Post on 27-Jan-2023
Seasonal and Spatial Distribution of SeveralEndocrine-Disrupting Compounds in the DouroRiver Estuary, Portugal
Claudia Ribeiro Æ Maria Elizabeth Tiritan ÆEduardo Rocha Æ Maria Joao Rocha
Received: 10 August 2007 / Accepted: 22 February 2008 / Published online: 27 March 2008
� Springer Science+Business Media, LLC 2008
Abstract Recent studies in the Douro River estuary show
signs of pollution in the area and of fish endocrine disruption.
However, the chemical nature of the local contamination has
not been fully investigated nor have studies checking for the
simultaneous presence of endocrine-disrupting chemicals
(EDCs), either of animal (estrone, E1; estradiol, E2),
pharmaceutical (17a-ethynylestradiol, EE2), vegetal
(daidzein, DAID; genistein, GEN; biochanin A, BIO-A), or
industrial (bisphenol A, BPA; 4–octylphenol, 4-OP; 4-no-
nylphenol, 4-NP) origins. Thus, the main objective of this
study was to examine the presence of these EDCs in estuarine
water samples collected, in every season of the year, at nine
sampling stations along the estuarine gradient. All samples
were processed by two-step solid-phase extraction (Oasis
HLB followed by silica) prior to high-performance liquid
chromatography with diode-array detection (HPLC-DAD)
and gas chromatography–mass spectrometry (GC-MS)
analyses. The current data showed that E1 and EE2, all
phytoestrogens, and BPA were identified and measured in
this estuary. In contrast, 4-OP was only detected by GC-MS
and E2 and 4-NP were not found. Additionally, E1 (up to
112.9 ng/L) and EE2 (up to 101.9 ng/L) were both measured
in biologically hazardous amounts in winter. In the year
sampled, the phytoestrogens suggested a possible seasonal
pattern of fluctuation. Both DAID (up to 888.4 ng/L) and
GEN (183.6 ng/L) were maximal in early summer, whereas
BIO-A (up to 191.4 ng/L) reached its highest concentrations
in winter. BPA (up to 10.7 lg/L) also attained highest
levels in winter. In December 2005, it is hypothesized that
E1, EE2, and BPA concentrations were atypically high due
to current drought conditions. Almost all assayed EDCs
existed in all seasons and, therefore, might have contributed
to endocrine disruption of aquatic animals, previously
documented by the high rate of ovotestis in fish caught in this
estuary.
Introduction
The Douro River flows along 900 km, from its source near
Duruelo de la Sierra (Spain) to its estuary at Porto
(Portugal). Due to its dimensions and huge watershed
(*98, 000 km2), this river passes throughout many differ-
ent regions that include densely inhabited areas, industrial
poles, and large agricultural fields, which, as a whole,
contribute to its pollution. This river has 51 large dams that
affect the drainage rates of the river to the Atlantic Ocean
(Vieira and Bordalo 2000). One dam, the Crestuma-Lever,
was built inside the innate estuary of the Douro River,
controlling the last 21.6 km of the river course and forming
a basin where the accumulation of pollutants is inevitable.
Thus, in addition to chemicals from upstream sources,
C. Ribeiro � M. E. Tiritan � M. J. Rocha (&)
Department of Pharmaceutical Sciences, Superior Institute of
Health Sciences (ISCS-N), Rua Central de Gandra, 1317,
4585-116 Gandra, PRD, Portugal
e-mail: mjsrocha@netcabo.pt
C. Ribeiro � E. Rocha � M. J. Rocha
Interdisciplinary Centre for Marine and Environmental Research
(CIIMAR), CIMAR Associate Laboratory, University of Porto
(UPorto), Porto, Portugal
C. Ribeiro � M. E. Tiritan
Centre of Studies of Organic Chemistry, Phytochemistry and
Pharmacology of Oporto University (CEQOFFUP), Porto,
Portugal
C. Ribeiro � E. Rocha
Institute of Biomedical Sciences Abel Salazar (ICBAS),
University of Porto (UPorto), Porto, Portugal
123
Arch Environ Contam Toxicol (2009) 56:1–11
DOI 10.1007/s00244-008-9158-x
contaminants leached from the untreated sewages and sew-
age treatment plants (SWTPs) located in the margins of this
highly industrialized and densely inhabited estuary (Vieira
and Bordalo 2000). In fact, recent studies involving the
measurement of pollutants such as insecticides (Ferreira
et al. 2002), inorganic metals (Mucha et al. 2004; Ramalhosa
et al. 2005), polycyclic aromatic hydrocarbon (PAHs)
(Ferreira et al. 2006), and bisphenol A (BPA) (Almeida et al.
2007) establish this estuary as extremely polluted. Estrogens
or estrogen mimics (phytoestrogens, alkylphenols, etc.),
which can contribute to diverse fish disorders (Mills and
Chichester 2005), were apparently never investigated in this
area. Those compounds, labeled herein as endocrine-dis-
rupting chemicals (EDCs), include several classes of natural
or pharmaceutical estrogens (17b-estradiol, estrone, and
ethynylestradiol), phytoestrogens (daidzein, genistein, and
biochanin A), and industrial pollutants (BPA, 4-octylphenol,
and 4-nonylphenol). Despite their different origins and dis-
tinct physical chemical properties, all of these EDCs could
cause endocrine disruption in fish (Kiparissis et al. 2003;
Mills and Chichester 2005). In fact, in vivo and in vitro
experiments have demonstrated that estrogens, either of
natural or pharmaceutical origins, and industrial pollutants
can exhibit additive or synergic toxic effects that are
responsible for the appearance of ovotestis in aquatic ani-
mals (Mills and Chichester 2005). Thus, because urban,
industrial, and agricultural pollution clearly exists in the
Douro River estuary and because studies in local fish, such as
the grey mullet (Mugil cephalus) (Ferreira et al. 2002),
suggest the presence of endocrine disruption in this area, the
main objectives of this study were to (1) identify and quantify
EDCs in Douro estuarine water, targeting natural and
pharmaceutical estrogens, phytoestrogens, BPA, and alkyl-
phenols, (2) determine possible seasonal variations for the
studied compounds, and (3) establish which compounds
were present in hazardous concentrations.
