at SciVerse ScienceDirect

Environmental Pollution 178 (2013) 220e228

Contents lists available

Environmental Pollution

journal homepage: www.elsevier .com/locate/envpol

Effects of pesticides and pharmaceuticals on biofilms in a highlyimpacted river

L. Proia a,*, V. Osorio b, S. Soley a, M. Köck-Schulmeyer a,b,c, S. Pérez b, D. Barceló b,c,A.M. Romaní a, S. Sabater a,c

a Institute of Aquatic Ecology, University of Girona, SpainbDepartment of Environmental Chemistry, Institute of Environmental Assessment and Water Research (IDAEA), Spanish National ResearchCouncil (CSIC), SpaincCatalan Institute for Water Research (ICRA), Spain

a r t i c l e i n f o

Article history:Received 16 June 2012Received in revised form1 February 2013Accepted 13 February 2013

Keywords:BiofilmPharmaceuticalsPesticidesLlobregat RiverTranslocation

* Corresponding author.E-mail address: proialorenzo@hotmail.it (L. Proia)

0269-7491/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.envpol.2013.02.022

a b s t r a c t

We investigated the effects of pharmaceuticals and pesticides detected in a Mediterranean river, onfluvial biofilms by translocation experiments performed under controlled conditions. Water was sampledfrom three sites along a pollution gradient. Biofilms grown in mesocosms containing relatively cleanwater were translocated to heavily polluted water. Several biofilm descriptors were measured before andafter translocations. Fifty-seven pharmaceuticals and sixteen pesticides compounds were detected inriver waters. The translocation from less to more polluted site was the most effective. Autotrophicbiomass and peptidase increased while phosphatase and photosynthetic efficiency decreased. Multi-variate analysis revealed that analgesics and anti-inflammatories significantly affected biofilm responses.Ibuprofen and paracetamol were associated with negative effects on photosynthesis, and with thedecrease of the green algae/cyanobacteria ratio, while diclofenac was associated with phosphatase ac-tivity. The effects of these emerging compounds on biofilms structure and function may cause importantalterations in river ecosystem functioning.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

A large number of organic compounds reach freshwaters andpose a risk to the structure and functioning of ecosystems. Althoughpesticides and pharmaceutical drugs are the most commonlydetected compounds (Azevedo et al., 2000; Daughton and Ternes,1999), they differ in the ways they enter water bodies. Pesticidesmainly originate from agricultural activities and enter aquatic en-vironments through diffuse sources via runoff. Their concentrationnormally peaks after rainfalls following spraying on agriculturalfields (Rabiet el al., 2010). Although various pesticides are currentlyincluded in the list of priority substances in the European Unionregulations (Decision 2455/2001/EC), many others are still unreg-ulated. Pharmaceuticals mainly enter aquatic environments viawastewater, and their concentrations in rivers are normally lowerthan those of pesticides (Petrovi�c et al., 2005). However, theirrelevance is related to the chronic character of their input, and theirconcentrations may increase as a consequence of lower dilution

.

All rights reserved.

situations characteristically occurring under water scarcity (Kusteret al., 2008). Pharmaceuticals are intrinsically bioactive compoundsand little is known about their effects on the aquatic ecosystemsresulting from long-term, low-dose exposure (Ginebreda et al.,2010). Although pharmaceuticals are considered as new emergingpollutants by the EC, only a few have been included in regulatorypolicies within the Water Framework Directive (EuropeanCommission, 2000). Environmental risk assessment (ERA) pro-cedures for both pesticides and pharmaceuticals are based onshort-term, single-species laboratory tests that only partially reflectreal ecosystem situations (Ginebreda et al., 2010).

To date, a few studies have investigated the effects of priorityand non-priority pollutants on real ecosystems (Hernando et al.,2006; Crane et al., 2006; Sanderson et al., 2004; Nunes et al.,2005; Pascoe et al., 2003), but less attention has been paid to theeffects on biofilms (Pesce et al., 2006; Ricart et al., 2010a). Riverbiofilms are complex microbial benthic communities composed ofautotrophic and heterotrophic organisms (Romaní, 2010), whichact as an interface between the water and the riverbed by inter-acting and responding rapidly to changes in environmental con-ditions (Sabater et al., 2007). Biofilms play a fundamental role in thetrophic web and in the biogeochemical cycles within aquatic

L. Proia et al. / Environmental Pollution 178 (2013) 220e228 221

ecosystems (Battin et al., 2003; Lock et al., 1993). The short life cycleof biofilm microorganisms and the trophic interactions betweenthe microbiota (algae, bacteria, fungi, protozoa) allow for thedetection of both short and long-term, and of direct and indirecteffects on the biofilm consortia (Proia et al., 2012a). River biofilmscan therefore be useful in determining the effects of pollutants onfreshwater ecosystems (Sabater et al., 2007).

This study aims to investigate the effects of pharmaceuticals andpesticides detected in the waters of the Llobregat River on thestructure and function of its biofilms. The Llobregat is the mostimportant source of drinking water for the city of Barcelona (Cat-alonia, NE Spain). However, this area is densely populated andaffected by intense industrial and agricultural activities. As aconsequence of these anthropogenic pressures, high concentra-tions of priority and emerging compounds occur in both water andsediments (Casas et al., 2003; Guerra et al., 2009). The environ-mental risk assessment of the pharmaceuticals detected in theLlobregat waters (Ginebreda et al., 2010), as well as the relation-ships between the occurrence of pharmaceuticals (Muñoz et al.,2009) and pesticides (Ricart et al., 2010a) and the structuralcomposition of benthic communities (macroinvertebrates and di-atoms) has been already carried out in this river. In search for causalevidence of the effects of these pollutants on biofilms, determiningthe effect of polluted water on reference communities can be ofhelp. This approach was applied in the present study in mesocosmsexperiments involving the translocation of biofilms from lesspolluted to more polluted waters, aiming to reveal the links be-tween the biofilm communities’ responses to pollutant concen-trations and to co-occurring environmental factors. Three samplingsites were selected to define a pollution gradient, translocationsfrom less to progressively more polluted waters were performed,and the effects on biofilms were determined. We hypothesized thatbioactive compounds affect the responses of biofilms, and themagnitude of the responses should be related to the incominghigher concentration of organic pollutants. To test our hypotheseswe related the observed biofilm responses in each translocationexperiment to the environmental factors and chemical waterquality of the sites.

2. Methods

2.1. Study sites

The Llobregat is a Mediterranean river located in NE Spain, which flows into theMediterranean Sea, south of Barcelona. The Llobregat is 165 km long and drains acatchment of 4948 km2 (Marcé et al., 2012). Its flow is characterized by a highvariability including periodic floods and droughts related to seasonal heavy rainfalland drought respectively (Ricart et al., 2010a). The mean annual discharge is693 Mm3 (Ginebreda et al., 2010), nearly 30% of which is used for drinking waterpurposes. As such, the Llobregat River is a paradigm of an overexploited river(Muñoz et al., 2009). Its watershed supports over 3 million people and receivessignificant inputs of industrial and urban wastewater (w137 Mm3/year, Ginebredaet al., 2010) as well as surface runoff from agricultural areas (Kuster et al., 2008).Moreover, salt inputs from the salt mines of the tributary Cardener have caused anincrease in water salinity, worsening the already poor conditions of the lower rea-ches of the river. In this study, three sampling sites were selected in the middle-lower part of the Llobregat following a pollution gradient: Castellbell (Reference,R) and Mina de Terrassa (Polluted, P), with low and moderate pollution respectively,and Sant Joan Despí (Highly Polluted, HP) being a pollution hotspot.

