Environmental Toxicology and Chemistry, Vol. 27, No. 3, pp. 519–528, 2008
䉷 2008 SETAC
Printed in the USA
0730-7268/08 $12.00 ⫹ .00
POLYCHLORINATED NAPHTHALENES AND OTHER DIOXIN-LIKE COMPOUNDS IN
ELBE RIVER SEDIMENTS
WERNER BRACK,*† LUDĚK BLÁHA,‡ JOHN P. GIESY,§㛳# MATTHIAS GROTE,† MONIKA MOEDER,†
STEFFI SCHRADER,† and MARKUS HECKER#
†UFZ Helmholtz Centre for Environmental Research, Permoserstraße 15, D-04318 Leipzig, Germany
‡Research Centre for Environmental Chemistry and Ecotoxicology (RECETOX), Masaryk University, Kamenice 3,
CZ62500 Brno, Czech Republic
§Department Veterinary Biomedical Science and Toxicology Centre, University of Saskatchewan, Saskatoon, Saskatchewan, Canada
㛳City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, Special Administrative Region, China
#Department of Zoology, National Food Safety and Toxicology Center, Center for Integrative Toxicology, Department of Zoology,
Michigan State University, East Lansing, Michigan 48824, USA
( Received 3 July 2007; Accepted 29 August 2007)
Abstract—Contamination of Elbe River (Germany) sediments with dioxin-like toxicants was investigated following the 500-year
flood (flood that statistically occurs once in 500 years) of 2002. It was hypothesized that large amounts of particulate matter from
river beds and associated dioxin-like toxicants were mobilized and transported during this flood event. The investigation focused
on polychlorinated naphthalenes (PCNs) that have not been determined previously in the Elbe River. The in vitro H4IIE-luc assay
was used as an overall measure for toxicants capable of binding to the aryl hydrocarbon receptor (AhR). The assay was combined
with congener-specific instrumental analyses and fractionation to quantify PCN contributions to total AhR-mediated activity relative
to polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) and polychlorinated biphenyls (PCBs). Penta- to octachloronaphthalene concentrations of 30 ng/kg dry weight up to 13 ␮g/kg dry weight were found in Elbe River sediments downstream of
Bitterfeld. Concentrations of penta- to octachloronaphthalenes, however, were only approximately 3 ␮g/kg dry weight at a site in
the vicinity of Bitterfeld, where a level of approximately 3 mg/kg dry weight was reported before the flood. Also, the congener
pattern of PCNs at this site changed after the flood, and PCN patterns reported previously for Bitterfeld and assigned to chloralkali electrolysis with graphite electrodes could now be observed at the sites from downstream of Bitterfeld and Magdeburg.
Whereas PCDD/Fs dominated the dioxin-like activity in the middle and lower Elbe River, PCNs contributed as much as 10% of
the total AhR-mediated activity. The contribution of PCBs was less significant (maximum, 0.2%). Thus, in Elbe River sediments,
PCNs should be considered as relevant contaminants and be included in future monitoring and risk assessment programs.
Keywords—Polychlorinated naphthalenes
Contamination
Dioxin-like activity
Flood 2002
Aryl hydrocarbon receptor
(PCNs) [17]. The contamination of Elbe River sediments with
PCDD/Fs and PCBs has been well described and known since
the early 1990s [4,18]. Toxicity equivalents (TEQs) of approximately 100 ng/kg dry weight for PCDD/Fs and 10 ng/kg dry
weight for PCBs have been found in sediments of the lower Elbe
River [4]. Studies conducted after the flood in 2002 reported
elevated PCDD/F concentrations in soils of floodplains downstream of Bitterfeld as well as in feedstuff produced on the floodplains and some products from animals fed with feedstuff grown
on the contaminated floodplains [19,20]. For PCDD/Fs, the industrial region of Bitterfeld has been identified as the primary
source, whereas for PCBs, sources and pollution patterns are less
clear. Polychlorinated naphthalenes have been found only recently
in sediments from Spittelwasser Creek, which drains the Bitterfeld
region [21]. Possible sources of PCNs include technical Halowax威 (Koppers, Pittsburg, PA, USA) and Aroclor威 (Monsanto,
St. Louis, MO, USA) mixtures, incineration processes, chloralkali electrolysis, and metallurgy processes [22,23]. The congener pattern observed previously in Spittelwasser Creek suggests
that chlor-alkali industry is a likely source of PCN contamination
in Elbe River sediments [21]. It was hypothesized that remobilization of sediments in the Bitterfeld region, such as by flood
events, could have contaminated the Elbe River downstream of
this region. To date, however, no data are available regarding
PCN contamination of sediments in the Elbe River and the contribution of PCNs to dioxin-like potency of sediment extracts.
INTRODUCTION
Sediments from the Elbe River are contaminated with a wide
variety of organic pollutants [1]. These include chlorinated hydrocarbons, such as dichlorodiphenyltrichloroethane and its metabolites, polychlorinated biphenyls (PCBs), hexachlorocyclohexanes, hexachlorobenzene, chlorostyrenes, and polychlorinated
dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) [2–5]; pesticides [6,7]; and organometallic compounds [8]. Elbe River sediments also are significantly toxic to alga, luminescent bacteria,
nematodes, macroinvertebrates, and fish embryos [9,10], and they
exhibit mutagenicity [11] and aryl hydrocarbon receptor (AhR)–
mediated (dioxin-like) activity [10].
A sensitive response to the binding of contaminants to the
AhR is the induction of ethoxyresorufin-O-deethylase (EROD)
[12]. Significantly elevated activities of EROD have been found
in bream caught at seven sites along the German stretch of the
Elbe River [13], confirming the relevance of dioxin-like toxicants
in this important Central European freshwater ecosystem. Major
groups of toxicants known to cause AhR-mediated responses of
vertebrates include, among many others, polycyclic aromatic hydrocarbons (PAHs) [14] and halogenated aromatic hydrocarbons,
such as coplanar (non- and mono-ortho-substituted congeners)
PCBs [15], PCDD/Fs [16], and polychlorinated naphthalenes
* To whom correspondence may be addressed
(werner.brack@ufz.de).
Published on the Web 10/31/2007.