Materials and Methods
Sampling Area
The Douro River estuary is about 22 km long and is
located on the west coast of Portugal (41�080 N, 8�400 W).
Its average depth is about 8 m, with a natural semidiurnal
tidal range of 2–3 m (Bordalo and Vieira 2005). For this
study, nine sampling stations (S1 to S9) were selected from
the river outlet, near the Atlantic Ocean, to the Crestuma-
Lever dam (Fig. 1). Sampling stations S1, S3, S6, and S9
were located on the north bank of the river at the Porto city
margin, whereas S2, S4, S5, S7, and S8 were located at the
opposite side, bordering the other highly industrialized and
densely inhabited district, the Gaia city.
Sampling Collection
At the peak of winter (December 2005), spring (March
2006), summer (July 2006), and autumn (October 2006),
2 L of estuarine water samples were systematically col-
lected at a depth of 1 m, using a peristaltic sampler pump
(Global Water, Model WS300), along the estuarine gradi-
ent. Here, sampling occurred at both high and low tides of
the Douro River estuary and mean water temperatures
ranged from *7.0�C (winter) to 25.0�C (summer). Other
physicochemical parameters such as pH, salinity (%), and
conductivity are shown in Table 1. During sampling, all
bottles were rinsed two or three times before the collection
of the water samples, which were immediately filtrated, to
eliminate particulate matter and other suspended solids,
through a 47-mm GF/C glass fiber filter, acquired from
Millipore (Ireland). After this procedure, each filter was
washed several times with small amounts of CH3OH that
were added to the latter filtrate. Finally, all samples were
acidified with H2SO4 to pH 2 to prevent biodegradation and
kept at *5�C during transport to the laboratory.
Chemicals
Estrone (E1), 17b-estradiol (E2), ethynylestradiol (EE2),
daidzein (DAID), genistein (GEN), biochanin A (BIO-A),
bisphenol A (BPA), 4-octylphenol (4-OP), and the deriv-
atizing reagent N-methyl-N-trimethylsilyltrifluoroacetamide
(MSTFA) were purchased from Sigma-Aldrich (Steinhein,
Germany), whereas 4-nonylphenol (4-NP) was purchased
from Riedel-de-Haen (Seelze-Hannover, Germany). Stock
solutions of individual standards were prepared by dissolv-
ing known amounts of each compound in CH3OH:CH3CN
Fig. 1 Map of the Douro River estuary showing the locations of the
nine sampling areas (S1 to S9) selected for this study, the main
industrial poles, and sewage treatment plants of this area
2 Arch Environ Contam Toxicol (2009) 56:1–11
123
(50:50, v/v) high-performance liquid chromatography
(HPLC) grade, acquired from Sigma-Aldrich (Steinhein,
Germany), to obtain final concentrations of 500 mg/L. All
other solvents were analytical grade from Sigma-Aldrich
(Steinhein, Germany). Ultrapure water was supplied by a
Milli-Q water system.
Sample Preparation
All water samples, previously filtered and acidified, were
preconcentrated by a solid extraction phase (SPE) proce-
dure using a 500-mg Oasis HLB Cartridge (polymer of
N-vinylpyrrolidone and divinylbenzene) purchased from
Waters Corporation (Milford, MA, USA). Prior to use, this
cartridge was sequentially washed with 25 mL of
CH2Cl2:CH3OH (50:50, v/v), 12 mL of CH3OH, and
25 mL of ultrapure Milli-Q water (Ribeiro et al. 2007).
Samples were applied to cartridges under vacuum to obtain
a constant flow rate of 5–7 mL/min. Cartridges were
washed with 25 mL of ultrapure Milli-Q water, followed
by 1 mL of CH3OH and eluates discarded. Cartridges were
dried kept under vacuum aspiration for 30 min and eluted
with 20 mL of CH2Cl2:CH3OH (50:50, v/v). The dark and
sticky eluate was further ‘‘purified’’ on a prewashed 1-g
Sep-Pak silica cartridge (CH2Cl2:CH3OH 50:50, v/v; from
Waters Corporation, Milford, MA, USA). Seven milliliters
of the same solvent composition were used to elute samples
into a round-bottomed tube, which was taken to dryness
under N2 in a thermostatic bath at 40�C. Samples were
resuspended in 200 lL of CH3OH:CH3CN (50:50, v/v).