2.2. Experimental design

The biofilm responses to increasing pollutant concentrations were investigatedby means of translocation experiments performed under controlled conditions. Theriver water was collected three times a week between the 16th of October and the19th of November 2009 from the three sites, and then used as inoculum to growbiofilms in 18 independent mesocosms installed in the laboratory. Biofilms werecolonized on glass slides (1 cm2 each) placed at the bottom of each mesocosm (35e40 slides per mesocosm). The mesocosms consisted of sterile glass jars (19 cm indiameter, 9 cm high), filled with 1.5 L of water recirculated by a submersible pump

(Hydor, Pico 300, 230 V 50 Hz, 4.5 W). All mesocosms were kept in an incubator(SCLAB) with controlled temperature (18 �C) and light irradiance (150e180 mmolphotons m�2 s�1; dark/light cycle of 12 h/12 h). After 25 days of colonization, thebiofilms were translocated to more polluted waters. Three translocation experi-ments were carried out (three replicate jars per translocation): glass jars with bio-films previously incubated with R water were then filled with P water (R/P), andHP water (R/HP). Glass jars previously incubated with P water were then incu-bated with HP water (P/HP). Finally, in nine replicate glass jars (three per site)previously incubated with R, P and HP water, the original conditions were main-tained and used as controls. The biofilms were sampled four times: one before thetranslocations (3 November, day 0) and three after; day 2 (5 November), day 9 (12November) and day 16 (19 November). During the experiments, the water wasreplaced three times a week with river water from the respective sites. At eachsampling date one glass slide for each biofilm metric was collected from eachreplicate mesocosm.

2.3. Environmental conditions

2.3.1. Physical and chemical parametersConductivity, temperature, pH and dissolved oxygen were measured with the

sensor probes (HACH LANGE GMBH, Germany) both in the field and in the meso-coms, before and after each water renewal (n ¼ 15). Water samples were collectedfrom the glass jars and filtered (Nylon Membrane Filters 0.2 mm, WHATMAN, UK)before and after water renewals, prior to analysis. Soluble reactive phosphorus wasmeasured following Murphy and Riley (1962). Samples for anion and cation werepreserved frozen until the analysis (n¼ 14) by ion chromatography (761 Compact IC,METROHM, Switzerland).

2.3.2. Pharmaceuticals and pesticidesThe concentrations of 66 pharmaceuticals were analysed in surface waters using

the multiresidue analytical method based on LC-MS/MS after solid-phase extractiondescribed by Osorio et al. (2012). The concentrations of 16 pesticides were analysedfollowing the method based on online SPEeLCeMS/MS described by Köck-Schulmeyer et al. (2012). The analysis of both pharmaceuticals and pesticides wasperformed on triplicate at each sampling date (n ¼ 18).

2.4. Biofilm metrics

2.4.1. Chlorophyll in vivo fluorescence measurementsThe chlorophyll fluorescence emission of the biofilms was measured with a

Phyto-PAM (Pulse Amplitude Modulated) chlorophyll fluorometer (Heinz WalzGmbH), which uses a set of light-emitting diodes that excite chlorophyll using fourdifferent wavelengths (470, 520, 645 and 665 nm). For each glass slide sampled fromeach glass jar, three measurements were performed to represent the small-scaleheterogeneity of biofilms. All measurements were based on the proceduredescribed by Serra et al. (2009). The photosynthetic efficiency (Yeff) and capacity(Ymax) of PSII were measured based on the fluorescence signal recorded at 665 nmand given as relative units of fluorescence. The minimum fluorescence level of thedark-adapted samples was used as an estimate of autotrophic biomass. This estimatewas based on the fluorescence recorded at four different excitation wavelengths (F1at 470 nm, F2 at 520 nm, F3 at 645 nm, and F4 at 665 nm). F1 is linked to green algae,whereas F2 is mostly related to diatoms. The F3 signal is related to cyanobacteria andthe F4 signal is related to the whole algal community (Ricart et al., 2010a). The ratiobetween F1 and F3 was calculated for each replicate as an indicator of changes in theautotrophic community structure.

2.4.2. Chlorophyll-a densityOn each sampling day, one glass slide from each glass jar was collected and the

chlorophyll-a was extracted using 90% acetone for 12 h. Sonication during twominutes (40 W power, 40 kHz frequency, SELECTA, Spain) improved the pigmentextraction. The chlorophyll-a concentration was determined using spectrophoto-metric measurements (UV, 1800 Shimadzu) following the method described inJeffrey and Humphrey (1975).

2.4.3. Extracellular enzyme activitiesThe activities of the extracellular enzymes leucine-aminopeptidase (EC 3.4.11.1),

alkaline phosphatase (EC 3.1.3.1e2) and b-D-1,4-glucosidase (EC 3.2.1.21) in thebiofilms were measured spectrofluorometrically as described in Proia et al. (2013).

Leucine-aminopeptidase and b-D-1,4-glucosidase are mainly bacterial activitieswhile alkaline phosphatase may be produced by both algae and bacteria.

2.4.4. Phosphorus uptake capacityPhosphorus (P) uptake capacity was estimated bymeasuring the decrease in SRP

after a calculated spike. In each experiment, background samples were analysed inadvance in order to reach 4e8 times increase of the basal phosphorus concentrationwith the spike. The phosphorus uptake rate (U, mg P cm�2 h�1) was calculated as themass of P per unit area per unit time (Proia et al., 2011). As U could depend onbiomass and the colonized area was different in each day, the values werenormalized by expressing the mass of P removed from water per chlorophyll-a

L. Proia et al. / Environmental Pollution 178 (2013) 220e228222

density for each replicate at each sampling date. Previous abiotic controls wereperformed and showed no abiotic decrease in SRP during the experimental timeunder the same conditions.

2.5. Data analysis

2.5.1. Biofilm responses to translocationsOn each sampling day, the responses of each biofilm metric to each trans-

locations were analysed independently by one-way ANOVA with treatment beingconsidered as a fixed factor. Effects were analysed by the Tukey’s b post hoc test.Statistical significance was set at p ¼ 0.05. Analysis was performed using SPSSVersion 15.0.