519
520
Environ. Toxicol. Chem. 27, 2008
W. Brack et al.
Fig. 1. Sampling sites (䢇) along the Elbe River were Königstein (KON), Dresden (DRE), Torgau (TOR), Wittenberg (WIT), Dessau Leopoldhafen
(LEO), Barby (BAR), Magdeburg (MAG), Arneburg (ARN), und Hitzacker (HIT). One sampling site was situated in Spittelwasser Creek (SPI),
a tributary to the Mulde River.
The flooding that occurred in the Elbe River basin in 2002 also
affected the Bitterfeld region and was hypothesized to have translocated large amounts of sediments [10]. Thus, this flooding could
have contributed to the contamination of downstream Elbe River
sediments, but it also may have changed the contamination patterns in Bitterfeld sediments.
The present study was conducted to identify possible floodinduced changes in contamination patterns of sediments from
Spittelwasser Creek (Bitterfeld), which is believed to be one
of the most relevant tributaries with respect to dioxin-like contamination in the Elbe River [4,24]; to evaluate concentrations
of dioxin-like compounds in sediments of Elbe River upstream
and downstream of Bitterfeld, with a major focus on PCNs;
and to evaluate the contribution of PCNs to the dioxin-like
activity in Elbe River sediments. To achieve these aims, concentrations of PCBs, PCDD/Fs, and PCNs, as determined by
congener-specific instrumental analyses, were combined with
group-specific fractionation of PCBs, PCDD/Fs, and PCNs
[25] and compared to the total potency of the AhR-mediated
compounds. The total AhR potency was determined by use of
bioanalytical assessment with H4IIE-luc cells, which have
been stably transfected with an AhR-dependent luciferase reporter system [26–28].
MATERIALS AND METHODS
Sample collection and preparation
Sediments were collected during March and April of 2003
from nine transect sites in the German part of the Elbe River,
and one additional sample was taken from the tributary Spittelwasser Creek downstream of Bitterfeld (SPI) [10] (Fig. 1).
Sampling was focused on sites located downstream of the
Czech border, which are characterized by exposure to effluents
of the petrochemical plants and other industries upstream of
the border, and on the middle course of the Elbe River, which
is characterized by increased exposure to a variety of contaminants from the major tributary rivers Mulde and Saale as well
as from a large number of known communal and industrial
point sources. Samples were frozen at ⫺20⬚C, freeze-dried,
and sieved (⬍63 ␮m). Subsequently, sediments were extracted
using accelerated solvent extraction with a toluene/acetone
mixture (70:30, v/v) at 140⬚C and a pressure of 10 MPa [29].
Before further analyses, sulfur was removed with activated
copper, because it interferes with both chemical analysis and
the cells in the H4IIE-luc assay [30].
PCNs in Elbe River sediments
Cleanup and fractionation
Desulfurized extracts were adsorbed to alumina neutral
(ICN Biomedicals, Eschwege, Germany) that had been deactivated with 4.5% double-distilled water and then separated
based on the methods described below. Samples were loaded
onto an alumina column, and nonpolar aliphatic compounds
were eluted with n-hexane (HEX). Subsequently, nonpolar aromatic compounds were eluted with a mixture of HEX and
dichloromethane (DCM; 90:10, v/v) [30]. Polar compounds,
including nitro-, oxy-, hydroxy-, and amino-substituted chemicals, were retained on the column. The nonpolar aromatic
fraction was separated into several fractions by preparative
high-performance liquid chromatography on a nitrophenylpropyl silica column (5-␮m Nucleosil 100-5 NO; Macherey and
Nagel, Düren, Germany) with 19 ml/min of HEX/DCM (95:
5, v/v) as a mobile phase according to the number of aromatic
rings [25]. The first fraction eluting within the initial 6 min
contained parent, alkylated, and halogenated naphthalenes, biphenyls, dibenzo-p-dioxins, dibenzofurans, and other compounds with two aromatic rings. This fraction was collected
for further assessment of dioxin-like toxicants, whereas PAHs
with more than two rings and longer retention times were
eluted and stored for future assessments.
The PCDD/Fs and PCNs with more than four chlorine atoms
were separated from the PCBs, PCDD/Fs, and PCNs that contained fewer chlorine atoms and nonchlorinated compounds by
use of electron donor–acceptor chromatography on a stainlesssteel guard column (10 ⫻ 20 mm) packed with 2-(1-pyrenyl)ethyldimethylsilylated silica (5-␮m Cosmosil PYE; Nacalai
Tesque, Kyoto, Japan) with an average pore diameter of 120 Å
at a mobile-phase flow rate of 8 ml/min. After isocratic elution
with HEX for 5 min, a gradient to 50:50 (v/v) HEX/DCM within
1 min was performed, followed by isocratic elution with 50:50
(v/v) HEX/DCM. Fraction F2.1.1, which contained PCBs, including the mono- and non-ortho-chlorinated congeners 156, 77,
126, and 169, was collected within the first 66 s. This elution
pattern was confirmed by use of standards. Nonhalogenated compounds and PCNs containing up to three or four chlorine substituents also were collected in this fraction. Polychlorinated
naphthalenes substituted with more chlorine atoms and PCDD/
Fs with four and more chlorine atoms were collected in fraction
F2.1.2 (from 66 s to 11 min). This fraction was demonstrated to
coelute with neat standards, including 2,3,6,7-tetrachlorodibenzofuran, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), 2,3,6,7-tetrachloronaphthalene,
1,2,3,4,6,7-hexachloronaphthalene,
1,2,3,4,5,6,7-heptachloronaphthalene, octachlorodibenzo-p-dioxin, and octachlorodibenzofuran. Fraction F2.1.2 was further separated by size-exclusion chromatography [25] on two stainlesssteel columns (25 ⫻ 600 mm) packed with porous polystyrene–
divinylbenzene copolymer with a pore size of 50 Å and a particle
size of 10 ␮m (PLgel; Polymer Laboratories, Waltrop, Germany)
with 10 ml/min of tetrahydrofuran as mobile phase. Two fractions
were collected: F2.1.2.1, which eluted between 33 and 40 min,
contained primarily PCDD/Fs, and coeluted with the standards
2,3,7,8-tetrachlorodibenzofuran, 2,3,7,8-TCDD, and octachlorodibenzo-p-dioxin; and F2.1.2.2, which eluted between 40 and 43
min, contained primarily PCNs, and coeluted with the standards
2,3,6,7-tetrachloronaphthalene, 1,2,3,6,7-pentachloronaphthalene, 1,2,3,4,6,7-hexachloronaphthalene, and 1,2,3,4,5,6,7-heptachloronaphthalene. The validation of the fractionation method has
been described by Brack et al. [25].
