Comparative field study on bioassay responses and micropollutant uptake of POCIS, Speedisk and SorbiCell polar passive samplers

https://doi.org/10.1016/j.etap.2020.103549Get rights and content

Highlights

  • 43 of 108 analyzed compounds were detected in POCIS, Speedisk or SorbiCell extracts.

  • Most compounds were detected in POCIS samplers, and least in SorbiCell.

  • Most frequent and intense bioassay responses were induced by POCIS extracts.

  • POCIS passive sampling leads to more exceedances of effect-based trigger values.

  • POCIS is best suited as passive sampler for effect-based water quality monitoring.

Abstract

Routine water quality monitoring is generally performed with chemical analyses of grab samples, which has major limitations. First, snapshot samples will not give a good representation of the water quality. Second, it is not sufficient to analyze only a limited number of (priority) pollutants. These limitations can be circumvented by an alternative environmental risk assessment that combines time-integrated passive sampling (PS) with effect-based methods. This study aimed to select which of three polar PS devices was best suited for effect-based monitoring strategies.

In the first part of this study, Speedisk, SorbiCell and POCIS polar PS devices were compared by simultaneous deployment at five sites. Chemical analyses of 108 moderately polar compounds (-1.82 < log D < 6.28) revealed that highest number of compounds, with the widest range of log KOW, log D and pKa, were detected in extracts of POCIS, followed by Speedisk. SorbiCell samplers accumulated the lowest numbers and concentrations of compounds, so they were not further investigated. In a follow-up study, bioassay responses were compared in extracts of POCIS and Speedisk devices deployed at eight sites. The passive sampler extracts were subjected to bioassays for non-specific toxicity, endocrine disruption, and antibiotics activities. More frequent and higher responses were induced by POCIS extracts, leading to more exceedances of effect-based trigger values for environmental risks. As POCIS outperformed Speedisk, it is better suited as PS device targeting polar compounds for semi-quantitative effect-based water quality monitoring.

Introduction

Routine water quality monitoring is generally performed by chemically analyzing grab samples for a limited number of compounds. The European Union Water Framework Directive (WFD), for instance, aims to achieve and ensure a good chemical surface water status by implementing a regular chemical monitoring program for 45 (groups of) priority compounds. Measured concentrations are compared to environmental quality standards (EQS) for the water phase as annual average (AA) and maximum allowable (MAC) concentrations (European commission, 2013). Since regular chemical monitoring programs with conventional methods have some serious limitations, more relevant alternatives need to be developed.

In the commonly applied grab sampling, water concentrations of target contaminants are reported as the ‘snapshot’ concentrations. The monthly or yearly frequency of grab sampling does not represent the temporal variations of compounds, as concentrations of most micropollutants typically vary over time (Jones et al., 2015; Brack et al., 2017). Another disadvantage of grab sampling is that the detection limits (LOD) of some pollutants are higher than their EQS values (Harman et al., 2012). Passive sampling (PS) can overcome these limitations by applying a time-integrated measurement of bioavailable micropollutant concentrations in water, which reflects the actual exposure conditions in the water body over extended periods (Vrana et al., 2005). Moreover, with the ability to extract large volumes of water, PS techniques can lower detection limits to overcome the limitations for chemicals with a LOD above their EQS (Jones et al., 2015; Terzopoulou and Voutsa, 2016). A drawback of PS, particularly for adsorption-based samplers such as POCIS or Chemcatcher, is the uncertainty in calculations of accurate time-weighted average (TWA) concentrations because sampling rates (RS) are influenced by environmental conditions such as temperature, water flow rate, salinity, and the formation of biofilms on the surface of the devices (Balaam et al., 2011; Harman et al., 2012; Jones et al., 2015; Roll and Halden, 2016). Moreover, the composition of the mixture extracted from the PS devices is not the same as the mixture in water because of compound-specific uptake rates and partitioning coefficients (Brack et al., 2016; Van der Oost et al., 2017b).