Twenty microliters of each sample were injected in tripli-
cate into the high-performance liquid chromatography with
diode-array detection (HPLC-DAD) system for quantita-
tive analysis (Ribeiro et al. 2007) and evaporated to
dryness and derivatizated with 50 lL of MSTFA before
injection in the gas chromatography–mass spectrometry
Table 1 Ranges of the physicochemical parameters measured in each of the sampling sites (S1 to S9) of the Douro River estuary
Parameters Sampling station Winter December 2005 Spring March 2006 Summer July 2006 Autumn October 2006
Low tide High tide Low tide High tide Low tide High tide Low tide High tide
pH S1 7.5 7.6 7.3 7.5 8.3 7.9 7.9 8.2
S2 7.9 7.9 7.4 7.5 7.9 7.7 7.8 8.1
S3 7.5 7.7 7.3 7.7 8.3 7.3 8.0 8.0
S4 7.7 7.8 7.7 7.6 8.2 7.9 8.0 8.2
S5 7.7 7.4 7.6 7.5 8.0 7.8 7.8 8.1
S6 7.8 7.8 7.5 7.3 7.9 7.9 8.0 8.1
S7 7.5 7.4 7.5 7.6 7.8 8.0 7.9 8.0
S8 7.5 7.4 7.6 7.4 7.6 7.9 7.8 7.8
S9 7.5 7.5 7.7 7.5 7.8 7.7 7.9 8.0
Salinity (%) S1 4.0 12.7 0.0 0.7 6.1 8.9 20.6 31.1
S2 5.8 6.0 0.0 0.0 7.1 8.5 10.0 30.3
S3 1.2 7.9 0.0 0.0 5.9 11.1 15.4 31.0
S4 8.1 10.9 0.0 0.0 6.1 11.1 13.5 31.4
S5 10.4 11.9 0.0 0.0 4.7 6.2 7.9 23.7
S6 2.0 2.3 0.0 0.0 1.8 1.9 3.4 14.5
S7 0.0 0.0 0.0 0.0 2.2 3.1 7.4 12.1
S8 0.0 0.0 0.0 0.0 0.0 0.0 1.3 1.5
S9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Conductivity (mS/cm) S1 7.3 21.2 0.8 1.1 10.8 15.3 33.0 47.7
S2 10.2 10.6 0.2 0.3 10.4 18.8 17.0 46.7
S3 1.9 13.7 0.2 0.8 12.5 14.6 25.3 47.6
S4 13.9 18.4 0.2 0.9 10.8 18.7 22.4 48.2
S5 17.7 19.9 0.2 0.2 8.5 11.0 13.7 37.4
S6 3.1 3.5 0.2 0.2 2.8 2.9 6.3 24.0
S7 0.2 0.2 0.2 0.2 4.2 5.7 7.4 12.1
S8 0.1 0.1 0.1 0.2 0.4 0.4 2.0 2.3
S9 0.1 0.3 0.2 0.2 0.6 0.6 0.5 0.9
Arch Environ Contam Toxicol (2009) 56:1–11 3
123
(GC-MS) equipment. Recovery assays, for the SPE steps,
demonstrated that all compounds had recoveries higher
than 69%, except 4-NP (53%) (Table 2).
Instrumental and Main Methodological Characteristics
The HPLC system consisted of a LiChroCART C18
reversed-phase analytical column 250 9 4-mm-inner
diameter, 5-lm particle size (Merck, Darmstadt, Germany)
and a Merck Hitachi HPLC apparatus, equipped with the
LaChrom diode array detector L-7455. Data acquisition was
performed by a HPLC System Manager HSM D-7000,
Version 3.0 (Merck-Hitachi). All analyses in this system
followed a previously validated method developed for the
present estuarine water samples (Ribeiro et al. 2007).
Briefly, the method was as follows: mobile-phase compo-
sition CH3CN:H2O (25:75, v/v) acidified with CF3CO2H,
pH 2, 1 mL/min with organic solvent increasing linearly, at
room temperature, as indicated in Table 3. After each
chromatographic run, the amount of CH3CN increased up to
100% and was maintained isocratically for 5 min before a
new injection. During the application of this method, sev-
eral quality control parameters were performed taking in
consideration the International Conference on Harmoniza-
tion (1996) rules; that is, the robustness of the method was
constantly checked. For that, the instrumental precision and
accuracy were monitored regularly by the measurement of
the peak areas of injections containing both standard mix-
ture and fortified matrix (relative standard deviation
(RSD) \ 1%). Furthermore, the maintenance of the reten-
tion times among samples and standards was always
verified as each analyzed sample was injected both indi-
vidually (n = 2) and spiked with all standards (n = 1).
Finally, only the areas of peaks showing a purity test higher
than 99% (this value was calculated automatically by the
HPLC-DAD software) were considered for the quantitative
analysis, which was based on calibration curves created for
this propose using a Douro River estuarine matrix spiked
with standards of all EDCs (Table 2).
The presence of compounds of interest in the samples
was confirmed by GC-MS on derivatized samples. The GC
system consisted in a Varian CP 3800 apparatus equipped
with a VF-5ms-type capillary column (30 m 9 0.25 mm
inner diameter; df: 0.25 lm) connected to an ion trap MS
(Varian Saturn 2200). The GC-MS analytical procedure
used for the current estuarine water samples was based in
several published protocols that were adjusted to the cur-
rent samples (Ballesteros et al. 2006; Lee et al. 2004;
Shareef et al. 2006; Silveira 2007). As the majority of the
EDCs analyzed (Fig. 2) contain several hydroxyl groups,
MSTFA was the preferred derivatization reagent used to
prepare the trimethylsilyl (TMS) derivatives (Lee et al.
2004; Shareef et al. 2006). Data acquisition was obtained
using the selected ion monitoring (SIM) mode. The iden-
tification of each chromatographic peak was achieved by
comparing the retention times with coinjected standards
and matching the characteristic ions of standards and
samples (Table 4).