Environ. Toxicol. Chem. 27, 2008
521
High-resolution gas chromatography with mass-selective
detection
Identification and quantification of individual compounds
were accomplished using a gas chromatograph (HP 5890 II;
Hewlett-Packard, Waldbronn, Germany) coupled with a highresolution mass spectrometer (Finnigan MAT 95, Bremen,
Germany). Compounds were separated on a fused silica capillary column (length, 60 m; inner diameter, 0.25 mm; film
thickness, 0.25 ␮m; DB-5MS; J&W Scientific, Folsom, CA,
USA) using a constant flow of helium of 1.5 ml/min with an
initial pressure of 1.86 ⫻ 105 Pa at 105⬚C. A guard capillary
was used in front of the analytical column to prevent its contamination. The column oven temperature was programmed to
increase from 105 to 180⬚C at a rate of 30⬚C/min and then to
260⬚C at 1.4⬚C/min, followed by a rate of 30⬚C/min to a final
temperature of 305⬚C with a final holding time of 10 min.
Injection was performed on-column. The transfer-line temperature was held at 290⬚C. The mass spectrometer was operated at electron-impact ionization (70 eV) and at mass resolution of 8,000. All compounds were quantified in the multiple-ion detection mode using the most abundant ions of the
molecular ion clusters for quantification and two further substance-typical ions as qualifier. Polychlorinated dibenzo-p-dioxins and dibenzofurans were quantified using 13C-labeled internal PCDD/F standards (EDF 8999 and EDF 957; Cambridge
Isotope Laboratories, Andover, MA, USA). For the quantification of PCBs and PCNs, 13C-labeled internal PCB standards
were used (EC1418, EC1419, and EC4064; Cambridge Isotope
Laboratories). Detection limits for PCBs, PCNs, and
PCDD/Fs were approximately 1.0 ng/kg dry weight sediment
calculated for the whole protocol using a defined signal to
noise ratio of 3:1. Mean standard deviations ranged from 6 to
18% (n ⫽ 3). No recovery standards were used to avoid artefact
responses in the in vitro H4IIE-luc assay.
Within the analyses series, after every fourth analysis, a
blank analysis (injection of 1 ␮l of pure toluene) was carried
out, and subsequently, a standard mixture (EC1418, EC1419,
and EC4064) was measured to check instrument performance.
Disturbing interferences and memory effects were prevented
by regularly maintenance of the injector port and the guard
capillary. The mass spectrometer was tuned every day with
heptacosa (perfluorotributylamine) purchased from Thermo
Scientific (Bremen, Germany).
In vitro analysis of dioxin-like activity
The potencies of the samples to induce AhR-mediated toxicity were determined with the H4IIE-luc bioassay. The details
of the luciferase reporter gene assay have been described elsewhere [27,28]. In brief, H4IIE-luc cells were seeded into 96well Culture View Plates娂 (Packard, Meriden, CT, USA).
Cells were exposed in triplicate to serial dilutions of sediment
extracts for 72 h. The amount of AhR-induced luciferase was
quantified using the Herta Luminometer with LucLite威 Reporter Gene Assay System (PerkinElmer, Nieuwerkerk a/d
IJssel, The Netherlands). Total concentrations of TCDD equivalents (TCDD-EQ) in sediments were determined by comparing the volume of extract required to cause a specified response
to the mass of TCDD required to cause the same response and
normalized to the mass of sediment that had been extracted.
The resulting units of quantification were pg TCDD-EQ/g dry
weight sediment. After an initial screening (range-finding experiments), full concentration–response curves for induction
of AhR-mediated responses were generated in triplicate.
522
Environ. Toxicol. Chem. 27, 2008
W. Brack et al.
Data analysis
For the calculation of TCDD-EQ from H4IIE-luc assay
results, the effect-equivalency approach was applied
[28,31,32]. Regression equations were derived for the loglinear portion of the standard TCDD curve and for each concentration–response curve of all sample/fractions. 2,3,7,8-Tetrachlorodibenzo-p-dioxin equivalents were then calculated
from the amount of sample producing a twofold induction of
luciferase.
Total TEQs were calculated as the sum of the product of
concentrations of individual congeners multiplied by their relative potencies (RePs) and normalized to sediment (dry wt)
and total organic carbon (TOC). For PCNs, no generally agreed
toxic equivalency factors are available. For a series of PCN
congeners, however, potencies relative to 2,3,7,8-TCDD have
been described using two test systems based on H4IIE cells
(H4IIE-EROD and H4IIE-luc) [17,33–36]. For TEQ calculations in the present study, RePs determined in the H4IIE-luc
assay (RePH4IIE-luc) were used when available. For those congeners for which RePH4IIE-luc were not available, RePs were
based on the induction of EROD in wild-type H4IIE cells. For
PCNs, in some cases values predicted for quantitative–structure activity relationships were used. In vitro RePs as well as
predicted in silico values were compiled recently [37].
For source analysis, concentrations of individual penta- to
heptachloronaphthalene congeners were normalized on total
contents in the respective sample and subjected to principal
components analysis together with sediment data from Spittelwasser Creek before the flood [21] and source-related samples published previously [23].
RESULTS
AhR-mediated activity (H4IIE-luc)
Fig. 2. Dose–response relationships of dioxin-like activity of the sediment-extract fractions F2.1.1 (polychlorinated biphenyl fraction),
F2.1.2.1 (polychlorinated dibenzo-p-dioxin and dibenzofuran fraction), and F2.1.2.2 (polychlorinated naphthalene fraction) from different sites at the Elbe River and the tributary Spittelwasser Creek in
the order of the course of the river given as fold-induction of luciferase. For comparison, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)
is shown. The following doses of TCDD were applied: From right,
240, 48, 9.6, 1.9, 0.38, and 0.07 pg/ml. The following samples were
taken: From right, 250, 100, 20, 5, 1, and 0.33 mg sediment dry wt/
ml. For site acronyms, see Figure 1.