2.5.2. Relationship between biofilm responses and environmental conditionsRedundancy analysis (RDA) was used to determine the influence of pharma-

ceuticals, pesticides and other environmental factors on biofilm responses. Thebiological dataset included biofilm metrics analysed after translocation (days 2, 9and 16). Biological data were square-root transformed. The environmental datasetincluded 31 variables: pH, conductivity, dissolved oxygen, NO3, SO4, SRP, K, Na, Mg,Cl, the concentrations of five pesticide classes (triazine, organophosphate, phenyl-urea, choroacetanilide and thiocarbamate) and the concentration of 15 pharma-ceutical classes (analgesic and anti-inflammatory drugs, b-Blockers, lipid regulators,psychiatric drugs, macrolid antibiotics, fluoroquinolone antibiotics, sulfonamideantibiotics, other antibiotics, tetracycline, stomach treatment drugs, barbiturates,blood pressure regulators, fungicides, anti-cancer and anti-diabetic drugs). Allenvironmental data were transformed by log10(x þ 1) to reduce skewed distribu-tions. Themaximum gradient length for the biofilmmetrics dataset was determinedusing detrended correspondence analysis (DCA). The maximum amount of variationwas 0.866, indicating that linear methods would be appropriate (ter Braak andSmilauer, 2002). To avoid co-linearity, the variables were selected based on the in-spection of variance inflation factors (VIF < 20) (ter Braak and Smilauer, 1998).Forward selection was used to reduce the environmental variables that significantlyexplained the responses of biofilms at a cut-off point of p ¼ 0.05. The significance ofthe RDA axes was assessed using theMonte Carlo permutation test (999 unrestrictedpermutations). The probabilities of multiple comparisons were counteracted byapplying the Bonferroni correction. To separate the effects of contaminants (phar-maceuticals and pesticides) from those of environmental variables on biofilm re-sponses, the variance partitioning technique was applied following Ricart et al.(2010a). This technique enabled us to assess the fractions of the explained vari-ance that are shared by two predictor variables, and to determine which of themcould be uniquely attributed to each of them (Borcard et al., 1992). The explanatoryvariables were therefore grouped into two subsets: (a) physical and chemical vari-ables and (b) contaminants. The following sequence of RDAs were performed: (a)RDA of the biofilm metrics matrix constrained by physical and chemical variables,(b) RDA of the biofilmmetrics matrix constrained by contaminants, (c) partial RDA ofthe biofilm metrics matrix constrained by physical and chemical variables using thecontaminants as co-variables and (d) partial RDA of the biofilm metrics matrixconstrained by contaminants using the physical and chemical variables as co-variables.

Based on the results of the first set of analyses, the contaminants dataset wasreduced, and the new dataset included concentrations of each single compound ofthe families of contaminants significantly explaining the biofilm response in the firstsequence of RDAs. With this new dataset of contaminants an additional sequence ofRDAs was performed: (a) RDA of the biofilm metrics matrix constrained by theselected compounds, (b) partial RDA of the biofilm metrics matrix constrained byphysical and chemical variables using the selected compounds as co-variables and(c) partial RDA of the biofilm metrics matrix constrained by the selected compoundsusing the physical and chemical variables as co-variables.

3. Results

3.1. Environmental conditions

3.1.1. Physical and chemical parametersDissolved oxygen (DO) and pH did not differ between sites,

while conductivity (Table 1) increased significantly from R to HP(p < 0.05). The concentration of soluble reactive phosphorus (SRP)gradually increased downstream, while the concentration of nitrate

Table 1Environmental variables measured at each sampling site during the experiment. Values

Flow (m3 s�1) pH Cond (mS cm�1)

R 6.93 (1.73) 8.43 (0.17) 1560.0 (195.0)P 6.16 (2.43) 8.40 (0.15) 1645.0 (274.3)HP 7.01 (12.53) 8.41 (0.18) 1819.0 (330.2)

(NO3) was similar in R and P and increased significantly in HP(p < 0.05, Table 1).

3.1.2. Pharmaceuticals and pesticidesA total of 57 pharmaceutical compounds from 14 different

therapeutic groups were detected at the three sites (Table 2). HPwas the most polluted site, while concentrations in R and P werelower and within a similar range. The highest concentrations ofcompounds at each sampling site corresponded to analgesics andanti-inflammatory drugs. On average, ibuprofen was the mostconcentrated compound in this group (Table 2). Psychiatric drugsand sulfonamide antibiotics were detected at concentrations higherthan 100 ng L�1 at each sampling site. The concentrations of bloodpressure regulators were 1928.2 � 561.4 ng L�1 in HP, and around200 ng L�1 at R and P. The maximum concentrations of the bloodpressure regulator hydrochlorothiazide, a thiazide diuretic, reached2930.9 ng L�1 at HP (Table 2).

A total of 16 compounds from five different pesticide classeswere detected in river water (Table 3). Concentrations were ingeneral low (<100 ng L�1), the highest at HP, and lower in R and P.The classes accounting for the higher concentrations were triazine,phenylurea and organophosphate. Those having the highest con-centrations at HP were terbuthylazine, diuron and diazinon,respectively (Table 3).

3.2. Biofilm responses

The differences between biofilms before translocation (day 0)were limited to the autotrophic compartment. For all the autotro-phic groups (green algae, diatoms and cyanobacteria), biofilmgrown with HP water had a significantly higher chlorophyll-adensity and fluorescence signal than those grown with R and Pwaters (Fig. 1, p< 0.05). Chlorophyll-a density of biofilms at HP was28.3 � 5.3 mg Chl cm�2, 1.9 and 1.7 times higher than thosemeasured in P and R respectively.

The biofilms’ responses to translocation depended on themagnitude of the differences in water chemistry. The translocationof biofilms from R to P was the least responsive. R/P biofilmsexhibited a significant decrease in extracellular peptidase activity,Fo and F1/F3 ratio on day 2 (Fig. 2), but from day 9 these metricsrecovered to values similar to R biofilms. Furthermore, the photo-synthetic efficiency of biofilms translocated between R/Pdecreased significantly on day 9, and recovered oneweek later (day16, Fig. 2).

In contrast, the translocation from P to HP caused structural andfunctional changes in the biofilm. Extracellular phosphatase activ-ity decreased to values significantly lower than P and HP on day 2,and the extracellular peptidase activity increased on days 2 and 16(Fig. 2). Moreover, the phosphate uptake capacity of HP biofilmswas lower than that of P biofilms. On day 16, the phosphate uptakerate of P/HP biofilm decreased to values closer to HP biofilms(Fig. 2). Chlorophyll-a density of translocated biofilms increased tovalues closer to HP, particularly on day 16. In contrast, Fo oftranslocated biofilms decreased significantly at day 2 and recoveredlater on (from day 9, Fig. 2).

Translocation from R to HP was the most effective in terms ofbiofilm responses. R/HP biofilms exhibited a significant increase

are expressed as mean values and SD in parenthesis (n ¼ 16).

DO (mg L�1) SRP (mg L�1) NO3 (mg L�1)

10.42 (1.04) 44.80 (57.18) 6.48 (1.58)10.37 (0.95) 70.85 (50.73) 7.32 (1.97)10.49 (1.00) 104.63 (38.04) 10.84 (2.62)

Table 2Maximum, minimum and average pharmaceutical concentrations measured at each sampling site during the experiment (n ¼ 18). Values are expressed in ng L�1.

