Evaluation of selected ubiquitous contaminants in the aquatic environment and their transformation products. A pilot study of their removal from a sewage treatment plant

Abstract

A simple method using direct sample injection combined with liquid chromatography tandem mass spectrometry has been developed for the simultaneous analysis of six alkaloid compounds in environmental samples. The target list includes two psychostimulants (nicotine and caffeine), three metabolites (cotinine, nicotinic acid and paraxanthine) and a coffee chemical (trigonelline). The analytical method was evaluated in three different matrices (surface water, influent and effluent wastewater). The method developed showed an adequate sensitivity, below 0.6 μg L⁻¹ for wastewater and 0.1 μg L⁻¹ for river matrices, without any prior treatment of the samples. Finally, the methodology was applied to real samples for evaluation of their removal from a sewage treatment plant and their persistence/fate in the aquatic environment. All compounds studied in this work were detected at all sampling points collected along the Henares River. However, nicotinic acid was only detected three times in treated sewage samples at levels above its detection limit.

Introduction

Coffee is one of the most popular beverages worldwide, appreciated not only for its characteristic taste and aroma, but also recently for its potentially beneficial effects on human health (Ranheim and Halvorsen, 2005). Caffeine, trigonelline and nicotinic acid are some of the coffee’s constituent compounds with the most relevant biological activity (Trugo, 2003). Caffeine is a xanthine alkaloid found in tea, cocoa, chocolate, energy drinks and other derived food. However, coffee has the highest caffeine content with regards to other dietary products. It is suggested that values range from 840 to 1500 mg of caffeine/100 g of commercial coffee (Perrone et al., 2008). In humans, caffeine is a central nervous system stimulant, and it is present in a large number of prescriptions because of its diuretic properties and benefits associated with improvements in alertness, learning capacity and exercise performance. About 80% of the caffeine dose is metabolized in the liver to paraxanthine (1,7-dimethylxanthine), 10% to theobromine (3,7-dimethylxanthine), and 4% to theophylline (1,3-dimethylxanthine). Trigonelline is an alkaloid found in barley, cantaloupe, corn, onions, peas, soybeans, tomatoes, crustaceans, fish, and mussels. It is also highly available in the coffee, the average content is 280–950 mg/100 g in commercial coffee (Perrone et al., 2008). In humans, this alkaloid possesses anti-cancer (cervix and liver), anti-migraine, antiseptic, hypoglycaemic and mutagenic properties. About 50–80% of the trigonelline is decomposed during coffee roasting, forming nicotinic acid, a water-soluble B vitamin also known as niacin, and other aromatic nitrogen compounds. However, only 5% of the nicotinic acid consumed is metabolized to trigonelline in humans (Zeiger, 1997). Nicotinic acid (vitamin B3 or niacin) is also an organic compound found in a variety of foods including liver, chicken, beef, fish, cereal, peanuts and legumes, but it is also present in prepared coffee (10–30 mg/100 g of commercial coffee) (Perrone et al., 2008). Niacin is produced as a vitamin supplement in tablet form as hypocholesterolemic and antihyperlipidemic (Zeiger, 1997).

Although nicotine is the most abundant alkaloid in tobacco (98% of the total alkaloids, 13–25 mg nicotine/cigarette) (Wu et al., 2002), there are some other plants which likewise contain nicotine, e.g. other members of the Solanaceae family such as potatoes, tomatoes, egg plants or chilli peppers; and members of the Camellia sinensis family such as tea products. A variety of black as well as green teas have been previously investigated by Siegmund et al. (1999) for estimation of the nicotine content in tea. Between 0.002 and 1.695 mg nicotine/kg were found in several kinds of teas with a mean value 0.5 μg/L. On the other hand, nicotine is rapidly and extensively metabolized in humans: approximately 70–80% of the nicotine absorbed by a smoker is transformed to cotinine and excreted in the urine (Bramer and Kallungal, 2003). Nicotine and cotinine can be measured in various biological fluids (blood, saliva, and urine) (Benowitz et al., 2002). In its pure form it is fast acting and has often been used as an agricultural insecticide. However, although it is a natural insecticide generated by plants as a defense against insects, nicotine-based insecticides have been banned in the U.S. since 2001 in order to prevent residues from contaminating foodstuffs (Environmental Protection Agency, EPA). Although nicotine can be oxidized to nicotinic acid with nitric acid (McElvain, 1941), a few publications have reported studying nicotine biosynthesis in cell-free systems at the enzymatic level, caused by a reduction reaction of nicotinic acid with NADPH (Friesen and Leete, 1990).