Except for sediments collected at Königstein, PCB fractions
(F2.1.1) of all samples induced luciferase more than twofold and
in a dose-dependent manner (Fig. 2) with concentrations of
TCDD-EQ that exceeded the lowest-observable-effect concentration. With the exceptions of the samples collected at Königstein, Dresden, and Barby that had less activity and the relatively
great concentrations at Hitzacker (HIT), all TCDD-EQ concentrations were in the range of 20 to 90 ng/kg dry weight (Table
1). The maximum TCDD-EQ concentration of 7.6 ⫻ 102 ng/kg
dry weight (1.2 ⫻ 104 ng TCDD-EQ/kg TOC) was observed at
HIT. Luciferase induction profiles of F2.1.1 varied along the Elbe
River and did not show a clear trend (Fig. 2).
Table 1. 2,3,7,8-Tetrachlorodibenzo-p-dioxin equivalent quantity (TCDD-EQ) per dry weight and normalized to total organic carbon (TOC) in
the sediment-extract fractions from nine sites on the Elbe River and one on the tributary Spittelwasser Creek based on lowest-effect concentrations
(twofold induction above background level)a
TCDD-EQ (ng/kg dry wt)
Site
KON
DRE
TOR
WIT
SPI
LEO
BAR
MAG
ARN
HIT
a
F2.1.1
⬍1.13
5.61
28.94
26.02
32.37
84.45
1.13
40.57
34.97
755.77
For site acronyms, see Figure 1.
TCDD-EQ (ng/kg TOC)
F2.1.2.1
F2.1.2.2
5.74
⬍1.13
⬍1.13
6.82
618.12
60.63
72.93
269.73
⬍1.13
8.12
⬍1.13
⬍1.13
⬍1.13
⬍1.13
195.11
14.05
19.25
25.42
2.26
⬍1.13
F2.1.1
⬍94.17
133.89
699.03
397.25
149.86
920.93
24.51
636.89
506.81
12,170.21
F2.1.2.1
478.33
⬍26.97
⬍27.29
104.12
2,861.67
661.18
1,582.00
4,234.38
⬍16.38
130.76
F2.1.2.2
⬍94.17
⬍26.97
⬍27.29
⬍17.25
903.29
153.22
417.57
399.06
32.75
⬍18.20
Environ. Toxicol. Chem. 27, 2008
PCNs in Elbe River sediments
523
Table 2. Contamination of Elbe River and tributary sediments with polychlorinated biphenyls (PCBs), polychlorinated dibenzo-p-dioxins and
dibenzofurans (PCDD/Fs), and polychlorinated naphthalenes (PCNs) in toxicity equivalents (TEQs) normalized to dry weight and total organic
carbon (TOC)a
Site
KON
DRE
TOR
WIT
SPI
LEO
BAR
MAG
ARN
HIT
a
TOC (%)
1.2
4.19
4.14
6.55
21.6
9.17
4.61
6.37
6.9
6.21
PCBs
PCDD/Fs
PCNs
ng TEQ/kg dry wt ng TEQ/kg TOC
ng TEQ/kg dry wt ng TEQ/kg TOC
ng TEQ/kg dry wt ng TEQ/kg TOC
0.10
0.89
1.43
1.92
29.54
1.95
1.30
1.03
0.70
1.40
8.34
21.24
34.54
29.31
136.76
21.26
28.20
16.17
10.14
22.54
0.28
1.61
1.23
0.31
1,564.24
103.96
55.82
254.39
4.23
6.33
23.33
38.42
29.71
4.73
7,241.85
1,133.70
1,210.85
3,993.56
61.30
101.93
0.26
0.11
0.82
0.29
0.68
4.00
5.77
15.64
0.06
0.68
21.67
2.63
19.81
4.43
3.15
43.62
125.16
245.53
0.87
10.95
For site acronyms, see Figure 1.
Concentrations of TCDD-EQ in F2.1.2.1 (PCDD/F fractions) upstream of Bitterfeld did not exceed 10 ng TCDD-EQ/
kg dry weight (Fig. 2). Downstream of Bitterfeld, concentrations of greater than 600 ng TCDD-EQ/kg dry weight occurred
at SPI and 60 to 270 ng TCDD-EQ/kg dry weight at Elbe
River sites downstream of Bitterfeld (Leopoldhafen [LEO] to
Magdeburg [MAG]) could be observed. Further downstream
(Arneburg and HIT), the activity again decreased below 10
ng/kg dry weight. Both on a dry-weight and a TOC basis,
TCDD-EQ concentrations were greatest at SPI, followed by
MAG and Barby (Table 1).
Ethoxyresorufin-O-deethylase activities measured in PCN
fractions (F2.1.2.2) were generally less upstream of Bitterfeld,
with less than 1 ng TCDD-EQ/kg dry weight (Table 1). The
greatest EROD activities were found in sediments collected at
SPI, with 195 ng TCDD-EQ/kg dry weight. Elevated TCDD-EQ
levels in Elbe River sediments were detected downstream of
Bitterfeld until MAG, with 10 to 30 ng TCDD-EQ/kg dry weight.
Sediment contamination with dioxin-like compounds
The sediment extract fractions described above also were
analyzed for planar PCBs, PCDD/Fs, and PCNs to explain the
AhR-mediated activity in these samples. The least proportion
of TEQ contributed by dioxin-like PCBs was determined to
be in sediments just below the Czech border at Königstein
(0.1 ng TEQ/kg dry wt; 8.3 ng TEQ/kg TOC) (Table 2). Concentrations increased further downstream to 20 to 30 ng TEQ/
kg TOC for the middle course of the Elbe River from Torgau
to MAG. The concentrations observed downstream of Bitterfeld were in the same range as those observed upstream despite
the 10-fold greater concentrations detected in SPI sediments.
Concentrations of PCDD/Fs from 0.28 to 1.6 ng TEQ/kg dry
weight and from 4.7 to 38 ng TEQ/kg TOC were found in sediments of the Elbe River upstream of the Bitterfeld region. In
SPI, more than 1.5 ⫻ 103 ng TEQ/kg dry weight and 7.2 ⫻ 103
ng TEQ/kg TOC were measured. Concentrations of TEQ contributed by PCDD/Fs in Elbe River sediments downstream of
Bitterfeld exceeded those upstream by factors of 30 to 800, with
maximum concentrations of 2.5 ⫻ 102 ng TEQ/kg dry weight
and 4.0 ⫻ 103 ng TEQ/kg TOC. Concentrations of TEQs in
sediments from MAG harbor, when expressed on a TOC basis,
were only slightly less than those in SPI sediments. At the lower
Elbe River sites Arneburg and HIT, PCDD/F concentrations were
still 2- to 10-fold greater than those in Elbe River sediments from
sites upstream of Bitterfeld.