R P HP

Min Max Average Min Max Average Min Max Average

Analgesic and anti-inflammatorydrugs

Ketoprofen 14.9 106.3 48.8 0.5 86.3 31.6 5.2 292.6 96.4Naproxen 68.9 198.9 127.7 85.0 165.4 111.8 126.8 258.5 185.5Ibuprofen 128.0 523.9 270.4 94.8 404.9 251.3 200.3 642.0 391.1Indometacine 12.7 26.6 20.1 9.8 109.0 25.1 17.2 123.8 48.8Diclofenac 86.5 202.9 142.5 83.7 184.0 125.5 128.6 445.2 311.0Acetaminophen 86.6 824.3 348.1 10.9 586.2 225.8 77.3 421.4 225.6Propiphenazone 2.2 4.6 3.6 2.2 4.8 3.7 6.0 36.3 17.4Phenazone 7.9 18.6 11.8 8.3 12.7 10.6 12.6 94.0 45.9Phenybutazone 2.3 6.5 4.9 1.0 6.6 3.6 4.0 50.5 18.2Codeine 7.0 30.6 15.9 5.5 26.8 14.4 3.8 122.7 55.7

b-Blockers Atenolol 21.2 47.8 37.4 17.2 45.1 33.3 62.5 251.2 153.8Sotalol 13.5 22.9 18.4 10.0 22.5 17.0 24.6 164.7 89.6Metoprolol 7.4 16.4 12.9 7.2 17.9 12.2 32.4 535.0 104.2Pindolol 0.0 0.2 0.1 0.1 0.3 0.1 0.2 0.4 0.3Carazolol 0.0 0.3 0.2 0.1 0.3 0.2 0.2 1.0 0.6Propanolol 8.8 25.8 14.3 11.9 22.5 16.2 10.3 70.4 37.8Timolol 1.2 153.5 14.7 1.0 2.9 1.8 2.2 10.3 6.8Nadolol 0.2 0.6 0.4 0.1 0.5 0.3 0.4 1.1 0.7

Lipid regulators Clofibric acid 1.3 18.7 3.4 1.1 2.4 1.8 6.8 40.1 20.8Gemfrobizil 12.0 25.4 18.3 9.0 24.3 16.2 21.2 152.0 85.5Benzafibrate 13.7 54.1 27.9 13.2 47.5 24.8 21.2 217.1 94.0Fenofibrate 24.8 97.6 47.1 19.3 56.6 34.9 35.1 277.6 148.8Atorvastatine 0.5 1.2 0.7 0.3 1.3 0.7 1.0 3.2 2.0Mevastatine 1.2 15.8 4.9 1.8 7.9 4.0 2.7 7.5 5.4

Psychiatric drugs Lorazepam 87.1 204.2 163.5 102.6 196.5 158.9 113.8 705.5 391.6Carbamazepine 35.3 59.2 51.0 36.1 62.8 51.9 53.2 278.2 173.6Diazepam 2.2 3.9 2.9 1.6 3.8 2.8 3.1 32.0 16.5Fluoxetine 12.8 29.0 16.1 4.3 34.1 14.2 10.3 53.5 33.1Paroxetine 1.3 4.0 2.1 1.1 9.6 4.4 3.8 145.1 24.4

Macrolid antibiotics Erytromicin 1.9 7.4 4.8 1.3 12.8 4.1 0.1 45.2 15.9Azythromicin 3.5 7.1 6.7 6.9 7.1 7.0 3.6 7.2 6.4Roxythromycin 0.4 1.1 0.7 0.2 0.9 0.5 0.7 8.1 3.6Clarithromicin 13.5 51.9 38.4 13.8 52.9 34.2 21.2 232.1 115.1Tylosin 1.8 4.3 2.8 1.0 4.7 2.7 2.2 30.3 8.1Josamycin 0.3 0.7 0.5 0.3 0.7 0.5 0.8 3.6 2.2Spyramicin 4.4 16.2 8.1 4.8 15.2 7.4 6.9 52.8 28.5Tilmicosin 0.3 370.5 31.8 0.8 95.8 8.8 1.9 96.6 11.8

Fluoroquinolone antibiotics Ofloxacine 11.6 87.5 29.9 11.9 32.2 20.8 24.1 337.8 156.4Ciprofloxacine 16.9 56.4 27.7 18.5 36.6 25.3 27.7 164.6 80.3Enoxacine 6.1 14.8 9.3 6.5 11.6 8.9 0.7 36.4 18.4Enrofloxacine 2.3 45.8 7.0 2.1 7.6 4.0 6.0 303.7 113.7Flumequine 0.2 0.9 0.4 0.2 0.8 0.4 0.3 0.9 0.5

Sulfonamide antibiotics Sulfamethoxazole 91.2 256.8 200.8 132.0 298.4 210.3 201.2 1576.0 717.2Sulfadiazine 2.7 17.8 6.5 2.6 45.7 12.4 7.1 43.4 30.1

Others antibiotics Trimethoprim 3.5 8.6 6.3 2.9 7.6 5.7 6.8 37.4 22.3Tetracycicline 3.5 12.4 7.3 2.8 35.1 14.7 25.6 788.8 247.1

Stomach treatment drugs Famotidine 0.6 1.0 0.9 0.7 1.0 0.9 0.2 7.5 3.0Ranitidine 1.1 3.3 2.3 0.7 3.4 2.2 0.1 115.8 15.1Cimetidine 1.1 3.6 2.3 0.3 3.9 2.3 0.2 42.3 9.7

Barbiturates Butalbial 1.9 14.6 7.0 1.8 27.3 11.5 2.2 4.3 3.0Pentobarbital 8.1 47.1 19.9 10.2 68.6 21.5 9.6 17.6 12.1Phenobarbital 2.3 25.4 9.2 2.6 22.1 11.0 2.3 12,1 5.8

Blood pressure regulators Enalapril 2.4 12.0 6.4 1.8 11.8 6.0 6.2 32.0 11.8Hydrochlorothiazide 140.7 286.2 219.2 100.8 236.7 185.5 272.8 2930.9 1283.1

Fungicides Metronidazole 0.2 0.6 0.3 0.1 0.6 0.3 0.6 4.5 2.7Anti-cancer drugs Tamoxifen 0.3 1.5 0.6 0.2 1.7 0.5 0.2 1.5 0.6Anti-diabetic drugs Glibenclamide 0.4 2.1 1.2 0.4 2.0 0.9 1.6 12.6 7.2

L. Proia et al. / Environmental Pollution 178 (2013) 220e228 223

in extracellular peptidase activity on day 9, and a significantdecrease in extracellular phosphatase activity on days 9 and 16(Fig. 2). The phosphate uptake capacity of HP biofilms was lowerthan that of R biofilms. The phosphate uptake rate of R/HPbiofilms decreased to resemble HP values (Fig. 2). The photo-synthetic efficiency of translocated biofilms decreased signifi-cantly on day 9 and recovered to R values on day 16. Thechlorophyll-a density of HP biofilms was significantly higher thanR and that of translocated biofilms tended to increase to valuescloser to HP from day 9 until the end of the experiment (Fig. 2).The Fo values of R/HP biofilms decreased significantly on day 2

and increased to values higher than R and HP on day 9 (Fig. 2).The F1/F3 ratio increased to values closer to HP biofilm on day 16(Fig. 2).

3.3. Relationships between biofilm responses and environmentalconditions

The RDA analysis showed that conductivity, analgesics andbarbiturates were the variables that had the greatest influence onthe responses of biofilms to translocation. This RDA accounted for57.3% of the variance.

Table 3Maximum, minimum and average concentrations of pesticides measured at each sampling site during the experiment (n ¼ 18). Values are expressed in ng L�1.