Due to the worldwide consumption of coffee and tobacco, or food and beverages which contain caffeine and nicotine, these organic compounds and their active metabolites may be continually introduced into the aquatic environment, via numerous and varied routes, such as industrial waste related to coffee processing or untreated, and treated, municipal wastewater. Fig. 1 shows a schematic overview of some of the possible routes of introduction of these contaminants into the aquatic system. A potential tracer or indicator of human impact is that substance clearly of anthropogenic origin and often has been detected in wastewater and surface water. A widespread detection of these chemicals in the environment may give weight to their potential relationship with water contamination due to anthropogenic sources. Therefore, a good marker should allow magnitude quantification as well as the unambiguous elucidation of the pollution source. Some of the requirements for being an adequate chemical marker are: (i) constant consumption, considering the population’s consumption habits or that the compound is not phased out in future and (ii) a regular detection: the quantities discharged into the environment should be sufficient to permit their detection after dilution/dissipation in the aquatic medium. As well as this, some of the possible removal processes to which these compounds may be subjected (sedimentation, volatilization or biotic/abiotic degradation in the environment) should be evaluated. Human endogenous metabolites, constituents in pharmaceuticals, personal care products or food are some of the substances which might be evaluated as possible markers (Buerge et al., 2003). Thus, in the last few years, caffeine, paraxanthine, nicotine and cotinine have been widely proposed as chemical markers for anthropogenic contamination processes (Buerge et al., 2003, Buerge et al., 2008, Martínez Bueno et al., 2010).

To our knowledge, up until now there has been no scientific literature on monitoring data for nicotinic acid or trigonelline in aquatic environment, in contrast to the wide-ranging information concerning the occurrence, persistence and fate of many pharmaceuticals and emergent contaminants in the environment processes (Martínez Bueno et al., 2010, Huerta-Fontela et al., 2007, Huerta-Fontela et al., 2008, Castiglioni et al., 2006, Ternes, 1998), as is the case for nicotine, cotinine, caffeine and paraxanthine. The analysis of these compounds in aqueous environmental samples has been performed using different methodologies, such as enzyme-linked immunosorbent assay (Nicolardi et al., submitted for pub), gas chromatography-mass spectrometry (GC-MS) (Buerge et al., 2003, Weigel et al., 2004, Gómez et al., 2007) or liquid chromatography-mass spectrometry (LC-MS) (Buerge et al., 2008, Martínez Bueno et al., 2010, Huerta-Fontela et al., 2007, Huerta-Fontela et al., 2008, Castiglioni et al., 2006, Ternes, 1998).

The continuing advances in analytical instrumentation, detection systems and separation techniques have provided analytical tools to reduce the treatment and pre-concentration of the sample, that allows very low detection limits (LODs). However, most of the above methods generally require time-consuming sample preparation procedures, such as liquid–liquid extraction (LLE) or solid-phase extraction (SPE) (Buerge et al., 2003, Buerge et al., 2008, Martínez Bueno et al., 2010, Huerta-Fontela et al., 2007, Huerta-Fontela et al., 2008, Castiglioni et al., 2006, Ternes, 1998, Nicolardi et al., submitted for pub, Weigel et al., 2004) prior to analysis of the sample, which is often tedious and time consuming. Direct injection methods offer the advantage of reduced sample preparation steps and therefore improved reproducibility and minimized potential contamination of the sample. Nevertheless, to date, the direct analysis of wastewater without prior sample treatment has not come into widespread use because of typically low sensitivity with respect to the levels present in the aquatic environment.

The aim of this work was to develop and validate a rapid LC-MS/MS method by direct sample injection to determine and evaluate nicotine, caffeine and some of their transformation products, ubiquitous in the aquatic environment as possible anthropogenic tracers. For that, the direct injection of a sample into a liquid chromatography-linear ion trap mass spectrometry (LC-QqQLIT-MS/MS) system was the procedure developed in this study. The more demanding requirements regarding mass spectrometric confirmation currently set by EU regulations were taken into account in carrying out the confirmation and quantification of target compounds (Commission Decision (2002/657/EC), 2002).