Polychlorinated naphthalenes could be detected along the
entire German stretch of the Elbe River. Concentrations of
PCNs upstream of Bitterfeld were in the range of 100 to 600
ng/kg dry weight or 0.11 to 0.82 ng TEQ/kg dry weight (Fig.
3 and Table 2). In sediments collected at SPI, PCNs have been
found at more than 2,000 ng/kg (Fig. 3), whereas concentrations of other congeners were not elevated relative to those in
Elbe River sediments. Concentrations of PCNs downstream of
Bitterfeld were greater than those detected at the sites further
upstream by a factor of 6 to 120. Sediment concentrations
increased from less than 4,000 ng PCN/kg dry weight (2.6 ng
TEQ/kg dry wt) to greater than 15,000 ng PCN/kg dry weight
(10 ng TEQ/kg dry wt) between LEO and MAG (Fig. 3 and
Table 2).
To identify possible sources of local PCN contamination,
congener-specific patterns were evaluated. Congener patterns
varied greatly upstream of Bitterfeld (TOE and Wittenberg)
(Fig. 4), but rather uniform patterns occurred downstream of
Bitterfeld (LEO, Barby, and MAG). These patterns are characterized by relatively great concentrations of PCNs 52/60, 66/
67, 73, and 75 (numbering according to Wiedmann and
Ballschmiter [38]). The pattern in SPI samples corresponded
neither to upstream nor to downstream patterns.
DISCUSSION
The analysis of AhR-mediated activity of PCB, PCDD/F,
and PCN fractions in combination with chemical analysis of
Fig. 3. Total concentrations of pentachloronaphthalenes (vertical
stripes), hexachloronaphthalenes (white), heptachloronaphthalenes
(horizontal stripes), and octachloronaphthalene (black) at 10 sites in
the Elbe basin. For acronyms, see Figure 1.
524
Environ. Toxicol. Chem. 27, 2008
W. Brack et al.
Fig. 4. Congener patterns of polychlorinated naphthalenes (PCNs) at selected sites. Numbering of PCNs is done according to Wiedmann and
Ballschmiter [38]. For site acronyms, see Figure 1.
these compounds demonstrated that the compounds follow a
diverse and fraction-specific pattern along the German stretch
of the Elbe River.
Polychlorinated biphenyls
The activity of the PCB fraction was rather constant over
the whole river stretch investigated in the present study. Exceptions were enhanced activities at LEO and, particularly,
HIT as well as lower activities at Königstein, Dresden, and
Barby. No indications were found that Bitterfeld was a significant source of or influenced PCB trends in the river stretch
related to dilution or accumulation processes. The pattern of
contamination observed indicated numerous small point sources or diffuse inputs of toxicants with AhR-mediated activity.
A comparison of TEQ concentrations based on coplanar PCBs
and TCDD-EQs in F2.1.1 (Fig. 5) suggests that except for Barby
and SPI, coplanar PCBs explain only a few percentage points of
the total concentration of TCDD-EQ in F2.1.1. This also was
true for the relatively great concentration of TCDD-EQ in sediments from HIT, where PCBs accounted for less than 0.2% of
the total TEQ concentration. Other compounds with great potential to bind to the AhR and, thus, to contribute to the total concentration of TCDD-EQ, such as PCDD/Fs, PCNs, and PAHs,
were confirmed by gas chromatography with mass-selective detection to be absent from F2.1.1, and these compounds can be
excluded as the cause of F2.1.1 activity (data not shown). The
fractionation scheme suggests unknown, nonpolar, small aromatic
compounds with relatively great AhR-mediated potency as the
cause of this effect.
Polychlorinated dibenzo-p-dioxins and dibenzofurans
Concentrations of TCDD-EQ in the PCDD/F fraction
(F2.1.2.1) were greater downstream of Bitterfeld relative to
the less polluted sites upstream (Table 1). The greatest activities were observed at SPI, a location close to the assumed
source. Concentrations of TCDD-EQ only slightly decreased
downstream until MAG, suggesting a relatively high concentration of TCDD-EQ over the whole stretch. At sites further
downstream, such as Arneburg and HIT, concentrations of
TCDD-EQ were similar to those in sediments at the sites upstream of Bitterfeld. Quantification of PCDD/Fs by gas chromatography with mass-selective detection was in agreement
with the H4II4E-luc results for the PCDD/F fraction, with
similar concentrations of TEQ and TCDD-EQ observed. The
difference between the concentrations of TCDD-EQ and TEQ
never exceeded twofold. Thus, major activities of F2.1.2.1
were explained by the presence of PCDD/Fs. The patterns of
TCDD-EQ and TEQ concentrations along transects confirmed
the Bitterfeld region as the major source of PCDD/F contamination and toxicity in the Elbe River. The proportion of TEQ
contributed by PCDD/Fs in SPI was two- to sevenfold less
than that reported before the flood [4,24], whereas the contamination of downstream Elbe River sites was of the same
order of magnitude as before the flood. Thus, no significant
influence of the 500-year flood in 2002 on PCDD/F contamination of the central and lower Elbe River basin was indicated.
Polychlorinated naphthalenes
Besides PCBs and PCDD/Fs, the present study demonstrated,
to our knowledge for the first time, that PCNs are a significant
contributor to TEQ in the central and lower Elbe River, with
approximately 10% of the total TEQ being contributed by PCNs
(relative to PCDD/Fs). Although no TCDD-EQ in the PCN fractions (F2.1.2.2) could be detected upstream of Bitterfeld, concentrations of TCDD-EQ approached almost 2.0 ⫻ 102 ng/kg dry
weight at SPI. Elevated concentrations of TCDD-EQ in PCN
PCNs in Elbe River sediments
Environ. Toxicol. Chem. 27, 2008
525
Fig. 5. Comparison of 2,3,7,8-tetrachlorodibenzo-p-dioxin equivalent quantities (ng/kg dry wt; black bars) and toxic equivalent quantities (ng/
kg dry wt; white bars) of coplanar polychlorinated biphenyls, polychlorinated dibenzo-p-dioxins and dibenzofurans, and polychlorinated naphthalenes for the fractions F2.1.1, F2.1.2.1, and F2.1.2.2, respectively. For fraction F2.1.2.2, toxic equivalent quantities of the sums of polychlorinated
naphthalenes and polychlorinated dibenzo-p-dioxins and dibenzofurans are given (shaded bars). For definitions of the fractions, see Figure 2.