R P HP

Min Max Average Min Max Average Min Max Average

Triazine Atrazine 0.02 3.12 0.27 0.02 0.46 0.19 0.04 0.43 0.16Cyanacine 0.20 2.06 0.34 0.19 0.19 0.19 0.25 0.25 0.25Deisopropilatrazine 3.57 9.58 3.90 4.20 4.20 4.20 7.33 36.89 8.97Desetilatrazine 0.19 2.73 0.38 0.17 0.17 0.17 0.40 1.33 0.48Simazine 0.25 4.07 0.73 0.04 1.05 0.60 0.86 3.40 1.94Terbuthylazine 0.04 50.08 30.97 0.04 65.74 33.88 0.05 82.84 34.78

Phenylurea Diuron 1.12 5.04 2.60 1.38 5.09 3.04 10.53 31.13 22.65Isoproturon 0.01 0.71 0.20 0.01 0.60 0.14 0.02 0.81 0.28Linuron 1.02 1.02 1.02 0.92 0.92 0.92 1.17 1.17 1.17Clortoluron 0.03 3.09 0.30 0.03 0.59 0.09 0.06 6.20 1.29

Organophosphate Malation 0.22 4.63 0.48 0.20 0.43 0.21 0.24 0.79 0.27Diazinon 0.82 9.12 3.56 0.78 5.72 3.08 7.12 79.03 24.72Dimetoate 0.50 2.37 0.70 0.49 1.06 0.52 0.78 3.52 1.08

Chloroacetanilide Alaclor 0.68 0.68 0.68 0.66 0.66 0.66 0.90 0.90 0.90Metolaclor 0.13 1.31 0.31 0.13 0.13 0.13 0.17 1.30 0.23

Tiocarbamate Molinate 0.52 4.22 0.72 0.50 0.50 0.50 0.59 0.59 0.59

L. Proia et al. / Environmental Pollution 178 (2013) 220e228224

The potential contribution of the two sets of variables (physico-chemical and contaminants) was estimated by means of a secondredundancy analysis. Amongst the former, conductivity and SRPconcentration significantly influenced biofilm responses (24.4% ofthe total variance). The contaminants that most significantlyinfluenced the biofilm responses were analgesics, triazines,

Green algae Diatoms Cyanobacteria

Fluo

resc

ence

Uni

ts

0

10

20

30

40

50

60RPHP

R P HP

Chl

orop

hyll

a de

nsity

(g

cm-2

)

0

10

20

30

40

a a

b

a a

ba a

b

b

a

c

µ

Fig. 1. Chlorophyll-a density (above) and fluorescence values indicative of the differentautotrophic groups (below), before translocation. Values are means and standard de-viation (n ¼ 3). Different letters represent statistical significance at p � 0.05 (one-wayANOVA).

barbiturates and organophosphates (60.4% of the total variance).The partial redundancy analysis performed to evaluate the co-variance explained by the six significant variables, revealed thatof the total variance (62.3%) conductivity and SRP only accountedfor 1.9%, while analgesics, triazines, barbiturates and organophos-phates accounted for 37.9%. The shared variance explained repre-sented 22.5%. The SRP concentration was related to higherchlorophyll-a and photosynthetic capacity as well as to thedecrease in the phosphate uptake rate (Fig. 3a). Higher conductiv-ities were associated with lower photosynthetic efficiency, phos-phatase, and b- Glucosidase activities (Fig. 3a). Analgesics, triazinesand barbiturates affected most of the biofilmmetrics, especially theautotrophic ones (Fig. 3b). In particular, analgesics were related to adecrease in the F1/F3 ratio and photosynthetic capacity, while theincrease of triazines and barbiturates was related to lower chloro-phyll-a, Fo, and phosphatase activity (Fig. 3b).

The last iteration with multivariate analysis revealed that thecompounds that most significantly influenced biofilm responseswere the analgesics diclofenac, ibuprofen and acetaminophen(paracetamol), accounting for the 25.4% of the explained variance.Ibuprofen and acetaminophen affected chlorophyll-a density,peptidase activity, Fo, photosynthetic capacity and F1/F3 ratio(Fig. 3c). In particular, the increase of acetaminophenwas related toa decrease in F1/F3, the increase in ibuprofen was related to lowerphotosynthetic capacity, and that of diclofenac was associated withhigher phosphatase activity.

4. Discussion

As many as 16 pesticides and 57 pharmaceutical compoundswere detected in the Llobregat at concentrations ranging from<1 ng L�1 to ca. 3 mg L�1. These values were comparable to con-centrations measured in previous studies in the Llobregat (Kusteret al., 2008; Quintana et al., 2001; Muñoz et al., 2009; Ginebredaet al., 2010), and confirm observed patterns of higher concentra-tions of pharmaceuticals than of pesticides in this river, whichvaried depending on the time of the year. Pesticides mostly enterriver waters from diffuse sources after rainfall episodes (Rabiet elal., 2010). This study was performed after a dry summer and nosignificant precipitations occurred during the experiment. Theseconditions during the study period determine the low flow regis-tered (<than 10 m3 s�1) that favoured the concentration of prod-ucts entering continuously from WWTP effluents (Kuster et al.,2008). Other studies performed on the same river confirmed theimportance of hydrologic seasonal patterns for the occurrence of

Fig. 2. Biofilm metrics changes in response of each translocation. R/P ¼ Translocation from Castellbell to Mina de Terrassa; R/HP ¼ Translocation from Castellbell to Sant JoanDespí; P/HP ¼ Translocation from Mina de Terrassa to Sant Joan Despí. Values are means and standard deviation (n ¼ 3). Different letters represent statistical significance atp � 0.05 (one-way ANOVA).

6.08.0-

8.06.0-

Pho

Pept

B glu

Chl a

F1/F3

F0

Ymax

Yeff

U

Analgesics

Barbiturics

Triazines

Organophosphates

6.06.0-4.0

3.0-

Pho PeptB glu

Chl aF1/F3

F0

Ymax

Yeff

U

Conductivity

SRP

8.06.0-

4.06.0-

Pho

Pept

B glu

Chl a

F1/F3

F0

Ymax

Yeff

U

Ibuprofen

Diclofenac

Acetaminophen

a)

b)c)

Fig. 3. Sequence of RDAs with the environmental and biological data. a) Partial RDA of the biofilm metrics matrix constrained by physical and chemical variables using thecontaminants as co-variables. b) Partial RDA of the biofilm metrics matrix constrained by families of contaminants using the physical and chemical variables as co-variables. c)Partial RDA of the biofilm metrics matrix constrained by individual compounds using the physical and chemical variables as co-variables. Only the variables that significantlyexplained the response of the biofilm to translocation are shown.

L. Proia et al. / Environmental Pollution 178 (2013) 220e228226

chemicals entering from punctual or diffuse sources (Osorio et al.,2012; Köck-Schulmeyer et al., 2012).

The pharmaceuticals detected in the Llobregat waters matchedthose most commonly consumed in Spain (Spanish Ministry ofPublic Health; http://www.aemps.gob.es). The most abundanttherapeutic group was the analgesic and anti-inflammatory drugs,represented by high concentrations of ibuprofen, diclofenac andacetaminophen (paracetamol). The most abundant pharmaceuticaldetected during our study was the blood pressure regulator hy-drochlorothiazide, a diuretic compound which is known to be toxicto fish cells (Caminada et al., 2006).

Several studies have investigated the presence and potentialeffects of pesticides (Ricart et al., 2009) and pharmaceuticals(Muñoz et al., 2009) on biological communities under field condi-tions. We used translocation as an experimental tool in order todefine the response of natural communities to different mixtures ofpesticides and pharmaceuticals. Translocations have been usedsuccessfully to describe the biofilms’ responses to various envi-ronmental and anthropogenic stressors (Victoria and Gómez, 2010;Ivorra et al., 1999; Tlili et al., 2011). Our translocation experimentsshowed that the mixtures of chemicals occurring in the LlobregatRiver affected the structure and function of biofilms, and that thisresponse was more pronounced when there was a higher concen-tration of pollutants. In particular, pharmaceuticals of the mostabundant therapeutic groups affected the biofilms structure andfunction. The responses varied in accordance with the time and themagnitude of the changes in water quality. Most of the measuredbiofilm metrics (both autotrophic and heterotrophic) responded totranslocation from the least (R) to the most (HP) polluted site to agreater extent than to the other translocations, since the HP water

was the most polluted in terms of nutrients (mainly phosphatesand nitrates) and contaminants (pesticides and pharmaceuticals)concentrations.

