Development of a rapid and comprehensive method for identifying organic micropollutants with high ecological risk to the aquatic environment

Highlights

• Organic micropollutants (OMPs) identified by targeted & nontargeted analysis.

• Our method (CSM) gives overall picture of OMPs pollution in the aquatic environment.

• Ecological risk evaluated via quantitative structure–activity relationship (QSAR).

• Candidate high-risk OMPs rapidly identified at low cost using CSM and QSAR.

• Phthalates, pharmaceuticals and personal care products priority OMPs in urban rivers in Japan.

Abstract

Currently, tens-of-thousands of chemicals are used in Japan, and their presence in and impact on aquatic ecosystems are poorly understood. Because conventional risk evaluation processes using target analysis and biological tests are time-consuming and costly, it is challenging to investigate all substances. Therefore, we aimed to develop a rapid and highly efficient screening scheme for identifying hazardous organic micropollutants (OMPs) in aquatic ecosystems. The scheme is divided into two steps: chemical analysis and risk evaluation. First, a comprehensive screening method (CSM) using gas chromatography (GC)–mass spectrometry (MS) and a database containing nearly 1000 compounds is used to identify known compounds, and nontargeted analysis is carried out using a GC × GC-time-of-flight (TOF)MS to detect compounds not registered in the database. Secondly, the predicted toxicity values obtained by quantitative structure–activity relationship (QSAR) are used to evaluate and rank the ecological risk of each detected OMPs and to identify priority compounds for detailed survey. To assess the proposed scheme, we surveyed representative urban rivers in Japan and ranked the potential toxicity of the identified compounds. The total number of compounds detected in water from each river ranged from 29 to 87, and the total concentrations ranged from 2.3 to 63 μg L⁻¹. Pharmaceuticals and personal care products, such as crotamiton and galaxolide, were identified in the urban rivers and found to have high ecotoxicity rankings. Thus, the scheme combining CSM and risk evaluation using QSAR is a novel screening that can identify candidates with high ecological risk in aquatic environment rapidly and efficiently.

Introduction

Economic and technological developments have improved the quality of life around the world, but the natural environment has been damaged by these changes. Based on the concept of planetary boundaries, which evaluates the effects of human activities on the global environmental, proposed by Rockström et al. (2009), climate change and biodiversity loss are already above their threshold limits. At present, no assessment of the state of chemical contamination has been reported. However, it is highly probable that chemical contamination is also at or near the global threshold because pollution by harmful chemicals, such as persistent organic pollutants (POPs), is widespread (Ludwicki et al., 2015; Zheng et al., 2016; Takahashi et al., 2017). In fact, concerns about environmental pollution caused by emerging chemicals have increased in recent years. Further, according to the American Chemical Society Chemical Abstracts Service, the number of registered substances exceeded 100 million in 2015 (American Chemical Society, 2015), and, the number and amounts of new chemicals produced has increased geometrically in recent years. Many of these chemicals have not been assessed for their environmental persistence and could have toxic effects on living organisms. In particular, environmental pollution by pharmaceutical and personal care products (PPCPs) and other emerging chemicals has been reported in various parts of the world (Nakata et al., 2012; Lopez et al., 2015; Machado et al., 2016; K’oreje et al., 2016). Furthermore, the heavy contamination of rivers with organic micropollutants (OMPs) is affecting many local river ecosystems (Zounkova et al., 2014; Maier et al., 2015).

Thus, the determination of the concentrations of ecotoxic chemicals in the environment and a comparison of their toxicity values, such as the predicted no-effect concentrations (PNEC), is necessary because this would allow the impact of contaminants on aquatic organisms to be assessed (von der Ohe et al., 2011; Guruge et al., 2019). However, it is challenging to apply conventional analytical and toxicity data acquisition methods to the rapidly increasing number of chemical substances. Importantly, conventional analytical techniques require complicated and time-consuming analytical operations and are limited in the range of concentrations and substances that can be analyzed. Further, toxicity assessment is also costly and time-consuming, requiring animal experiments, and the amount of available toxicity data is limited. Therefore, to protect aquatic ecosystems from the proliferation of chemicals, improvements in analytical methods and toxicity assessment are required.

Kadokami et al. (2005) developed an automated identification and quantification system (AIQS) using gas chromatography–mass spectrometry (GC-MS) combined with a database, and this system can determine nearly 1000 semi-volatile organic compounds (SVOCs) in environmental samples. However, because AIQS analysis is a target-based technique, only compounds registered in the database can be measured. To investigate environmental contaminants comprehensively, screening methods that do not limit the target compounds are required. Nontargeted analysis is particularly good for identifying unknown compounds in the environment; for example, some studies have used a GC-Orbitrap (Krätschmer et al., 2018) or two-dimensional GC (GC × GC)-time-of-flight (TOF) MS (Jia et al., 2015; Prebihalo et al., 2015). This latter method has the advantages described above but has not been used for comprehensive quantification because there are few databases, such as AIQS, for identification and quantification. Thus, we focused on the qualitative ability of nontargeted analysis with a GC × GC-TOFMS and the quantitative performance of AIQS, and devised a novel analysis system combining the advantages of both approaches. The first objective of this study is to develop an analytical scheme for the rapid and efficient identification and quantification of OMPs in river water. The analytical scheme is as follows. First, target screening was carried out by target screening analysis (TSA) using AIQS (TSA-AIQS), and, subsequently, the samples were subjected to nontargeted analysis with a GC × GC-TOFMS to identify unknown compounds not registered in the AIQS database. Subsequently, we conducted retrospective analysis and quantified these newly identified compounds after their registration in the AIQS. The novel screening method using TSA-AIQS-assisted nontargeted analysis is denoted the comprehensive screening method (CSM) in this study.

The final objective of this study is to identify hazardous OMPs rapidly and at low cost. Regarding toxicity data for ecological risk evaluation, we employed the quantitative structure–activity relationship (QSAR) technique. QSAR is a method for predicting the harmfulness of a chemical compound from the relationship between the chemical structure of the compound and biological activity. Therefore, QSAR has attracted attention as a toxicity assessment method that could replace animal experiments. Indeed, QSAR-based methods have been used for the examination of new chemical compounds in accordance with the United States Toxic Substances Control Law (Zeeman et al., 1995). However, the predictive power of QSAR is not sufficient because it is not based on the results of biotoxicity tests, although it does allow rapid and easy calculation of the predicted toxicity values of many compounds. Thus, we used the predictive power of QSAR to estimate the ecological risk of the OMPs detected by CSM, rank their toxicities, and identify priority compounds potentially affecting aquatic organisms.

Our proposed survey scheme is characterized by the combination of chemical analysis using CSM and risk evaluation using QSAR. Previous studies have reported environmental screening using TSA-AIQS (Matsuo et al., 2019), whereas our scheme includes not only targeted analysis using AIQS, but also non-targeted analysis, which allows wider range of compound detection. Furthermore, by applying predicted value of QSAR to detected compounds, ecotoxicity screening can be performed rapidly and at low cost. The proposed survey scheme is novel and simple, making it suitable for the comprehensive evaluation of the ecological risk of a large number of chemicals present in the environment. In this study, we used our method to carry out an initial environmental survey on water samples from four urban rivers and a suburban river in Japan to identify compounds potentially affecting the aquatic ecosystem. Additionally, after ranking the compounds in order of risk, we report OMPs that should be investigated with priority in each studied river.