For site acronyms, see Figure 1.
526
Environ. Toxicol. Chem. 27, 2008
W. Brack et al.
Fig. 6. Loading plots of the principal components (PCs) 1 and 2 (A) and PCs 1 and 3 (B) from principal component analysis on sediment data
from the present study (䉱) (for site acronyms see Fig. 1) together with sediment data from Spittelwasser Creek before the flood (SPIbf; *) [21]
and source-related data (⽧) from Järnberg et al. [23], including H14 (Halowax 1014威; Koppers, Pittsburg, PA, USA), H51 (Halowax 1051威;
Koppers), 1242 (Aroclor 1242威; Monsanto, St. Louis, MO, USA), A40 (Clophen威 A40; Bayer AG, Leverkusen, Germany), Pap (polychlorinated
biphenyl containing carbonless paper), Mwi (fly ash of a municipal waste incinerator), and Gra (graphite sludge from chlor-alkali industry). The
corresponding score plots are inset in the upper right corner. Numbering of polychlorinated naphthalenes is done according to Wiedmann and
Ballschmiter [38].
PCNs in Elbe River sediments
fractions also were observed in sediments from LEO, Barby, and
MAG, all of which are downstream of Bitterfeld. The relatively
great concentrations of TEQ in F2.1.2.2 observed at SPI, LEO,
Barby, and MAG, however, could be explained only in part by
the presence of PCNs. Major portions of the concentrations of
TEQ observed at SPI were caused by the presence of PCDD/Fs
that were not completely separated from PCNs by the size-exclusion technique used. In the case of the relatively great concentrations of TEQ contributed by PCDD/Fs relative to PCNs (a
factor of ⬃2,000), the resolution of this technique [25] obviously
needs further improvement to exclude false-positive results
caused by coelution of minor components from the adjacent fraction. In the F2.1.2.2 from the LEO, Barby, and MAG sediments,
30 to 60% of the TCDD-EQ could be explained by PCNs (Fig.
5). Polychlorinated naphthalenes and coeluting PCDD/Fs together
were responsible for 60 to 70% of the F2.1.2.2 TCDD-EQ. This
is well within the deviations of RePs that may vary by factors
up to approximately 20 even for the same test system [36].
To associate PCN contamination of Elbe River sediments
with possible sources, congener patterns of penta- to octachloronaphthalenes were measured for all sites (Fig. 4) and subjected to principle components analysis (Fig. 6) together with
source-related PCN patterns reported previously [23] and the
PCN pattern found before the flood in Bitterfeld at the identical
site (SPIbf) [21]. Bitterfeld was hypothesized to be a major
source for these compounds, because in SPI sediments before
the flood, concentrations of up to 2.5 ␮g of penta- to heptachloronaphthalenes per gram dry weight had been measured,
whereas the congener pattern suggested chlor-alkali electrolysis as a probable source [21].
Principal components analysis of concentrations of pentato heptachloronaphthalenes revealed three principal components (PCs) that together explained 87% of the variance (PC1,
47%; PC2, 28%; PC3, 12%). The resulting loading plots of
PC2 versus PC1 and PC3 versus PC1 are shown in Figure 6
together with the corresponding score plots (shown as inserts).
Highest absolute score values were found for the heptachloronaphthalene PCN 73 (PC1), the pentachloronaphthalenes PCN
52 and PCN 60 (PC2), and the pentachloronaphthalene PCN
50 (PC3). In both plots, principle components analysis grouped
the profiles of LEO, MAG, Barby, and Torgau together with
graphite sludge from chlor-alkali industry [23] and SPIbf [21].
Thus, pattern analysis and substantially increased PCN concentrations in downstream sediments suggest Bitterfeld as a
major source of PCN contamination in LEO, MAG, and Barby
sediments. Because chlor-alkali electrolysis in Bitterfeld
ceased years ago, historical burdens from Bitterfeld appear to
be the main source for the contamination of long stretches of
the central and lower Elbe River with this compound class.
Although additional downstream sources emitting the same
pattern as Bitterfeld cannot be excluded completely, the present results suggest a translocation of sediments contaminated
with PCNs from the Bitterfeld region downstream of the Elbe
River. Interestingly, the chlor-alkali pattern was found not only
downstream of Bitterfeld but also upstream at Torgau. Because
Bitterfeld can be excluded as a source of contamination, this
suggests that local emissions are responsible for the contamination at Torgau.
Analysis of sediment contamination at SPI in the present
study revealed major differences relative to the situation before
the flood of 2002 at the same site (SPIbf). Total concentrations
of PCNs were approximately 1,000-fold less. Principle components analysis clearly separates the SPI pattern from down-
Environ. Toxicol. Chem. 27, 2008
527
stream patterns and from the previous pattern at the identical
site (SPIbf). In both loading plots, the SPI patterns appear
close to Mwi (i.e., fly ash from a municipal waste incinerator)
[23]. Dominating peaks are PCNs 52/60 and 50.
The behavior of PCDD/Fs and PCNs in SPI sediments was
different, with basically unchanged contamination by
PCDD/Fs but dramatically altered PCN patterns and concentrations. It may be hypothesized that both contaminations were
associated with different portions (probably layers) of sediment. Whereas the upper layer containing PCNs from a chloralkali plant might have been removed by flood-induced erosion, the PCDD/F contamination could be relatively constant
over a much greater depth and may have been only partly
eroded during the flood event. This supports the hypothesis of
different sources for PCDD/Fs and PCNs, which are believed
to originate from metallurgical processes [4] and chlor-alkali
industry [21], respectively. A comprehensive sampling campaign, including recording depth profiles for PCNs and
PCDD/Fs, could help to prove this hypothesis and make conclusions regarding total contamination of Spittelwasser Creek
and its flood plain. The PCN contamination found in the present study after the 2002 flood at SPI obviously has a different
source and seems to be less significant. These results suggest
a highly dynamic situation regarding the qualitative and quantitative PCN contamination in sediments of different age and/
or origin that have been sampled at an identical site according
to an identical protocol with a period of approximately three
years between samplings.