The direction of biofilms’ responses could be explained by bothdirect and indirect effects of environmental factors and chemicalpollution on community structure and function. The general in-crease of autotrophic biomass observed in translocated biofilmsmay be explained by the increasing availability of nutrientsdownstream. Considering the importance of microbial interactionwithin biofilms this result may indirectly help to explain thebehaviour of extracellular enzymatic activities in translocatedcommunities. In particular, the increase of peptidase activity couldbe explained by the increasing availability of algal exudates inbiofilm with higher autotrophic biomass. In fact, the high weightmolecules present in algal exudates are the substrate of the enzy-matic peptidase activity and this activity increase in thicker andmore complex biofilms (Proia et al., 2012b). In contrast, thereduction of the phosphatase activity in translocated biofilms maybe explained by end product inhibition. In fact, phosphatase ca-talyses the hydrolysis of phosphate esters, liberating inorganicphosphorus available for microbial uptake. The increase of availableinorganic phosphorus inhibits the alkaline phosphatase activity ofalgae and bacteria in biofilms as already observed (Proia et al.,2012b; Romaní et al., 2004; Sabater et al., 2005). Nevertheless,some direct or indirect effects of the huge number of priority andemerging pollutants detected in Llobregat waters may also inter-ferewith the observed results. For example the reduction of biofilmphotosynthetic efficiency may be due to the direct action of someherbicide on autotrophs (i.e. Diuron) while reduction of phosphateuptake capacity may reflect a general loss of biofilm performance in

L. Proia et al. / Environmental Pollution 178 (2013) 220e228 227

removing nutrients from water column caused by the conjoint ofdifferent effects of pollutants on target and non-target organisms(Proia et al., 2013). All these responses in biofilm structure andfunction may have consequences at ecosystem level as biofilmsplay an important role in freshwater ecosystems for organic matterre-mineralization and inorganic nutrient fluxes through hydro-graphic web.

The redundancy analyses revealed that biofilm responses totranslocation were affected by both physico-chemical and organiccontamination factors. Among the physico-chemical variables,conductivity and SRP explained a high percentage of the variabilityof the biofilm responses, whereas amongst the contaminants, tri-azines, analgesics and anti-inflammatories had the greatest influ-ence. Those pollutants and other physical and chemical variablescontributing to the biological responses in the Llobregat River werepreviously described for invertebrates and diatom communities(Muñoz et al., 2009; Ricart et al., 2010a). Significant relationshipsbetween SRP and triazines and the diatom community compositionwere observed in epilithic and epipsammic biofilms (Ricart et al.,2010a). The composition of the invertebrate community in theLlobregat River was significantly related to conductivity (Ricartet al., 2010a), but also to analgesic and anti-inflammatory concen-trations (Muñoz et al., 2009). In our study, biofilm metrics identi-fied three analgesic and anti-inflammatory compounds, whichexplained a significant percentage of the variance in biofilm re-sponses to translocation. The products detected (diclofenac, para-cetamol and ibuprofen) commonly occur in aquatic ecosystems(Daughton and Ternes, 1999; Ellis, 2006; Fent et al., 2006; Hebereret al., 2002), and their acute toxicity has been tested on severalaquatic organisms (Cleuvers et al., 2003; David et al., 2009; Ferrariet al., 2004).

In our study, acetaminophen (paracetamol) and ibuprofenmainly affected autotrophic descriptors. Acetaminophen wasrelated to a decrease in the F1/F3 ratio of biofilms, suggesting anegative effect on green algae and/or a positive effect on cyano-bacteria. Although data on the direct effects of paracetamol onalgae and biofilms are not available, those for other aquatic or-ganisms (Brain et al., 2004; David et al., 2009; Kim et al., 2007,2010) describe its potential toxicity. Brain et al. (2004) describedacetaminophen toxicity to growth rates and pigment content oftwo species of aquatic macrophytes, when mixed with seven otherpharmaceuticals, and David et al. (2009) reported acute toxicity tothe embryonic development of zebrafish at concentrations in therange of those measured in the Llobregat River. Moreover, acet-aminophen was the most toxic pharmaceutical tested on Daphniamagna and its toxicity increases with a rise in water temperature(Kim et al., 2010). Ibuprofen affected the biomass of photosyntheticorganisms in biofilm as well as their photosynthetic capacity, andwas also related to the decrease in the F1/F3 ratio. Ibuprofenshowed a stimulatory effect on cyanobacteria growth at concen-trations within the range of those measured in our study (Pomatiet al., 2004). The negative effects of ibuprofen on autotrophic bio-film biomass described in this study are in line with the findings ofother authors, and may result from both direct and indirect effects(Lawrence et al., 2005). Furthermore, ibuprofen was seen to be oneof the most effective compounds to affect the invertebrate com-munity structure of the Llobregat (Muñoz et al., 2009), and hasshown significant effects on the food web in the long-term(Richards et al., 2004). The acute toxicity of ibuprofen combinedwith that of diclofenac may affect algal growth (Cleuvers, 2003).These compounds may also act non-specifically via non-polarnarcosis (Cleuvers, 2003) on autotrophs (Desmodesmus spp.) andconsumers (Daphnia spp). Diclofenac is one of the most toxiccompounds at several trophic levels of the aquatic food web(Cleuvers, 2003; Fent et al., 2006; Ginebreda et al., 2010). Our study

showed a limited relationship between diclofenac and biofilmautotrophic metrics. Only the increase of biofilm phosphatase ac-tivity could be associated with diclofenac concentrations. Diclofe-nac effects on biofilm heterotrophs could either be due to a directimpact on their activities or indirectly derived from the structuralefunctional relationships between autotrophs and heterotrophswithin biofilms, as widely described in several studies (Bonnineauet al., 2010; Proia et al., 2011, 2013; Ricart et al., 2010b).

This study showed that biofilm translocation along a pollutiongradient is a sensitive tool for investigating the effects of realcontaminant mixtures. The biofilm responses accurately reflectedthe water quality of the sites. The continuous arrival of bioactivecompounds at low concentrations may lead to chronic effects onbiological communities. This can be particularly relevant in riverssubject to water scarcity, since toxicants reach higher concentra-tions during low flow periods. The biofilm responses detected inthis study show that the effects of these pollutants on the basalcompartment of the river food web do affect its structure andfunction.

Acknowledgements

Financial support was provided by the EU projects, MODELKEY(SSPI-CT-2003-511237-2) and KEYBIOEFFECTS (MRTN-CT-2006e035695), and the Spanish projects, SCARCE (Consolider-IngenioCSD2009-00065), FLUMED-HOT-SPOTS(CGL2011-30151-C02-01)and VIECO (009/RN08/011).