A second limitation of regular monitoring programs is that lists of target priority compounds are generally not representative of present-day contamination, and therefore provide limited information on the relationships between pollution and risks to aquatic organisms (Altenburger et al., 2019). Many priority compounds are being phased out or banned, and their emissions are decreasing (Altenburger et al., 2015; Fliedner et al., 2016) while industries have switched to alternative compounds that may have a serious impact on chemical water quality (Schwarzenbach et al., 2006; Busch et al., 2016). At present, more than 350,000 chemicals are registered for production and use on the global market (Wang et al., 2020). It can be expected that a large number of these chemicals, as well as their transformation products, will end up in the water cycle (Brack et al., 2017). In addition, environmental mixtures, with potential synergism or antagonism, may still cause adverse effects although the concentrations of individual chemicals are below the EQS values (Carvalho et al., 2014). As a result, a large portion of the toxic effects observed in surface waters cannot be explained by compounds that water authorities are required to monitor (Escher et al., 2013; Brack et al., 2016; Tousova et al., 2017). On the other hand, chemical analysis of the myriad of compounds present in the aquatic environment is practically and economically impossible. Therefore, mixture toxicity assessment should be included in water quality monitoring strategies. Bioassays integrate the combined effects of all bioactive compounds in a water sample and are thus recommended (Carvalho et al., 2014; Wernersson et al., 2015; Brack et al., 2017; Van der Oost et al., 2017a; Novák et al., 2018). For several decades, both in vivo and in vitro bioassays have been applied in water quality assessment and have been proven successful in benchmarking water quality (Escher et al., 2013; Neale et al., 2015; Di Paolo et al., 2016; Van der Oost et al., 2017b).

Given the limitations of the present chemical water quality assessment, there is an urgent need for a time-integrated effect-driven monitoring strategy that employs a combination of PS and bioassays. In recent years, the suitability of the combination of PS and subsequent effect monitoring in water quality assessment has received much attention, and accordingly, multiple monitoring strategies applying such an approach have been described (Altenburger et al., 2015; Van der Oost et al., 2017a; Hamers et al., 2018; De Baat et al., 2019, 2020). In the Dutch SIMONI strategy (Smart Integrated Monitoring), a suite of bioanalyses is exposed to the extracts of two types of passive samplers (Van der Oost et al., 2017a). Effect-based trigger values (EBT) can be subsequently used to quantify the environmental risks based on the bioassay responses, making this method suitable for routine monitoring programs (Van der Oost et al., 2017a, 2017b).

Due to passive sampler specific affinities for the wide variety of organic compounds, the adequate selection of the types of passive samplers is crucial for their successful application in water quality monitoring strategies (Ahrens et al., 2015). None of the currently known PS devices will effectively accumulate compounds from the full range of the hydrophobic (non-polar) to hydrophilic (polar) spectrum (Ahrens et al., 2015). Hence, usually a combination of passive samplers targeting non-polar and polar compounds is employed in surface water quality monitoring (Petty et al., 2004; Booij et al., 2013; Van der Oost et al., 2017a; Hamers et al., 2018; De Baat et al., 2019). Partitioning or equilibrium PS devices are generally used for the sampling of non-polar compounds and adsorptive passive samplers are generally used to collect more polar compounds (Brack et al., 2017).

Consistent results for non-polar compounds have been obtained by partitioning samplers such as silicone rubbers (SR) and semi permeable membrane devices (SPMDs) (Allan et al., 2010; Booij et al., 2016), but a selection for the most suited passive sampler for polar compounds is still under debate. Many contaminants of emerging concern, such as pharmaceuticals & personal care products (PPCPs), polar pesticides, and endocrine disruptors (EDCs), are polar substances that may exert toxic effects on aquatic organisms in the ng-μg/L range (Daughton, 2005; Fauvelle et al., 2014). This underlines the need to standardize the employment of polar passive samplers for time-integrated effect-based monitoring strategies. Therefore, the present paper is focused on adsorptive samplers targeting polar compounds. The two kinetic passive samplers most often used for polar compounds are the polar organic chemical integrative sampler (POCIS) and Chemcatcher® (Alvarez et al., 2004; Kingston et al., 2006; Mills et al., 2014). Alternatively, an increased use of Speedisk and SorbiCell devices as passive samplers for polar compounds was reported in recent years. Speedisk is a solid-phase extraction (SPE) disk that is also applied as a passive sampler for polar compounds (Hamers et al., 2018; Zwart et al., 2017). SorbiCell is a flow-through cartridge containing an adsorbent and a tracer salt that indicates the volume of water sampled during deployment (De Jonge and Rothenberg, 2005). Unlike POCIS and Speedisk, the uptake mechanism of SorbiCell is based on the advection flow of water through the sampler (Rozemeijer et al., 2010). The system does not need a pump, since this flow is produced by the hydrostatic pressure which is influenced by the depth of the sampler below the surface water (Rozemeijer et al., 2010).

