentration in muscle tissue can be used as an indicator of concentrations in liver. Based on this, it is probable that the above-mentioned assumption is true. In this context, it is important that octylphenol, as the most potent estrogenic disruptor of the analyzed AP and APE (Servos, 1999), showed no spatial differentiation and very low values. This is further evidence that AP and APE did not impact highly the induction of vtg in the liver of male breams. Consequently, it remains ambiguous which chemical substance in the effluent caused the elevated vtg concentrations in male breams in the River Saar between Kleinblittersdorf and Rehlingen. Furthermore, it is difficult to prove to what extent chemical substances determined by the German ESB (chlorinated hydrocarbons and some elements) contribute to the production of vtg because these data are available only for Gu¨dingen and Rehlingen, both of which show elevated vtg concentrations. Besides AP and APE, natural estrogens (17X-estradiol und esterone) and synthetic estrogens (17a-ethynyl-estradiol) are the main components of estrogen disruptors in municipal waste water (Rodgers-Gray et al., 2001). Therefore, we cannot determine to what extent they each affect the vtg concentrations found. Overall, one can conclude that vtg is a reliable indicator of exposure to estrogenic disruptors and of a first estrogen effect in organisms but not the overall relevance of estrogenic disruptors for biota.
Acknowledgments This study was funded by the German Environmental Agency (UBA) in Berlin. We thank Ms. Petra Willems and Mr. Bernd Fontaine from the University of Trier for ELISA analyses and fishing, respectively.
References Allen, Y., Matthiessen, P., Scott, A.P., Haworth, S., Feist, S., Thain, J.E., 1999. The extent of oestrogenic contamination in the UK estuarine and marine environments—further surveys of flounder. Sci. Total Environ. 233, 5–20. Arukwe, A., Kullmann, S.W., Hinton, D.E., 2001a. Differential biomarker gene and protein expressions in nonylphenol and estradiol-17X treated juvenile rainbow trout (Oncorhynchus mykiss). Comp. Biochem. Physiol. 129, 1–10. Arukwe, A., Yadetie, F., Male, R., Goksoyr, A., 2001b. In vivo modulation of nonylphenol-induced zonagenesis and vitellogenesis by the antiestrogen, 3,30 4,40 -tetrachlorobiphenyl (PCB-77) in juvenile fish. Environ. Toxicol. Pharmacol. 10, 5–15. Bartel, M., Klein, R., Nentwich, K., Paulus, M., Wagner, G., 2002. Beitra¨ge zur Probenahme und Probenbeschreibung fu¨r die Umweltprobenbank des Bundes—Jahresbericht 2001. Univ. Trier, Trier, Germany, unpublished, 145pp. Bo¨hmer, W., Wenzel, A., 2001. Teilbericht II: Alkylphenole und Alkylphenolethoxylate. In: Umweltbundesamt (Hrsg.), Organische Zinnverbindungen, Alkylphenole und Bisphenol A in marinen und limnischen Biota der Umweltprobenbank. UBA Texte 06/01, pp. 63–122. Folmar, L.C., Denslow, N.D., Rao, V., Chow, M., Crain, D.A., Enblom, J., Marcino, J., Guilette Jr., L.J., 1996. Vitellogenin induction and reduced serum testosterone concentrations in feral male carp (Cyprinus carpio) captured near a major metropolitan sewage treatment plant. Environ. Health Perspect. 104/10, 1096–1101. Folmar, L.C., Denslow, N.D., Kroll, K., Orlando, E.F., Enblom, J., Marcino, J., Metcalfe, C., Guillette Jr., L.J., 2001a. Altered serum sex steroids and vitellogenin induction in walleye (Stizostedion vitreum) collected near a metropolitan sewage treatment plant. Arch. Environ. Contam. Toxicol. 40, 392–398. Folmar, L.C., Gardner, G.R., Schreibman, M.P., Magliulo-Cepriano, L., Mills, L.J., Zaroogian, G., Gutjahr-Gobell, R., Haebler, R., Horowitz, D.B., Denslow, N.D., 2001b. Vitellogenin-induced pathology in male Summer Flounder (Paralichthys dentatus). Aquat. Toxicol. 51, 431–441. Harries, J., Jobling, S., Matthiessen, P., Sheahan, D.A., Sumpter, J.P., 1995. Severity and extent of contamination of some English rivers and drinking water reservoirs by estrogenic substances. Report to UK Department of the Environment, Ref. PECD 7/7/384. Harries, J.E., Sheahan, D.A., Jobling, S., Matthiessen, P., Neall, P., Routledge, E.J., Rycroft, R., Sumpter, J.P., Tylor, T., 1996. A survey of estrogenic activity in United Kingdom inland waters. Environ. Toxicol. Chem. 15/11, 1993–2002. Harries, J.E., Sheahan, D.A., Jobling, S., Matthiessen, P., Neall, P., Sumpter, J.P., Tylor, T., Zaman, N., 1997. Estrogenic activity in five United Kingdom rivers detected by measurement of vitellogenesis in caged male trout. Environ. Toxicol. Chem. 16/3, 534–542. Hecker, M., 2001. Natu¨rliche Variabilita¨t endokriner Funktionen und ihre Modulation durch anthropogene Einflu¨sse—Untersuchungen am Beispiel des Brassen (Abramis brama [L.]) in der Elbe und in einem Referenzgewa¨sser. Berichte aus dem Zentrum fu¨r Meeresund Klimaforschung Reihe E, Nr 16, 149.