References

Azevedo, D., Lacorte, S., Vinhas, T., Viana, P., Barceló, D., 2000. Monitoring of prioritypesticides and other organic pollutants in river water from Portugal by gaschromatographyemass spectrometry and liquid chromatographyeatmosphericpressure chemical ionization mass spectrometry. Journal of Chromatography A879, 13e26.

Battin, T.J., Kaplan, L.A., Newbold, J.D., Hansen, C.M.E., 2003. Contributions of mi-crobial biofilms to ecosystem processes in stream mesocosms. Nature 426(6965), 439e442.

Bonnineau, C., Guasch, H., Proia, L., Ricart, M., Geiszinger, A., Romaní, A.M.,Sabater, S., 2010. Fluvial biofilms: a pertinent tool to assess b-blockers toxicity.Aquatic Toxicology 96 (3), 225e233.

Borcard, D., Legendre, P., Drapeau, P., 1992. Partialling out the spatial component ofecological variation. Ecology 73, 1045e1055.

Brain, R.A., Johnson, D.J., Richards, S.M., Hanson, M.L., Sanderson, H., Lam, M.W.,Young, C., Mabury, S.A., Sibley, P.K., Solomon, K.R., 2004. Microcosm evaluationof the effects of an eight pharmaceutical mixture to the aquatic macrophytesLemna gibba and Myriophyllum sibiricum. Aquatic Toxicology 70, 23e40.

Caminada, D., Escher, C., Fent, K., 2006. Cytotoxicity of pharmaceuticals found inaquatic systems: comparison of PLHC-1 and RTG-2 fish cell lines. AquaticToxicology 79, 114e123.

Casas, J.M., Rosas, H., Solé, M., Lao, C., 2003. Heavy metals and metalloids in sedi-ments from the Llobregat basin, Spain. Environmental Geology 44, 325e332.

Cleuvers, M., 2003. Aquatic ecotoxicity of pharmaceuticals including the assessmentof combination effects. Toxicology Letters 142, 185e194.

Crane, M., Watts, C., Boucard, T., 2006. Chronic aquatic environmental risks fromexposure to human pharmaceuticals. Science of the Total Environment 367, 23e41.

Daughton, C.G., Ternes, A.T., 1999. Pharmaceuticals and personal care products inthe environment: agents of subtle change? Environmental Health Perspectives107, 907e938.

David, A., Pancharatn, K., 2009. Effects of acetaminophen (paracetamol) in theembryonic development of zebrafish Danio rerio. Journal of Applied Toxicology29, 597e602.

Ellis, J.B., 2006. Pharmaceutical and personal care products (PPCPs) in urbanreceiving waters. Environmental Pollution 144, 184e189.

European Commission, 2000. Directive 2000/60/EC of the European Parliament andof the Council e Establishing a Framework for Community. Action in the Field ofWater Policy. European Commission, Belgium.

Fent, K., Weston, A.A., Caminada, D., 2006. Ecotoxicology of human pharmaceuti-cals. Aquatic Toxicology 76, 122e159.

Ferrari, B., Mons, R., Vollat, B., Fraysse, B., Paxéus, N., Lo Giudice, R., Pollio, A.,Garric, J., 2004. Environmental risk assessment of six human pharmaceuticals:are the current environmental risk assessment procedures sufficient for theprotection of the aquatic environment? Environmental Toxicology and Chem-istry 23, 1344e1354.

L. Proia et al. / Environmental Pollution 178 (2013) 220e228228

Ginebreda, A., Muñoz, I., López de Alda, M., Brix, R., López-Doval, J., Barceló, D.,2010. Environmental risk assessment of pharmaceuticals in rivers: relationshipsbetween hazard indexes and aquatic macroinvertebrate diversity indexes in theLlobregat River (NE Spain). Environment International 36, 153e162.

Guerra, P., Eljarrat, E., Barceló, D., 2009. Analysis and occurrence of emergingbrominated flame retardants in the Llobregat River basin. Journal of Hydrology383, 39e43.

Heberer, T., 2002. Occurrence, fate, and removal of pharmaceutical residues in theaquatic environment: a review of recent research data. Toxicology Letters 131,5e17.

Hernando, M.D., Mezcua, M., Fernandez-Alba, A.R., Barceló, D., 2006. Environmentalrisk assessment of pharmaceutical residues in wastewater effluents, surfacewaters and sediments. Talanta 69, 334e342.

Ivorra, N., Hettelaar, J., Tubbing, G.M.J., Kraak, M.H.S., Sabater, S., Admiraal, W., 1999.Translocation of microbenthic algal assemblages used for in situ analysis ofmetal pollution in rivers. Archives of Environmental Contamination and Toxi-cology 37, 19e28.

Jeffrey, S., Humphrey, G.F., 1975. New spectrophotometric equations for deter-mining chlorophylls a, b, c1 and c2 in higher plants, algae and natural phyto-plankton. Biochemie und Physiologie der Pflanzen 167, 191e194.

Kim, Y., Choi, K., Jung, J., Park, S., Kim, P., Park, J., 2007. Aquatic toxicity of acet-aminophen, carbamazepine, cimetidine, diltiazem and six major sulfonamides,and their potential ecological risks in Korea. Environment International 33,370e375.

Kim, J., Park, J., Kim, P., Lee, C., Choi, K., Choi, K., 2010. Implication of global envi-ronmental changes on chemical toxicity effect of water temperature, pH, andultraviolet B irradiation on acute toxicity of several pharmaceuticals in Daphniamagna. Ecotoxicology 19, 662e669.

Köck-Schulmeyer, M., Ginebreda, A., González, S., Cortina, J.L., López de Alda, M.,Barceló, D., 2012. Analysis of the occurrence and risk assessment of polarpesticides in the Llobregat River Basin (NE Spain). Chemosphere 86, 8e16.

Kuster, M., López de Alda, M.J., Hernando, M.D., Petrovic, M., Martín-Alonso, J.,Barceló, D., 2008. Analysis and occurrence of pharmaceuticals, estrogens, pro-gestogens and polar pesticides in sewage treatment plant effluents, river waterand drinking water in the Llobregat river basin (Barcelona, Spain). Journal ofHydrology 358, 112e123.

Lawrence, J.R., Swerhone, G.D.W., Wassenaar, L.I., Neu, T.R., 2005. Effects of selectedpharmaceuticals on riverine biofilm communities. Canadian Journal of Micro-biology 51, 655e669.

Lock, M.A., 1993. Attached microbial communities in rivers. In: Fored, T.E. (Ed.),Aquatic Microbiology: an Ecological Approach. Blackwell Scientific Publications,Oxford, pp. 113e138.

Marcé, R., Honey-Rosés, J., Manzano, A., Moragas, L., Catllar, B., Sabater, S., 2012. TheLlobregat River Basin: a paradigm of impaired rivers under climate changethreats. In: Sabater, S., Ginebreda, A., Barceló, D. (Eds.), The Llobregat River: thestory of a polluted river, vol. 21. The Handbook of Environmental Chemistry,Springer Verlag, Berlin Heidelberg, pp. 1e26.

Muñoz, I., López-Doval, J.C., Ricart, M., Villagrasa, M., Brix, R., Geszinger, A.,Ginebreda, A., Guasch, H., López de Alda, M., Romaní, A.M., Sabater, S.,Barceló, D., 2009. Bridging levels of pharmaceuticals in river water with bio-logical community structure in the Llobregat river basin (NE Spain). Environ-mental Toxicology and Chemistry 28 (12), 2706e2714.