Results
Data are displayed in Figs. 3 and 4 and in Tables 1, 2, and
4–7. In Table 5, it is shown that E1 and EE2 were found in
Table 2 Chromatographic data and calibration results obtained by the standard addition technique
Chemicals
(EDCs)
Wavelength
(nm)
Retention
times
(tM, min)
Linear
dynamic
range (lg/mL)
Intercept Slope R2 Recovery
(%)
LOD
(ng/L)
LOQ
(ng/L)
DAID 246 8.10 0.40–20.0 5,172.3 57,460 0.998 104 10.0 31.7
GEN 260 10.60 0.40–20.0 9,622.9 52,857 0.999 100 3.2 9.8
BPA 278 12.60 0.90–135.0 1,723.5 7,817.9 0.998 99 8.0 24.5
E2 280 13.60 0.80–20.0 1,236.3 3,530.1 0.999 87 7.0 21.3
EE2 280 14.40 1.00–10.0 –152.7 4,056.8 0.999 108 18.0 54.5
E1 280 15.20 0.80–10.0 1,497.3 3,398.0 0.992 116 15.0 44.0
BIO-A 260 15.70 0.12–10.0 –403.9 36,004 0.998 92 12.4 37.5
4-OP 278 25.90 2.00–140.0 30.1 5,340.9 0.997 69 3.8 12.0
4-NP 278 27.80 2.00–140.0 -53.6 4,598.0 0.999 53 7.0 21.8
LOD = limit of detection; LOQ = limit of quantificaion
Table 3 Organic gradient for the chromatographic separation of nine
EDCs by HPLC-DAD
Time (min) Organic solvent (%)
0 25
5 40
14 55
17 57
30 90
36 100
4 Arch Environ Contam Toxicol (2009) 56:1–11
123
the Douro River estuarine waters. These chemicals, quan-
tified only in winter (Fig. 3A), showed their highest levels
at the sampling point S4 (up to 112.9 ng/L in low tide) for
E1 and at the sampling station S5 (up to 101.9 ng/L in high
tide) for EE2. Later, during all other analyzed seasons, E1
and EE2 became undetectable by HPLC-DAD (Table 2),
whereas their presence were confirmed by GC-MS, as
demonstrated by the retention time and fragmentation
patterns characteristic of both E1 and EE2 shown in
Table 4 and Fig. 4A and 4B. In contrast, the presence of E2
was not detected on any occasion either by HPLC-DAD or
GC-MS. The measurement, by HPLC-DAD, of both E1
and EE2 in winter can be related with the absence of
pluviosity and low temperatures registered in this period.
This observation is also supported by the values of salinity
and conductivity found in winter (Table 1).
Table 6 shows the concentrations of all phytoestrogens
measured. Only BIO-A was quantifiable in all seasons
(Fig. 3A–3C). Therefore, in every season of the year, the
highest levels of BIO-A were measured at the following
sampling stations: S1 (up to 112.1 ng/L in low tide) in
winter, S7 (up to 191.4 ng/mL in low tide) in spring, S3 (up
Fig. 2 Chemical structure of all
EDCs investigated in the Douro
River estuary
Table 4 Ions and fragment ratios used for identification of the nine proposed analytes by GC-MS
EDCs Main
sourceaMass weight
(g/mol)
Retention
times (min)
Ions selected in
standard mixtures (m/z)
(% relative abundance)
4-OP Industrial and municipal effluents 206 12.0 278 (30), 179 (100)
4-NP Industrial and municipal effluents 220 13.1 292 (34), 179 (100)
BPA Industrial effluents 228 15.7 372 (6), 357 (100)
E1 Urban and municipal effluents 270 19.7 342 (100), 257 (45)
E2 Urban and municipal effluents 272 19.9 416 (100), 285 (75)
EE2 Municipal effluents 296 21.4 425 (100), 285 (40)
BIO-A Agricultural runoff 284 22.4 428 (8), 413 (100)
DAID Agricultural runoff 254 23.4 398 (100), 383 (85)
GEN Agricultural runoff 270 23.7 471 (1), 399 (100)
a Data from Lintelmann et al. (2003) and Lagana et al. (2004)
Arch Environ Contam Toxicol (2009) 56:1–11 5
123
to 23.1 ng/L in high tide) in summer, and at S3 (87.3 ng/L
in low tide) in autumn. In contrast, the levels of DAID and
GEN were very low or undetectable by HPLC-DAD
(Table 2) in winter, spring, and autumn but showed note-
worthy concentrations in summer (Table 6). Thus, in July
(Fig. 3C), the highest levels of DAID were measured at the
sampling station S9 (up to 888.4 ng/L in high tide) and
those of GEN at the sampling station S5 (up to 197.4 ng/L
in high tide). Here, it is important to stress that the identity
of all phytoestrogens referred in this study were all con-
firmed by GC-MS, as is shown in Fig. 4C–4E. Thus, the
retention times and fragmentation patterns characteristic of
DAID, GEN, and BIO-A (Table 4) were found in all ana-
lyzed samples.
In Table 7, it is demonstrated that BPA was the main
industrial pollutant measured in our study, as 4-OP was
under the detection limits of the HPLC-DAD method and
4-NP was never detected. Therefore, BPA was measured in
all seasons at the sampling points S1 and S3 and S5 and S7
located near the industrial poles and sewage treatment
plants of the Porto and the Gaia city, respectively (Fig. 1).
Mostly in summer and autumn, BPA levels became, in
several sampling stations, under the detection limits of the
current HPLC-DAD method (Table 2). Nonetheless, the
current GC-MS analysis demonstrated that BPA and 4-OP
were always present in this estuary, whereas 4-NP was not.
In Table 4 and Fig. 4F and 4G are shown the fragmentation
peaks characteristic of the BPA and 4-OP.
Discussion
The data of this study are the first of their kind advanced
for the estuary of the Spanish–Portuguese Douro River.