The present results suggest that PCDD/Fs are the predominant dioxin-like chemicals in the central and lower Elbe River.
Contamination of Elbe River sediments with PCNs, which to
our knowledge has been examined for the first time in the
present study, equals approximately 10% of total TEQ concentrations contributed by PCDD/Fs, whereas the contribution
of PCBs was approximately 2- to 10-fold less. These results
suggest PCNs as a relevant contaminant in the lower Elbe
River. Sediment concentrations in the River Elbe were on the
same order of magnitude as concentrations at contaminated
sites of the Gdañsk Basin in the Baltic Sea [39] and Lake
Vänern in Sweden [23] but 100- to 1,000-fold less than concentrations in sediments adjacent to a former chlor-alkali plant
in southeastern coastal Georgia, USA [22]. The accuracy of
the conclusions is influenced by the uncertainty of the toxic
equivalency factors for PCNs relative to those for PCDD/Fs
and PCBs. Thus, the precautionary principle suggests further
investigations regarding PCN contamination and effects and
the inclusion of PCNs in monitoring and risk assessment programs for the Elbe River.
Acknowledgement—The present study was supported by the Federal
Ministry of Education and Research (Berlin, Germany) in a project
to assess consequences of the Elbe/Mulde flooding in the summer of
2002 (PTJ 0330492) and the European Union–funded project Models
for Assessing and Forecasting the Impact of Environmental Key Pollutants on Marine and Freshwater Ecosystems and Biodiversity (MODELKEY; contract 511237-GOCE).
REFERENCES
1. Franke S, Hildebrandt S, Schwarzbauer J, Link M, Francke W.
1995. Organic compounds as contaminants of the Elbe River and
its tributaries. Part II: GC/MS screening for contaminants of the
Elbe water. Fresenius J Anal Chem 353:39–49.
2. Bester K, Biselli S, Ellerichmann T, Hühnerfuss H, Möller K,
Rimkus G, Wolf M. 1998. Chlorostyrenes in fish and sediment
samples from the river Elbe. Chemosphere 37:2459–2471.
528
Environ. Toxicol. Chem. 27, 2008
3. Gandraß J, Zoll M. 1996. Chlorinated hydrocarbons in sediments
of the Elbe catchment area—Analytical methods and status of
pollution. Acta Hydrochim Hydrobiol 24:212–217.
4. Götz R, Steiner B, Friesel P, Roch K, Walkow F, Maaß V, Reincke
H, Stachel B. 1998. Dioxin (PCDD/F) in the river Elbe—Investigations of their origin by multivariate statistical methods. Chemosphere 37:1987–2002.
5. Heemken OP, Stachel B, Theobald N, Wenclaviak BW. 2000.
Temporal variability of organic micropollutants in suspended particulate matter of the river Elbe at Hamburg and the river Mulde
at Dessau. Arch Environ Contam Toxicol 38:11–31.
6. Gandraß J, Bormann G, Wilken RD. 1995. N-/P-pesticides in the
Czech and German part of the river Elbe—Analytical methods
and trends of pollution. Fresensius J Anal Chem 353:70–74.
7. Pietsch J, Schmidt W, Sacher F, Fichtner S, Brauch HJ. 1995.
Pesticides and other organic micropollutants in the River Elbe.
Fresenius J Anal Chem 353:75–82.
8. Jantzen E, Prange A. 1995. Organometallic species of the elements tin, mercury, and lead in sediments of the longitudinal
profile of the River Elbe. Fresenius J Anal Chem 353:28–33.
9. Heininger P, Pelzer J, Claus E, Pfitzner S. 2003. Results of longterm sediment-quality studies on the river Elbe. Acta Hydrochim
Hydrobiol 31:356–367.
10. Grote M, Altenburger R, Brack W, Moschütz S, Mothes S, Michael C, Narten G-B, Paschke A, Schirmer K, Walter HA, Wennrich R, Wenzel K-D, Schüürmann G. 2005. Ecotoxicological profiling of transect river Elbe sediments. Acta Hydrochim Hydrobiol
3:555–569.
11. Vahl HH, Karbe L, Westendorf J. 1997. Genotoxicity assessment
of suspended particulate matter in the Elbe River: Comparison
of Salmonella microsome test, arabinose resistance test, and umutest. Mutat Res 394:81–93.
12. Arinç E, Sen A, Bozcaarmutlu A. 2000. Cytochrome P4501A and
associated mixed-function oxidase induction in fish as a biomarker for toxic carcinogenic pollutants in the aquatic environment. Pure Appl Chem 72:985–994.
13. Jedamski-Grymlas J, Kammann U, Tempelmann A, Karbe L, Siebers D. 1995. Biochemical responses and environmental contaminants in breams (Abramis brama L.) caught in the river Elbe.
Ecotoxicol Environ Saf 31:49–56.
14. Billiard S, Hahn ME, Franks DG, Peterson RE, Bols NC, Hodson
PV. 2002. Binding of polycyclic aromatic hydrocarbons (PAHs)
to teleost aryl hydrocarbon receptors (AhRs). Comp Biochem
Physiol B Biochem Mol Biol 133:55–68.
15. Ahlborg UG, Becking GC, Birnbaum LS, Brouwer A, Derks
HJGM, Feeley M, Golor G, Hanberg A, Larsen JC, Liem AKD,
Safe SH, Schlatter C, Waern F, Younes M, Yrjänheikki E. 1994.
Toxic equivalency factors for dioxin-like PCBs. Chemosphere 28:
1049–1067.
16. Safe S. 1986. Comparative toxicology and mechanism of action
of polychlorinated dibenzo-p-dioxins and dibenzofurans. Annu
Rev Pharmacol Toxicol 26:371–399.
17. Blankenship AL, Kannan K, Villalobos SA, Villeneuve DL, Falandysz J, Imagawa T, Jacobsson E, Giesy J. 2000. Relative potencies of individual polychlorinated naphthalenes and halowax
mixtures to induce Ah receptor–mediated responses. Environ Sci
Technol 34:3153–3158.