Murphy, J., Riley, J.P., 1962. A modified single solution method for the determinationof phosphate in natural waters. Analytica Chimica Acta 27, 31e36.

Nunes, B., Carvalho, F., Guilhermino, L., 2005. Acute toxicity of widely used phar-maceuticals in aquatic species: Gambusia holbrooki, Artemia parthenogeneticaand Tetraselmis chuii. Ecotoxicology and Environmental Safety 61, 413e419.

Osorio, V., Pérez, S., Ginebreda, A., Barceló, D., 2012. Pharmaceuticals on a sewageimpacted section of a Mediterranean 5 River (Llobregat River, NE Spain) andtheir relationship with hydrological conditions. Environmental Science andPollution Research. http://dx.doi.org/10.1007/s11356-011-0603-4.

Pascoe, D., Karntanut, W., Muller, C.T., 2003. Do pharmaceuticals affect freshwaterinvertebrates? A study with the cnidarian Hydra vulgaris. Chemosphere 51,521e528.

Pesce, S., Fajon, C., Bardot, C., Bonnemoy, F., Portelli, C., Bohatier, J., 2006. Effects ofthe phenylurea herbicide diuron on natural riverine microbial communities inan experimental study. Aquatic Toxicology 78, 303e314.

Petrovi�c, M., Dolores Hernando, M., Díaz-Cruz, M.S., Barceló, D., 2005. Liquidchromatographyetandem mass spectrometry for the analysis of pharmaceu-tical residues in environmental samples: a review. Journal of Chromatography A1067, 1e14.

Pomati, F., Netting, A.G., Calamari, D., Neilan, B.A., 2004. Effects of erythromycin,tetracycline and ibuprofen on the growth of Synechocystis sp. and Lemna minor.Aquatic Toxicology 67, 387e396.

Proia, L., Morin, S., Peipoch, M., Romaní, A.M., Sabater, S., 2011. Resistance and re-covery of river biofilms receiving short pulses of Triclosan and Diuron. Scienceof the Total Environment 409, 3129e3137.

Proia, L., Cassiò, F., Pascoal, C., Tlili, A., Romaní, A.M., 2012a. The use of attachedmicrobial communities to assess ecological risks of pollutants in river ecosys-tems. The role of heterotrophs. In: Guasch, H., Ginebreda, A., Geiszinger, A.(Eds.), Emerging and Priority Pollutants in Rivers: Bringing Science into RiverManagement Plans. Springer Verlag, Berlin Heidelberg, pp. 55e83.

Proia, L., Romaní, A.M., Sabater, S., 2012b. Nutrients and light effects on streambiofilms: a combined assessment with CLSM, structural and functional pa-rameters. Hydrobiologia 695, 281e291.

Proia, L., Vilches, C., Boninneau, C., Kantiani, L., Farré, M., Romaní, A.M., Sabater, S.,Guasch, H., 2013. Drought episode modulates the response of river biofilm totriclosan. Aquatic Toxicology 127, 36e45.

Quintana, J., Martí, I., Ventura, F., 2001. Monitoring of pesticides in drinking andrelated waters in NE Spain with a multiresidue SPE-GCeMS method includingan estimation of the uncertainty of the analytical results. Journal of Chroma-tography A 938, 3e13.

Rabiet, M., Margoum, C., Gouy, V., Carluer, N., Coquery, M., 2010. Assessing pesticideconcentrations and fluxes in the stream of a small vineyard catchment e effectof sampling frequency. Environmental Pollution 158, 737e748.

Ricart, M., Barceló, D., Geiszinger, A., Guasch, H., López de Alda, M., Romaní, A.M.,Vidal, G., Villagrasa, M., Sabater, S., 2009. Effects of low concentrations of thephenylurea herbicide diuron on biofilm algae and bacteria. Chemosphere 76,1392e1401.

Ricart, M., Guasch, H., Barceló, D., Brix, R., Conceição, M.H., Geiszinger, A., López deAlda, M.J., López-Doval, J.C., Muñoz, I., Postigo, C., Romaní, A.M., Villagrasa, M.,Sabater, S., 2010a. Primary and complex stressors in polluted Mediterraneanrivers: pesticide effects on biological communities. Journal of Hydrology 383,52e61.

Ricart, M., Guasch, H., Alberch, M., Barceló, D., Bonnineau, C., Geiszinger, A.,Farré, M., Ferrer, J., Ricciardi, F., Romaní, A.M., Morin, S., Proia, L., Sala, L.,Sureda, D., Sabater, S., 2010b. Triclosan persistence through wastewater treat-ment plants and its potential toxic effects on river biofilms. Aquatic Toxicology100, 346e353.

Richards, S.M., Wilson, C.J., Johnson, D.J., Castle, D.M., Lam, M., Mabury, S.A.,Sibely, P.K., Solomon, K., 2004. Effects of pharmaceutical mixtures in aquaticmicrocosms. Environmental Toxicology and Chemistry 23 (4), 1035e1042.

Romaní, A.M., 2010. Freshwater biofilms. In: Dürr, S., Thomason, J.C. (Eds.),Biofouling, first ed. Wiley-Blackwell, Oxford.

Romaní, A.M., Giorgi, A., Acuña, V., Sabater, S., 2004. The influence of substratumtype and nutrient supply on biofilm organic matter utilization in streams.Limnology and Oceanography 49, 1713e1721.

Sabater, S., Acuña, V., Giorgi, A., Guerra, E., Munoz, I., Romaní, A.M., 2005. Effects ofnutrient inputs in a forested Mediterranean stream under moderate lightavailability. Archive für Hydrobiologie 163, 479e496.

Sabater, S., Guasch, H., Ricart, M., Romaní, A.M., Vidal, G., Klünder, C., Schmitt-Jansen, M., 2007. Monitoring the effect of chemicals on biological communities.The biofilm as an interface. Analytical and Bioanalytical Chemistry 387, 1425e1434.

Sanderson, H., Johnson, D.J., Reitsma, T., Brain, R.A., Wilson, C.J., Solomon, K.R., 2004.Ranking and prioritization of environmental risks of pharmaceuticals in surfacewaters. Regulatory Toxicology and Pharmacology 39 (2), 158e183.

Serra, A., Corcoll, N., Guasch, H., 2009. Copper accumulation and toxicity in fluvialperiphyton: the influence of exposure history. Chemosphere 74, 633e641.

Ter Braak, C.J.F., Smilauer, P., 1998. CANOCO Reference Manual and User’s Guide toCanoco for Windows: Software for Canonical Community Ordination (Version4). Microcomputer Power, Ithaca, New York, 352 pp.

Ter Braak, C.J.F., Smilauer, P., 2002. CANOCO Reference Manual and CanoDraw forWindows User’s Guide: Software for Canonical Community Ordination (Version4.5). Microcomputer Power, Ithaca, New York, 500 pp.

Tlili, A., Corcoll, N., Bonet, B., Morin, S., Montuelle, B., Bérard, A., Guasch, H., 2011. Insitu spatio-temporal changes in pollution-induced community tolerance to zincin autotrophic and heterotrophic biofilm communities. Ecotoxicology 20, 1823e1839.

Victoria, S.M., Gómez, N., 2010. Assessing the disturbance caused by an industrialdischarge using field transfer of epipelic biofilm. Science of the Total Environ-ment 408, 2696e2705.