Fig. 3 Chromatograms of several the samples: (A) S5, low tide,
winter (December 2005), (B) S1, low tide, spring (March 2006), (C)
S9, high tide, summer (July 2006), and (D) S4, low tide, autumn
(October 2006) spiked with the standard mixture containing 0.10 lg/L
of DAID and GEN, 0.23 lg/L of BPA, 0.20 lg/L of E2 and E1,
0.25 lg/L of EE2, 0.03 lg/L of BIO-A, and 0.5 lg/L of 4-OP and
4-NP
6 Arch Environ Contam Toxicol (2009) 56:1–11
123
Our study showed that the water contained varying
amounts of natural estrogens and of xenoestrogens, which,
in accordance with recent bibliographic data, at times
reached deleterious amounts of EDCs (Mills and Chich-
ester 2005). Estrogens of natural and pharmaceutical origin
found in winter samples were similar to those reported in
Japanese wastewaters (Komori et al. 2004). Such values
(Table 5), commonly accepted as hazardous for the aquatic
fauna (Metcalfe et al. 2001; Segner et al. 2003), were
extremely high when compared with other European
estuaries (Belfroid et al. 1999; Noppe et al. 2007; Vethaak
et al. 2005). Nonetheless, we believe that the estrogenic
load might be atypical because in 2005 there occurred the
most severe regional drought in the last 60 years in this
region, and on such occasions, the drainage rates of the
river to the Atlantic Ocean are very low, ranging from 0 to
13,000 m3/s (Bordalo et al. 2006; Vieira and Bordalo
2000). This hypothesis is also supported by the data
obtained later for all other sampling seasons, when the
levels of both E1 and EE2 became undetectable by HPLC-
DAD (Table 2). It is important to stress that the precipi-
tation levels in this area returned to normal later
Fig. 4 MS spectra of all
detected compounds in the
Douro River estuary. The
arrows point to the peaks (m/z)
used for the identification of E1
(A), EE2 (B), DAID (C), GEN
(D), BIO-A (E), BPA (F), and
4-OP (G)
Arch Environ Contam Toxicol (2009) 56:1–11 7
123
Table 5 Concentrations of E1 and EE2 in each of the sampling sites (S1 to S9) of the Douro River estuary
Estrogens (ng/L) Sampling station Winter December 2005 Spring March 2006 Summer July 2006 Autumn October 2006
Low tide High tide Low tide High tide Low tide High tide Low tide High tide
E1 S1–S3 \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a
S4 112.9 \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a
S5 99.8 \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a
S6–S7 \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a
S8 103.9 \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a
S9 \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a \15.0a
EE2 S1 \18.0a 56.0 \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a
S2–S3 \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a
S4 97.7 \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a
S5 \18.0a 101.9 \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a
S6 83.1 62.3 \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a
S7 \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a
S8–S9 \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a \18.0a
a Less than detection limit
Table 6 Concentrations of DAID, GEN, and BIO-A in each of the sampling sites (S1 to S9) of the Douro River estuary
Phytoestrogens (ng/L) Sampling station Winter December 2005 Spring March 2006 Summer July 2006 Autumn October 2006
Low tide High tide Low tide High tide Low tide High tide Low tide High tide
DAID S1 \10.0a \10.0a \10.0a \10.0a 231.3 495.9 \10.0a \10.0a
S2 \10.0a \10.0a \10.0a \10.0a 597.4 205.3 \10.0a \10.0a
S3–S4 \10.0a \10.0a \10.0a \10.0a \10.0a \10.0a \10.0a \10.0a
S5 25.9 \10.0a \10.0a \10.0a 245.1 603.5 \10.0a \10.0a
S6 \10.0a 14.6 \10.0a \10.0a \10.0a \10.0a \10.0a \10.0a
S7 \10.0a \10.0a \10.0a \10.0a \10.0a \10.0a \10.0a \10.0a
S8 \10.0a \10.0a \10.0a \10.0a 202.1 292.5 \10.0a \10.0a
S9 \10.0a \10.0a \10.0a \10.0a 145.3 888.4 \10.0a \10.0a
GEN S1 \3.2a \3.2a \3.2a \3.2a \3.2a 72.4 \3.2a \3.2a
S2 \3.2a \3.2a \3.2a \3.2a 183.6 83.9 \3.2a \3.2a
S3 \3.2a \3.2a \3.2a \3.2a \3.2a 177.3 \3.2a \3.2a
S4 \3.2a \3.2a \3.2a \3.2a \3.2a \3.2a \3.2a \3.2a
S5 \3.2a \3.2a \3.2a \3.2a \3.2a 197.4 \3.2a \3.2a
S6–S7 \3.2a \3.2a \3.2a \3.2a \3.2a \3.2a \3.2a \3.2a
S8 \3.2a \3.2a \3.2a \3.2a 48.4 90.5 \3.2a \3.2a
S9 \3.2a \3.2a \3.2a \3.2a 29.1 122.8 \3.2a \3.2a
BIO-A S1 112.1 57.1 134.4 109.6 \12.4a 18.7 61.4 17.1
S2 \12.4a \12.4a 124.2 50.9 \12.4a \12.4a \12.4a \12.4a
S3 \12.4a 76.1 108.8 107.9 \12.4a 23.1 87.3 \12.4a
S4 94.0 \12.4a 93.5 95.9 \12.4a \12.4a 43.2 \12.4a
S5 78.0 75.1 49.6 72.0 \12.4a \12.4a 28.9 \12.4a
S6 \12.4a \12.4a 46.8 104.0 \12.4a \12.4a 38.9 37.1
S7 52.2 63.1 191.4 82.1 \12.4a \12.4a 56.2 31.2
S8 38.7 41.2 90.1 52.5 \12.4a \12.4a 24.1 17.9
S9 56.2 79.3 60.6 122.6 \12.4a \12.4a 33.3 \12.4a
a Less than detection limit
8 Arch Environ Contam Toxicol (2009) 56:1–11
123
(Cararmelo and Orgaz 2007), causing a visible increase of
the caudal of this estuary in spring and a significant
decrease in the levels of the salinity and conductivity
during this period (Table 1). These parameters, closely
related with pluviosity (Ellis et al. 1977), indicate an
obvious dilution effect of all local pollutants. However,
because we demonstrated, by GC-MS, the continuous
presence of E1 and EE2 in all seasons, their potential
deleterious effects cannot be ruled out, as both estrogens
might produce endocrine disruption and reproductive dis-
turbances even in very low nanograms per liter ranges
(Metcalfe et al. 2001; Segner et al. 2003). These results
are in accordance with recent studies that classified the
Douro estuary as the second most polluted of Portugal
(Vasconcelos et al. 2007), with documented intersex in
local estuarine fish (Ferreira et al. 2002). Thus, the present
data confirm the need of regular local monitoring studies
for evaluating natural and pharmaceutical estrogens, as it is
well known that these chemicals induce vitellogenesis,
intersex, and feminization in male fish and affects human
reproduction (Waring and Harris 2005). In contrast, E2 was
never detected in these estuarine waters, possibly due to
rapid degradation of E2 to E1 and/or because humans
excrete more E1 than E2 in their urine (Johnson et al.