18. Götz R, Schumacher E, Kjeller LO, Bergqvist PA, Rappe C. 1990.
Polychlorinated dibenzo-para-dioxins and polychlorinated dibenzofurans in sediments and fish in the harbor of Hamburg. Chemosphere 20:51–73.
19. Stachel B, Christoph EH, Gotz R, Herrmann T, Kruger F, Kuhn
T, Lay J, Loffler J, Papke O, Reincke H, Schroter-Kermani C,
Schwartz R, Steeg E, Stehr D, Uhlig S, Umlauf G. 2006. Contamination of the alluvial plain, feeding-stuffs and foodstuffs with
polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans (PCDD/Fs), dioxin-like polychlorinated biphenyls (DLPCBs), and mercury from the river Elbe in the light of the flood
event in August 2002. Sci Total Environ 364:96–112.
20. Umlauf G, Bidoglio G, Christoph EH, Kampheus J, Kruger F,
Landmann D, Schulz AJ, Schwartz R, Severin K, Stachel B, Stehr
D. 2005. The situation of PCDD/Fs and dioxin-like PCBs after
the flooding of river Elbe and Mulde in 2002. Acta Hydrochim
Hydrobiol 33:543–554.
W. Brack et al.
21. Brack W, Kind T, Schrader S, Möder M, Schüürmann G. 2003.
Polychlorinated naphthalenes in sediments from the industrial
region of Bitterfeld. Environ Pollut 121:81–85.
22. Kannan K, Imagawa T, Blankenship AL, Giesy JP. 1998. Isomerspecific analysis and toxic evaluation of polychlorinated naphthalenes in soil, sediment, and biota collected near the site of a
former chlor-alkali plant. Environ Sci Technol 32:2507–2514.
23. Järnberg U, Asplund L, de Wit C, Egebäck AL, Wideqvist U,
Jakobsson E. 1997. Distribution of polychlorinated naphthalene
congeners in environmental and source-related samples. Arch Environ Contam Toxicol 32:232–245.
24. Brack W, Schirmer K, Kind T, Schrader S, Schüürmann G. 2002.
Effect-directed fractionation and identification of cytochrome
P4501A–inducing halogenated aromatic hydrocarbons in a contaminated sediment. Environ Toxicol Chem 21:2654–2662.
25. Brack W, Kind T, Hollert H, Schrader S, Möder M. 2003. A
sequential fractionation procedure for the identification of potentially cytochrome P4501A–inducing compounds. J Chromatogr
A 986:55–66.
26. Sanderson JT, Aarts JMMJG, Brouwer A, Froese KL, Denison
MS, Giesy JP. 1996. Comparison of Ah receptor–mediated luciferase and ethoxyresorufin-O-deethylase induction in H4IIE
cells: Implications for their use as bioanalytical tools for the detection of polyhalogenated aromatic hydrocarbons. Toxicol Appl
Pharmacol 137:316–325.
27. Giesy JP, Hilscherova K, Jones PD, Kannan K, Machala M. 2002.
Cell bioassays for detection of aryl hydrocarbon (AhR)– and
estrogen receptor (ER)–mediated activity in environmental samples. Mar Pollut Bull 45:3–16.
28. Hilscherova K, Kannan K, Kang YS, Holoubek I, Machala M,
Masunaga S, Nakanishi J, Giesy JP. 2001. Characterization of
dioxin-like activity of sediments from a Czech river basin. Environ Toxicol Chem 20:2768–2777.
29. Hubert A, Wenzel K-D, Manz M, Weissflog L, Engewald W,
Schüürmann G. 2000. High extraction efficiency for POPs in real
contaminated soil samples using accelerated solvent extraction.
Anal Chem 72:1294–1300.
30. Brack W, Altenburger R, Ensenbach U, Möder M, Segner H,
Schüürmann G. 1999. Bioassay-directed identification of organic
toxicants in river sediment in the industrial region of Bitterfeld
(Germany)—A contribution to hazard assessment. Arch Environ
Contam Toxicol 37:164–174.
31. Villeneuve DL, Blankenship AL, Giesy JP. 2000. Derivation and
application of relative potency estimates based on in vitro bioassay results. Environ Toxicol Chem 19:2835–2843.
32. Brack W, Segner H, Möder M, Schüürmann G. 2000. Fixed-effectlevel toxicity equivalents—A suitable parameter for assessing
ethoxyresorufin-O-deethylase induction potency in complex environmental samples. Environ Toxicol Chem 19:2493–2501.
33. Falandysz J. 1998. Polychlorinated naphthalenes: An environmental update. Environ Pollut 101:77–90.
34. Hanberg A, Wärn F, Asplund L, Haglund E, Safe S. 1990. Swedish
dioxin survey: Determination of 2,3,7,8-TCDD toxic equivalent
factors for some polychlorinated biphenyls and naphthalenes using biological tests. Chemosphere 20:1161–1164.
35. Hanberg A, Stahlberg M, Georgellis A, de Wit C, Ahlborg UG.
1991. Swedish dioxin survey: Evaluation of the H-4-II E bioassay
for screening environmental samples for dioxin-like enzyme induction. Pharmacol Toxicol 69:442–449.
36. Villeneuve DL, Kannan K, Khim JS, Falandysz J, Nikiforov VA,
Blankenship AL, Giesy JP. 2000. Relative potencies of individual
polychlorinated naphthalenes to induce dioxin-like responses in
fish and mammalian in vitro bioassays. Arch Environ Contam
Toxicol 39:273–281.
37. Puzyn T, Falandysz J, Jones PD, Giesy JP. 2007. Quantitative
structure–activity relationships for the prediction of relative in
vitro potencies (RePs) for chloronaphthalenes. J Environ Sci
Health A Tox Hazard Subs Environ Eng 42:573–590.
38. Wiedmann T, Ballschmiter K. 1993. Quantification of chlorinated
naphthalenes with GC-MS using the molar response of electronimpact ionization. Fresenius J Anal Chem 346:800–804.
39. Falandysz J, Strandberg L, Bergqvist PA, Kulp SE, Strandberg
B, Rappe C. 1996. Polychlorinated naphthalenes in sediment and
biota from the Gdansk basin, Baltic Sea. Environ Sci Technol
30:3266–3274.