The present study aimed to examine which of three devices (POCIS, Speedisk and SorbiCell) is recommended to be applied as a PS device for a semi-quantitative (using average extracted water volumes for all compounds) effect-based risk assessment of a wide range of moderately polar compounds, by i) comparing the amounts, numbers, and ranges of target micropollutants that accumulated in the three types of PS devices, and ii) comparing the responses of a bioassay battery to extracts of the two types of PS devices that appeared to be most suited based on the accumulation of target compounds (POCIS and Speedisk). PS devices were simultaneously deployed in two separate sampling campaigns at sites likely to be contaminated with polar compounds. For the chemical survey, POCIS, Speedisk, and SorbiCell extracts were analyzed for 108 compounds with log KOW values ranging from -0.8 to 6.6. For the biological survey, POCIS and Speedisk extracts were subjected to a suite of bioassays responsive to polar organic compounds, thus generating toxicity profiles. Ideally, the two approaches should be combined, but unfortunately the two studies for chemical and bioassay data were disconnected from each other, performed in different campaigns at different sites. Therefore, no direct comparisons could be made between chemical concentrations and bioassay responses. Establishing exact time-weighted average chemical concentrations and their relationships with bioassay responses, however, are not the primary aims of the present paper. The main objective is to provide recommendations, based on the chemical and biological results, on the preferred type of polar passive sampler for semi-quantitative effect-based monitoring strategies.

Section snippets

Sampling sites

Different types of passive samplers were deployed at various surface water sites in two separate sampling campaigns, a chemical campaign performed in 2015, and an effect-based campaign performed in 2016. The selected sampling sites listed in Table 1 were located within the management areas of the Waternet and Rijnland water boards, near the city of Amsterdam. Sites were selected for their known contamination with polar compounds, originating from urban land use, agricultural activity, or WWTP

Campaign 1: chemical analyses

Out of the 108 targeted compounds (Table S2), 43 compounds were detected in extracts of the three samplers (Table 2). Details on concentrations of compounds detected in each sampler per site are shown in Table S3. The overall distribution of these compounds over the samplers is shown in Fig. 1.

The highest number of compounds was detected in extracts of POCIS (n = 38), followed by Speedisk (n = 25), and SorbiCell (n = 19). Fifteen compounds were detected in extracts of all three samplers. Most

Chemical analyses

Summarizing the literature data on compound affinities for passive samplers, silicone rubbers and SPMDs are reported to effectively take up the hydrophobic compounds (logKOW range 3–7), whereas hydrophilic compounds accumulate more effectively in Speedisk (logKOW range -0.7 to 4.5) and in POCIS (logKOW range -2.6 to 4.0) (Mazzella et al., 2007; Vermeirssen et al., 2012; Morin et al., 2012, 2013; Fauvelle et al., 2014; Booij et al., 2016; Silvani et al., 2017). In fact, several hydrophobic

CRediT authorship contribution statement

M. Thao Nguyen: Formal analysis, Investigation, Data curation, Writing - original draft, Writing - review & editing, Visualization. Milo L. De Baat: Data curation, Writing - original draft, Writing - review & editing. Ron Van Der Oost: Conceptualization, Methodology, Validation, Writing - review & editing, Supervision, Funding acquisition. Willie Van Den Berg: Writing - review & editing, Funding acquisition. Pim De Voogt: Writing - review & editing.

Declaration of Competing Interest

The authors report no declarations of interest.

Acknowledgements

The present study was funded by Waterproef Foundation, Waternet Institute for the Urban Water Cycle, and the Amsterdam Water Science (AWS) collaboration. Part of the POCIS samplers were kindly provided by Patrick Bäuerlein of KWR. We thank Rob Visee for conducting most field experiments and Maria Vasileiadou and Hannah van de Kerkhof for their assistance in the field and the laboratory. Gerrit van der Honing and Bram Aalbregtse are acknowledged for performing the chemical analyses for this

References (72)

  • A. Fliedner et al.

    Current levels and trends of selected EU Water Framework Directive priority substances in freshwater fish from the German environmental specimen bank

    Environ Pollut

    (2016)
  • T. Hamers et al.

    Time-Integrative Passive sampling combined with Toxicity Profiling (TIPTOP): an effect-based strategy for cos-effective chemical water quality assessment

    Environ Toxicol Pharma

    (2018)
  • M. Isidori et al.