2000).
Phytoestrogens are about 10-2- to 10-3-fold less potent
than natural and pharmaceutical estrogens (Benassayag
et al. 2002; Gutendorf and Westendorf 2001). Nonetheless,
because they are still able to induce endocrine disruption in
fish (Kiparissis et al. 2003) and DAID, GEN, and BIO-A
were never measured in the Douro River, this study reports
the levels of these compounds in this estuary. Thus, in
winter, spring, and autumn, occasions when both agricul-
tural activities are at the resting stage, both DAID and GEN,
which are among the most important components of fruits,
cereals, and vegetables (Liggins et al. 2000a, b, 2002), were
undetectable or showed very low levels (Table 6). In con-
trast, the concentrations of BIO-A were highest in winter
and early spring, which is consistent with the seasonal
variation of the indigenous flora of this estuary. In fact, this
estuary not only contains considerable amounts of red clo-
ver (Trifolium pratense L., Fabaceae) and papilonaceous
plants (Papilionaceae) but also holds many species of
grasses that are rich sources of BIO-A at that time of the
year (Booth et al. 2006), which, by leaching, can easily
reach the estuarine waters. Later, in early summer, this
depiction reversed and, although the concentrations of both
DAID and GEN were particularly high, those of BIO-A
become minimal. These data are consistent with both the
vast vineyards and orchards (mainly of apple, orange, and
almond) located in both margins of this river. In autumn, the
seasonal cycle of the studied phytoestrogens seemed to be
completed with the rise of BIO-A concentrations and the
remarkable decrease of both DAID and GEN.
Because the concentrations of all measured phytoestro-
gens were below the hazardous levels described in the
in vitro experiments conducted by Kiparissis et al. (2003)
and because the last study demonstrated that only con-
centrations above 1000 lg/L GEN produce gonadal
intersex in the Japanese medaka (Oryzias latipes), our
current data suggest that, individually, all measured phy-
toestrogens might not exert any adverse effect in the local
fauna. However, further biological studies are needed in
this estuary to confirm if the total load of all these com-
pounds have potential for synergistic activity, either with
other phytoestrogens or other estrogenic chemicals able to
produce endocrine disruption in aquatic species.
As for industrial pollution, the Douro estuary is partic-
ularly affected by discharges of debris from plating, soap,
textile, and tannery industries located near its margins or in
its main tributaries (Fig. 1). Extremely high levels of BPA
(Table 7) were found in winter, possibly due to the drought
Table 7 Concentrations of BPA measured in each sampling sites (S1 to S9) of the Douro River estuary
Industrial pollutants (lg/L) Sampling station Winter December 2005 Spring March 2006 Summer July 2006 Autumn October 2006
Low tide High tide Low tide High tide Low tide High tide Low tide High tide
BPA S1 10.7 5.5 6.6 7.0 1.0 4.2 1.7 \0.08a
S2 \0.08a \0.08a 6.3 2.5 \0.08a 1.3 \0.08a \0.08a
S3 0.25 9.0 5.2 5.5 0.7 3.1 2.5 \0.08a
S4 \0.08a \0.08a 4.8 5.0 \0.08a \0.08a 1.1 \0.08a
S5 9.9 7.8 1.9 4.6 \0.08a 1.6 0.9 \0.08a
S6 \0.08a \0.08a 0.1 3.9 \0.08a \0.08a 0.9 \0.08a
S7 5.5 7.6 0.3 4.8 0.7 2.3 0.7 \0.08a
S8 4.3 4.7 5.1 0.7 0.7 2.1 \0.08a \0.08a
S9 6.2 7.6 3.3 4.7 \0.08a 1.7 \0.08a \0.08a
a Less than detection limit
Arch Environ Contam Toxicol (2009) 56:1–11 9
123
conditions of December 2005. In fact, later the levels of
BPA decreased, becoming similar to those reported in other
polluted rivers either in Portugal (Azevedo et al. 2001;
Quiros et al. 2005), Spain (Cespedes et al. 2005), or The
Netherlands (Vethaak et al. 2005). The permanent presence
of BPA in the sampling points S1 and S3, and S5 and S7,
located near highly urbanized and industrial poles (Fig. 1),
confirm the impact of the human activities in both margins
of this estuary. Relative to the possible hazardous effects of
BPA, the concentration in the Douro estuary were lower
than those reported for endocrine disruption in several
studies (i.e., lower than 160 lg/L) (Mills and Chichester
2005; Staples et al. 2002). Nonetheless, we think that fur-
ther studies should be conducted with water from this
estuary in order to investigate the possible ‘‘cocktail
effect’’ (i.e., total estrogenic effect) promoted by the mix-
ture of all EDCs detected and/or quantified in this area.