ARTICLE IN PRESS R. Klein et al. / Environmental Research 98 (2005) 55–63 Jobling, S., Nolan, M., Tyler, C.R., Brighty, G., Sumpter, J.P., 1998. Widespread sexual disruption in wild fish. Environ. Sci. Technol. 32, 2498. Jobling, S., Sumpter, J.P., 1993. Detergent components in sewage effluent are weakly oestrogenic to fish: an in vitro study using rainbow trout (Oncorhynchus mykiss) hepatocytes. Aquat. Toxicol. 27, 361–372. Karbe, L., Westendorf, J., Beck, M.-P., Gutendorf, B., Hecker, M., 2000. O¨kotoxikologische Relevanz endokrin wirksamer Stoffe in der aquatischen Umwelt—Untersuchungen am Modell Elbe. AbschluXbericht FE-Vorhaben 29765001/02, Univ. Hamburg, p. 113. Karels, A., Markkula, E., Oikari, A., 2001. Reproductive, biochemical, physiological, and population responses in perch (Perca fluviatilis L.) and roach (Rutilus rutilus L.) downstream of two elemental chlorine-free pulp and paper mills. Environ. Toxicol. Chem. 20/7, 1517–1527. Klein, R., Paulus, M., 1996. Richtlinie zur Probenahme und Probenbearbeitung, Transport und Lagerung—Limnische Probenarten—Brassen (Abramis brama). In: Umweltbundesamt (Hrsg.), Umweltprobenbank des Bundes—Verfahrensrichtlinien: Teil 6. Erich Schmidt Verlag, Berlin, pp. 1–26. Lehmann, J., Paris, F., Stu¨renberg, F.-J., Blu¨m, V., 2000. O¨kotoxikologische Untersuchungen an freilebenden Brassen, LO¨BFJahresbericht 1999, pp. 127–131. Nichols, K.M., Miles-Richardson, S.R., Snyder, E.M., Criesy, J.P., 1999. Effects of exposure to municipal wastewater in situ on the reproductive physiology of the fathead minnow (Pimephales promelas). Environ. Toxicol. Chem. 18, 2001–2012.
63
Oehlmann, J., Schulte-Oehlmann, U., Stroben, E., Bauer, B., Bettin, C., Fioroni, P., 1995. Androgene Effekte zinnorganischer Verbindungen bei Mollusken. In: Umweltbundesamt (Ed.), Umweltchemikalien mit endokriner Wirkung, UBA-Texte 65/95, Berlin, pp. 77–85. Purdom, C.E., Hardiman, P.A., Bye, V.J., Eno, N.C., Tyler, C.R., Sumpter, J.P., 1994. Estrogenic effects of effluents from sewage treatment works. Chem. Ecol. 8, 275–285. Rodgers-Gray, T.P., Jobling, S., Kelly, C., Morris, S., Brighty, G., Waldock, M.J., Sumpter, J.P., Tyler, C.R., 2001. Exposure of juvenile roach (Rutilus rutilus) to treated sewage effluent induces dose-dependent and persistent disruption in gonadal duct development. Environ. Sci. Technol. 35/3, 462–470. Servos, M.R., 1999. Review of the aquatic toxicity, estrogenic responses and bioaccumulation of alklylphenols and alkylphenol polyethoxylates. Water Qual. Res. J. Can. 34/1, 123–177. Sumpter, J.P., Jobling, S., 1995. Vitellogenesis as a biomarker for estrogenic contamination of the aquatic environment. Environ. Health Perspect. 103, 173–178. Tyler, C.R., van der Eerden, B., Joblin, S., Panter, G., Sumpter, J.P., 1996. Measurement of vitellogenin, a biomarker for exposure to oestrogenic chemicals, in a wide variety of cyprinid fish. Comp. Biochem. Physiol. B 166, 418–426. Tyler, C.R., Jobling, S., Sumpter, J.P., 1998. Endocrine disruption in wildlife: a critical review of the evidence. Crit. Rev. Toxicol. 28/4, 319–361. Wagner, G., Quack, M., Paulus, M., Nentwich, K., Klein, R., Bartel., M., 2004. Beitra¨ge zur Probenahme und Probenbeschreibung fu¨r die Umweltprobenbank des Bundes–Jahresbericht 2003. Univ. Trier, Trier, Germany, unpublished, 152pp.