Elsevier

Journal of Chromatography A

Volume 1141, Issue 2, 9 February 2007, Pages 174-181
Journal of Chromatography A

Analysis of estrogens in river sediments by liquid chromatography–electrospray ionisation mass spectrometry: Comparison of tandem mass spectrometry and time-of-flight mass spectrometry

https://doi.org/10.1016/j.chroma.2006.12.045Get rights and content

Abstract

A method has been developed and optimised for the determination of two natural estrogens, estrone (E1) and 17β-estradiol (E2), and one synthetic estrogen, 17α-ethynylestradiol (EE2), in river sediments at the sub-ng/g level. This procedure includes microwave-assisted solvent extraction (MASE), solid-phase extraction and high-performance liquid chromatography–tandem mass spectrometry (LC–MS/MS) with electrospray ionisation. Using sediments spiked with the three estrogens at 10 ng/g wet weight, efficient extraction (>92%) of all the three analytes was achieved by MASE, and whole-procedure recoveries ranged from 82 to 98%. Optimisation of the LC separation allowed for substantial reduction of ionisation suppression in the electrospray source to a final level of <18% suppression. Time-of-flight mass spectrometry (TOF-MS) and MS/MS were compared for the analysis of sediment extracts, with the latter technique proving to be the most selective. The method detection limits achieved by LC–MS/MS were 15, 30 and 40 pg/g for E1, E2 and EE2, respectively, which were 13-fold lower than those obtained by LC–TOF-MS. Analysis of river sediments collected from the River Ouse, UK, showed the presence of the natural estrogens at sub-ng/g level. E1 levels ranged from 0.40 ng/g (dry weight) to 3.30 ng/g while E2 levels ranged from <0.03 to 1.20 ng/g and EE2 was never detected (<0.04 ng/g).

Introduction

There is an increasing body of evidence demonstrating the occurrence of estrogenic substances in aquatic systems impacted by anthropogenic activities. Exposure of freshwater or estuarine fish to these chemicals may result in the alteration of their sexual function, particularly in ecosystems receiving high levels of poorly diluted wastewater treatment plant (WWTP) effluents [1], [2], [3], [4]. Natural estrogens such as estrone (E1) and 17β-estradiol (E2), along with the synthetic estrogen 17α-ethynylestradiol (EE2), contribute to a large extent to the estrogenicity of WWTP effluents [5], [6], [7]. Despite the fact that estrogens are relatively weak sorbates [8], [9], [10], [11], a number of field studies have demonstrated that sediments could act as a sink for these compounds in riverine [12], [13], [14], [15], [16], estuarine [17], and marine [18], [19] environments, with concentrations up to 1000 times higher in bed sediments than in the overlying water column [16]. In these studies, levels of steroidal estrogens in bed sediments were reported to range from <0.12 to 22.80 ng/g in rivers [12], [13], [14], [15], [16], from <0.05 to 2.52 ng/g in estuaries [17] and from <0.05 to 3.6 ng/g in coastal marine areas [18], [19]. The accurate measurement of estrogen concentrations in sediments is fundamental to a number of investigations, e.g. for estimating their role in the removal of steroidal estrogens from the water column [8], for determining their persistence and risk of permeation to ground waters [20] and to determine the risk to benthic organisms from exposure to estrogens for [21]. Furthermore, in case of sediment resuspension and estrogen remobilization, sediments may act as a secondary source of exposure to aquatic organisms living in the water column.

Current analytical methods for sediment-associated estrogens include gas chromatography–mass spectrometry (GC–MS) where the limits of detection (LODs) have been reported to range from 0.02 to 0.4 ng/g for GC–MS [12], [14], [19], and 0.06–0.1 ng/g for tandem GC–MS/MS [15], [22]. However, high-performance liquid chromatography–mass spectrometry (LC–MS) methods are favoured over GC–MS methods as they do not require a derivatisation step prior to analysis and the ion sources used in LC–MS are less prone to fouling. Reported LODs for LC–MS or LC–MS/MS analyses of sediment-associated estrogens range between 0.02 and 1.0 ng/g [13], [16], [18] and, for LC–time-of-flight mass spectrometry (LC–TOF-MS) analyses, between 0.03 and 0.04 ng/g [17]. However, LC–MS methods often require extensive clean-up steps to remove interferences arising from the matrix that can result in ion suppression of the target analytes. In order to detect estrogens at the ng/g level in sediments, most analytical methods include a number of solid-phase extraction (SPE) steps alongside more time-consuming purification steps such as gel permeation chromatography [18], [22], normal-phase preparative chromatography or immunoaffinity cleanup [17].