Finally, no typical gradient and influence of tides were
found along this estuary for the assayed EDCs. These
observations were probably due to the artificial hydrody-
namics of this estuary, a result of the Crestuma-Lever dam
as suggested by Bordalo et al. (2006) in studies concerning
the water quality of this estuary. However, the combination
of drought and diminished discharges of the dam, like those
registered in December 2005, warns that in the Douro
River estuary, moderate or low concentrations of EDCs
might suddenly become extremely high and then be
potentially risky to wildlife.
Conclusions
Our data (1) showed the presence of natural and pharma-
ceutical estrogens, phytoestrogens, and selected industrial
pollutants in the Douro River estuarine waters, (2) dem-
onstrated the occurrence of hazardous amounts of E1 and
EE2 in December 2005 as well as their continuous pres-
ence in this estuary, (3) reported, for the first time, the
environmental concentrations of DAID, GEN, and BIO-A
in the Douro estuary, (4) suggested a possible seasonal
fluctuation trend for the studied phytoestrogens, (5)
revealed that BPA was a significant pollutant of this estu-
ary, whereas 4-OP and 4-NP were not (6) supported the
need for further monitoring studies targeting estrogenic
chemicals, and (7) warned that biologic studies are needed
to evaluate possible endocrine disruptive ‘‘cocktail effects’’
of the EDCs in the local fauna.
Acknowledgments This study was financially supported by the
Fundacao para a Ciencia e Tecnologia (FCT; PhD Grant SFRH/BD/
18231/2004/SLU, attributed to Claudia Ribeiro) and by the Cooper-
ativa de Ensino Superior, Politecnico e Universitario, CRL (CESPU)/
ISCS-N (Research Projects 1F/13/2005/CESPU and 2F/02/2006/
CESPU).
References
Almeida C, Serodio P, Florencio MH, Nogueira JMF (2007) New
strategies to screen for endocrine-disrupting chemicals in the
Portuguese marine environment utilizing large volume injection-
capillary gas chromatography-mass spectrometry combined with
retention time locking libraries (LVI-GC-MS-RTL). Anal Bio-
anal Chem 387:2569–2583
Azevedo DA, Lacorte S, Viana P, Barcelo D (2001) Analysis of
priority pesticides and phenols in Portuguese river water by
liquid chromatography-mass spectrometry. Chromatographia
53:113–118
Ballesteros O, Zafra A, Navalon A, Vilchez JL (2006) Sensitive gas
chromatographic-mass spectrometric method for the determina-
tion of phthalate esters, alkylphenols, bisphenol A and their
chlorinated derivatives in wastewater samples. J. Chromatogr A
1121:154–162
Belfroid AC, Van der Horst A, Vethaak AD et al (1999) Analysis and
occurrence of estrogenic hormones and their glucuronides in
surface water and waste water in The Netherlands. Sci Total
Environ 225:101–108
Benassayag C, Perrot-Applanat M, Ferre F (2002) Phytoestrogens as
modulators of steroid action in target cells. J Chromatogr B
777:233–248
Booth NL, Overk CR, Yao P et al (2006) Seasonal variation of red
clover (Trifolium pratense L., Fabaceae) isoflavones and estro-
genic activity. J Agric Food Chem 54:1277–1282
Bordalo AA, Teixeira R, Wiebe WJ (2006) A water quality index
applied to an international shared river basin: the case of the
Douro river. Environ Manage 38:910–920
Bordalo AA, Vieira MEC (2005) Spatial variability of phytoplankton,
bacteria and viruses in the mesotidal salt wedge Douro Estuary
(Portugal). Est Coast Shelf Sci 63:143–154
Cararmelo L, Orgaz MDM (2007) A study of precipitation variability
in the Duero Basin (Iberian Peninsula). Int J Climatol 27:327–339
Cespedes R, Lacorte S, Raldua D, Ginebreda A, Barcelo D, Pina B
(2005) Distribution of endocrine disruptors in the Llobregat
River basin (Catalonia, NE Spain). Chemosphere 61:1710–1719
Ellis J, Kanamori S, Laird PG (1977) Water pollution studies on Lake
Illawarra. I. Salinity variations and estimation of residence time.
Marine Freshwater Res 28:467–477
Ferreira M, Antunes P, Gil O, Vale C, Silva L, Reis-Henriques MA
(2002) Testis-ova in mullet (Mugil cephalus) exposed to organic
contaminants in the Douro Estuary, Portugal. In: Proceedings of
the 21th conference of European comparative endocrinologists,
Germany
Ferreira M, Moradas-Ferreira P, Reis-Henriques MA (2006) The
effect of long-term depuration on phase I and phase II
biotransformation in mullets (Mugil cephalus) chronically
exposed to pollutants in River Douro estuary, Portugal. Marine
Environ Res 61:326–338
Gutendorf B, Westendorf J (2001) Comparison of an array of in vitroassays for the assessment of the estrogenic potential of natural
and synthetic estrogens, phytoestrogens and xenoestrogens.
