Analysis of multi-class pharmaceuticals in fish tissues by ultra-high-performance liquid chromatography tandem mass spectrometry

Highlights

• Exhaustive method optimization for determination of 20 multi-class pharmaceuticals in fish.

• Method application to different fish tissues.

• In-depth evaluation of several strategies to correct observed matrix effects.

• First monitoring campaing performed in fish in Mediterranean rivers.

• First detection of diclofenac in fish tissues.

Abstract

A new sensitive method based on pressurized liquid extraction (PLE) and purification by gel permeation chromatography (GPC) prior to ultra-high-performance liquid chromatography coupled to tandem mass spectrometry (UHPLC–MS/MS) was developed for the determination in fish homogenate, liver and muscle of twenty pharmaceuticals compounds and metabolites from seven commonly used therapeutic families. An extensive matrix effect evaluation was performed in order to select the best approach when analyzing such complex matrices. Limits of detection (MDLs) for the target compounds were in the range of 0.03–0.50 ng/g for fish homogenate, 0.01–0.42 ng/g for fish muscle, and 0.08–0.98 ng/g for fish liver. The method was applied to fish tissues of eleven fish species from four heavily impacted Mediterranean rivers. Nine compounds from five therapeutic families were measured at concentrations higher than MDLs. Highest levels were found in trout liver, with a maximum concentration of 18 ng/g for carbamazepine, whereas the most ubiquitous compound was diclofenac.

Introduction

In recent years, the occurrence, fate, and adverse effects of pharmaceutical residues in aquatic organisms have become a noteworthy issue. In Europe, the legislative proposal for amending the list of priority substances that represent a significant risk to or via the aquatic environment was presented by the European Commission on 31 January 2012, and included for the first time the pharmaceutical substances 17-α-ethinylestradiol (EE2), 17-ß-estradiol (E2) and diclofenac [1]. The U.S. Environmental Protection Agency (EPA) has also included some pharmaceutical substances in the Drinking Water Contaminant Candidate List, such as the antibiotic erythromycin and the estrogenic hormones 17-α-estradiol, estriol and estrone [2]. In addition, many reports have highlighted the potential of pharmaceuticals and/or their metabolites to accumulate in tissues of aquatic organisms, such as crustaceans, molluscs and fish, as a consequence of their chronic exposure in aquatic ecosystems [3], [4], [5]. In general, pharmaceutically active compounds (PhACs) are highly hydrophilic, and their bioaccumulation potential might be considered irrelevant, particularly when compared to other pollutants, such as pesticides and persistent organic compounds (POPs). These conventional pollutants have been reported in a vast number of studies to bioaccumulate in different organisms because of their lipophilicity and tendency to bind to organic matter [6], [7], [8], [9], [10] and are considered in many priority pollutants lists [11], [12]. However, some studies have indicated that the bioaccumulation of PhACs is not only determined by chemical lipophilicity, and other processes should also be considered, such as active transport through biological membranes or uptake and depuration kinetics [13], [14], [15].

Analytical techniques used for the detection of PhACs presence at (ultra)trace quantities in environmental matrices have advanced significantly in the last few years and have been summarized in recent reviews [16], [17], [18], [19], [20], [21], [22], [23], [24]. Even though an increasing number of analytical procedures have been reported for several therapeutic families in biota in the last years, they are still sparse, probably due to the challenges associated with the complexity of the biological matrices [25]. Groups of PhACs analyzed in biota so far include psychiatric drugs [5,26–29], synthetic hormones [30], and antibiotics [3], [29], [31], [32]. Exhaustive sample preparation followed by sensitive detection techniques is required in these cases, due to the very low concentration of analytes in biological matrices [33]. Both, ultrasonication and pressurized liquid extraction (PLE) have been often used for the extraction of PhACs in aquatic organisms, such as crustaceans, mussels, algae, and fish [27], [28], [30], [34], [35], [36], [37]. Regarding the crucial purification step of the sample extract, different clean-up procedures have been used: solid-phase extraction (SPE) in Florisil columns [38], Oasis HLB cartridges [39] or techniques like gel-permeation chromatography (GPC) [30], [40]. Last step in the analytical process includes the identification and determination of PhACs, usually based on liquid (LC) or gas chromatography (GC) in combination with mass spectrometry (MS) detection, because it provides high selectivity, specificity, and sensitivity required for this type of studies [5], [27], [29].

This work describes the development, optimization and validation of a method for the determination of 20 multi-class pharmaceuticals and metabolites in fish (Table 1), selected according to their detection frequency in water and sediment in Mediterranean rivers [41], [42] as well as to their potential negative effects in aquatic organisms [43], [44], [45], [46]. Selected compounds included analgesics and non-steroidal anti-inflammatories (NSAIDs), diuretics, psychiatric drugs, anti-histaminics, antihelmintics, β2-adrenergic receptor agonists and β-blockers. The method developed is based on an extraction step using pressurized liquid extraction (PLE) followed by gel permeation chromatography (GPC) clean-up and ultra performance liquid chromatography–triple quadrupole mass spectrometry (UPLC–MS/MS) for the detection of target compounds. Critical steps in method development involved (i) the selection and optimization of the most appropriate sample pre-treatment step that allowed the simultaneous extraction of selected compounds from fish homogenates, liver and muscle tissues, (ii) lipid removal from the extract to reduce matrix interferences during analysis and (iii) comparison of several strategies to correct the observed matrix effects.

The method was subsequently applied to assess the occurrence of target compounds in fish samples collected from different sites of four Mediterranean rivers in Spain: Ebro, Llobregat, Júcar and Guadalquivir, all subjected to intensive anthropogenic activity. Nine pharmaceutical compounds belonging to five different therapeutic families were detected for the first time in fish from Mediterranean rivers.