The aim of this work was to develop a simple and reliable method for the determination of E1, E2 and EE2 in surface sediment samples at sub-ng/g level using LC–MS techniques. The developed method is based on MASE, SPE and LC–MS with an electrospray (ESI) interface. Several key points such as MASE conditions, sequential SPE steps and HPLC separation were optimised and are discussed. Whilst LC–MS/MS is now widely used for the determination of organic contaminants at trace levels in the environment, LC–TOF-MS is still scarcely used, even though it is considered to be a promising technique for the study of emerging contaminants [17], [23]. Therefore, the relative selectivity and sensitivity of HPLC–TOF and LC–MS/MS techniques for estrogen analysis in spiked sediments were compared. The optimised analytical procedure was then applied to the quantification of estrogens in sediment samples collected from the River Ouse, Sussex, UK.

Section snippets

Chemicals

Estrogen standards were purchased from Sigma–Aldrich (Gillingham, UK): E1, purity >99%), E2 and EE2 (purity >98%). [2,4,16,16-2H4]E1 (E1-d4), [2,4,16,16-2H4]E2 (E2-d4) and [2,4,16,16-2H4]EE2 (EE2-d4) (isotope purity 96%, chemical purity >98%) were used as surrogates and were obtained from Cambridge Isotope Labs. (Andover, MA, USA). All solvents were of LC-grade, purchased from Rathburn (Walkerburn, UK). Estrogen solutions (stock solution: 1 mg/mL; working solution: 0.1 μg/L) were prepared in

Choice of microwave-assisted solvent extraction conditions

Since no reference material is available for estrogens in sediment, MASE recovery was determined from the analysis of spiked sediment. Approximately 1 g of wet sediment was spiked with estrogens (10 ng of each standard), then well mixed with a spatula and left to stand for 2 h. These samples were then subjected to MASE, deuterated surrogates were added (5 ng each) and extracts were purified by SPE. Spiked samples were incubated for 2 h prior to extraction as significant sorption of estrogens to

Conclusions

A simple analytical method based on MASE, SPE and LC–MS/MS has been developed for the determination of estrogens in river sediment at pg/g level. Analysing estrogens at such low levels in sediment is challenging because of matrix effects and interferences. Sample preparation, cleanup and chromatographic separation had to be optimized to lower detection limits and reduce matrix-induced ion suppression in the electrospray source. Without optimization of the LC separation, dramatic signal

Acknowledgments

This work was funded by the European Regional Development Fund INTERREG IIIA programme (project no 162/025/266). We are grateful to Mr. M. Andrews who assisted with sediment sampling and analyses.

References (40)

  • G.-G. Ying et al.

    Water Res.

    (2003)
  • R. Liu et al.

    J. Chromatogr. A

    (2004)
  • T. Isobe et al.

    Environ. Poll.

    (2006)
  • O. Braga et al.

    Chemosphere

    (2005)
  • G.-G. Ying et al.

    Environ. Int.

    (2002)
  • K.M. Lai et al.

    Sci. Total Environ.

    (2002)
  • I. Ferrer et al.

    Trends Anal. Chem.

    (2003)
  • M. Kuster et al.

    Trends Anal. Chem.

    (2004)
  • X. Peng et al.

    J. Chromatogr. A

    (2006)
  • R. Gomes et al.

    Trends Anal. Chem.

    (2004)
  • J.-P. Antignac et al.

    Anal. Chim. Acta

    (2005)
  • I.-C. Beck et al.

    J. Chromatogr. A

    (2005)
  • T. Reemtsma

    Trends Anal. Chem.

    (2001)
  • R. Kitamura et al.

    J. Chromatogr. B

    (2001)
  • C.P. Czajka et al.

    Sci. Total Environ.

    (2006)
  • S. Jobling et al.

    Environ. Sci. Technol.

    (1998)
  • S. Jobling et al.

    Biol. Reprod.

    (2002)
  • L.D. Arcand-Hoy et al.

    Environ. Toxicol. Chem.

    (1998)
  • K.E. Liney et al.

    Environ. Health Perspect.

    (2005)
  • C. Desbrow et al.

    Environ. Sci. Technol.

    (1998)
  • Cited by (113)

    • Engineered microbes as effective tools for the remediation of polyaromatic aromatic hydrocarbons and heavy metals

      2022, Chemosphere
      Citation Excerpt :

      As a result, accurate and sensitive analytical procedures are required. Water samples are routinely used in studies documenting pesticide or pharmaceutical quantities in the environment, whereas sediment samples are rarely used (Labadie and Hill, 2007). Furthermore, the approaches given here are limited to a specific examination of a few chemical families of contaminants (Fernandez et al., 2001).

    View all citing articles on Scopus
    View full text