Toxicology 166:79–89
International Conference on Harmonisation (ICH) (1996) Validation of
analytical procedures: methodology, Q2B (CPMP/ICH/281/96)
Johnson AC, Belfroid A, Di Corcia A (2000) Estimating steroid
oestrogen inputs into activated sludge treatment works and
observations on their removal from the effluent. Sci Total
Environ 256:163–173
Kiparissis Y, Balch GC, Metcalfe TL , Metcalfe CD (2003) Effects of
the isoflavones genistein and equol on the gonadal development
of Japanese medaka (Oryzias latipes). Environ Health Perspect
111:1158–1163
10 Arch Environ Contam Toxicol (2009) 56:1–11
123
Komori K, Tanaka H, Okayasu Y, Yasojima M, Sato C (2004)
Analysis and occurrence of estrogen in wastewater in Japan.
Water Sci Technol 50:93–100
Lagana A, Bacaloni A, De Leva I, Faberi A, Fago G, Marino A (2004)
Analytical methodologies for determining the occurence of
endocrine disrupting chemicals in sewage treatment plants and
natural waters. Anal Chim Acta 501:79–88
Lee SH, Jung BH, Kim SY, Chung BC (2004) Determination of
phytoestrogens in traditional medicinal herbs using gas chroma-
tography-mass spectrometry. J Nutr Biochem 15:452–460
Liggins J, Bluck LJC, Runswick S, Atkinson C, Coward WA,
Bingham SA (2000a) Daidzein and genistein content of fruits
and nuts. J Nutr Biochem 11:326–331
Liggins J, Bluck LJC, Runswick S, Atkinson C, Coward WA,
Bingham SA (2000b) Daidzein and genistein contents of
vegetables. Br J Nutr 84:717–725
Liggins J, Mulligan A, Runswick S, Bingham SA (2002) Daidzein
and genistein content of cereals. Eur J Clin Nutr 56:961–966
Lintelmann J, Katayama A, Kurihara N, Shore L, Wenzel A (2003)
Endocrine disruptors in the environment (IUPAC Technical
Report). Pure Appl Chem 75:631–681
Metcalfe CD, Metcalfe TL, Kiparissis Y et al (2001) Estrogenic
potency of chemicals detected in sewage treatment plant
effluents as determined by in vivo assays with Japanese medaka
(Oryzias latipes). Environ Toxicol Chem 20:297–308
Mills LJ, Chichester C (2005) Review of evidence: are endocrine-
disrupting chemicals in the aquatic environment impacting fish
populations? Sci Total Environ 343:1–34
Mucha AP, Bordalo AA, Vasconcelos M (2004) Sediment quality in
the Douro river estuary based on trace metal contents, macro-
benthic community and elutriate sediment toxicity test (ESTT). J
Environ Monit 6:585–592
Noppe H, Verslycke T, De Wulf E et al (2007) Occurrence of
estrogens in the Scheldt estuary: a 2-year survey. Ecotoxicol
Environ Safety 66:1–8
Quiros L, Cespedes R, Lacorte S et al (2005) Detection and
evaluation of endocrine-disruption activity in water samples
from Portuguese rivers. Environ Toxicol Chem 24:389–395
Ramalhosa E, Pereira E, Vale C, Valega M, Monterroso P, Duarte AC
(2005) Mercury distribution in Douro estuary (Portugal). Marine
Pollut Bull 50:1218–1222
Ribeiro C, Tiritan ME, Rocha E, Rocha MJ (2007) Development and
validation of a HPLC-DAD method for determination of several
endocrine disrupting compounds in estuarine water. J Liq
Chromatogr Relat Technol 30:2729–2746
Segner H, Navas JM, Schafers C, Wenzel A (2003) Potencies of
estrogenic compounds in in vitro screening assays and in life
cycle tests with zebrafish in vivo. Ecotoxicol Environ Safety
54:315–322
Shareef A, Angove MJ, Wells JD (2006) Optimization of silylation
using N-methyl-N-(trimethylsilyl)-trifluoroacetamide, N,O-bis-
(trimethyl.silyl)-trifluoroacetamide and N-(tert-butyldimethylsi-
lyl)-N-methyltrifluoroacetamide for the determination of the
estrogens estrone and 17 alpha-ethinylestradiol by gas chroma-
tography-mass spectrometry. J Chromatogr A 1108:121–128
Silveira M (2007) Identification of endocrine disruptors in superficial
waters of Douro River estuary, University of Porto, Portugal
Staples CA, Woodburn K, Caspers N, Hall AT, Klecka GM (2002) A
weight of evidence approach to the aquatic hazard assessment of
bisphenol A. Hum Ecol Risk Assess 8:1083–1105
Vasconcelos RP, Reis-Santos P, Fonseca V et al (2007) Assessing
anthropogenic pressures on estuarine fish nurseries along the
Portuguese coast: a multi-metric index and conceptual approach.
Sci Total Environ 374:199–215
Vethaak AD, Lahr J, Schrap SM et al (2005) An integrated
assessment of estrogenic contamination and biological effects
in the aquatic environment of The Netherlands. Chemosphere
59:511–524
Vieira MEC, Bordalo AA (2000) The Douro estuary (Portugal): a
mesotidal salt wedge. Oceanol Acta 23:585–594
Waring RH, Harris RM (2005) Endocrine disrupters: a human risk?
Mol Cell Endocrinol 244:2–9
Arch Environ Contam Toxicol (2009) 56:1–11 11
123