    Toxic and genotoxic evaluation of six antibiotics on non-target organisms

    Sci. Total Environ.

    (2005)
  • V. Jálová et al.

    Estrogen-, androgen- and aryl hydrocarbon receptor mediated activities in passive and composite samples from municipal waste and surface waters

    Environ. Int.

    (2013)
  • S. Lissalde et al.

    Liquid chromatography coupled with tandem mass spectrometry method for thirty-three pesticides in natural water and comparison of performance between classical solid phase extraction and passive sampling approaches

    J. Chromatogr. A

    (2011)
  • S. Lissalde et al.

    Polar organic chemical integrative samplers for pesticides monitoring: impacts of field exposure conditions

    Sci. Total Environ.

    (2014)
  • E. Martínez-Carballo et al.

    Environmental monitoring study of selected veterinary antibiotics in animal manure and soils in Austria

    Environ Pollut

    (2007)
  • N. Mazzella et al.

    Determination of kinetic and equilibrium regimes in the operation of polar organic chemical integrative samplers. Application to the passive sampling of the polar herbicides in aquatic environments

    J. Chromatogr. A

    (2007)
  • N. Morin et al.

    Chemical calibration, performance, validation and applications of the polar organic chemical integrative sampler (POCIS) in aquatic environments

    Trends Analyt. Chem.

    (2012)
  • N. Morin et al.

    Determination of uptake kinetics and sampling rates for 56 organic micropollutants using “pharmaceutical” POCIS

    Talanta

    (2013)
  • A.J. Murk et al.

    Chemical-activated luciferase gene expression (CALUX): a novel in vitro bioassay for ah receptor active compounds in sediments and pore water

    Fundam. Appl. Toxicol.

    (1996)
  • J. Novák et al.

    Effect-based monitoring of the Danube River using mobile passive sampling

    Sci. Total Environ.

    (2018)
  • J.D. Petty et al.

    A holistic passive integrative sampling approach for assessing the presence and potential impacts of waterborne environmental contaminants

    Chemosphere

    (2004)
  • M.G. Pikkemaat et al.

    A new microbial screening method for the detection of antimicrobial residues in slaughter animals: the Nouws antibiotic test (NAT-screening)

    Food Control

    (2008)
  • I.B. Roll et al.

    Critical review of factors governing data quality of integrative samplers employed in environmental water monitoring

    Water Res.

    (2016)
  • L. Silvani et al.

    Monitoring alkylphenols in water using the polar organic chemical integrative sampler (POCIS): determining sampling rates via the extraction of PES membranes and Oasis beads

    Chemosphere

    (2017)
  • H. Sun et al.

    Antiandrogenic activity of pyrethroid pesticides and their metabolite in reporter gene assay

    Chemosphere

    (2007)
  • Z. Tousova et al.

    European demonstration program on the effect-based and chemical identification and monitoring of organic pollutants in European surface waters

    Sci. Total Environ.

    (2017)
  • A.D. Vethaak et al.

    An integrated assessment of estrogenic contamination and biological effects in the aquatic environment of the Netherlands

    Chemosphere

    (2005)
  • B. Vrana et al.

    Passive sampling techniques for monitoring pollutants in water

    Trends Analyt. Chem.

    (2005)
  • N. Zwart et al.

    Improved androgen specificity of AR-EcoScreen by CRISPR based glucocorticoid receptor knockout

    Toxicol. In Vitro

    (2017)
  • R. Altenburger et al.

    Future water quality monitoring: improving the balance between exposure and toxicity assessments of real-world pollutant mixtures

    Environ. Sci. Eur.

    (2019)
  • D.A. Alvarez et al.

    Development of a passive, in situ, integrative sampler for hydrophilic organic contaminants in aquatic environments

    Environ. Toxicol. Chem.

    (2004)
  • AZUR Environmental

    Microtox® Basic Test Procedure

    (1998)
  • J. Balaam et al.

    Project C3461 – from marine strategy framework directive to good environmental status: the use of passive sampling for monitoring offshore waters, CEFAS

    United Kingdom.

    (2011)
  • P.S. Bäuerlein et al.

    Sorption behavior of charged and neutral polar organic compounds on solid phase extraction materials: Which functional group governs sorption?

    Environ. Sci. Technol.

    (2012)
  • Cited by (0)

    1

    Equal contribution.

    View full text