Correlations of Eisenia fetida metabolic responses to extractable phenanthrene concentrations through time Jennifer R. McKelvie, David M. Wolfe, Magda Celejewski, André J. Simpson, Myrna J. Simpson * Department of Physical and Environmental Sciences, University of Toronto,1265 Military Trail, Toronto, Ontario, Canada M1C 1A4 Metabolic responses of Eisenia fetida earthworms to phenanthrene exposure are better correlated to total phenanthrene concentrations than to cyclodextrin-extractable concentrations through time. article info Article history: Received 17 September 2009 Received in revised form 15 February 2010 Accepted 27 February 2010 Keywords: Metabolomics Bioavailability Earthworms Partial least squares fitting Polyaromatic hydrocarbons abstract Eisenia fetida earthworms were exposed to phenanthrene for thirty days to compare hydroxypropyl- b-cyclodextrin (HPCD) extraction of soil and 1 H NMR earthworm metabolomics as indicators of bioavailability. The phenanthrene 28-d LC 50 value was 750 mg/kg (632e891, 95% confidence intervals) for the peat soil tested. The initial phenanthrene concentration was 319 mg/kg, which biodegraded to 16 mg/kg within 15 days, at which time HPCD extraction suggested that phenanthrene was no longer bioavailable. Multivariate statistical analysis of 1 H NMR spectra for E. fetida tissue extracts indicated that phenanthrene exposed and control earthworms differed throughout the 30 day experiment despite the low phenanthrene concentrations present after 15 days. This metabolic response was better correlated to total phenanthrene concentrations (Q 2 ¼ 0.59) than HPCD-extractable phenanthrene concentrations (Q 2 ¼ 0.46) suggesting that 1 H NMR metabolomics offers considerable promise as a novel, molecular- level method to directly monitor the bioavailability of contaminants to earthworms in the environment. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction Assessing the risk associated with environmental contamination is challenging as several studies have shown that chemical concen- trations in soil are poor predictors of toxicity (Alexander, 2000; Dean and Scott, 2004). This is because hydrophobic chemicals can sorb strongly to soil organic matter and become unavailable to soil organisms (Kelsey and Alexander, 1997; Alexander, 2000). However, earthworms ingest soil directly and can potentially take in contam- inants that are considered non-bioavailable to other organisms (Tang et al., 1998). Numerous soil extraction techniques, including hydroxypropyl-b-cyclodextrin (HPCD) extraction, have been devel- oped to estimate contaminant bioavailability. HPCD extraction has shown excellent correlations to hydrophobic chemical bioavailability to microorganisms (Reid et al., 2000; Doick et al., 2005). However, investigations testing correlations of HPCD extraction with contam- inant bioavailability to earthworms have shown contaminant- and species-specific results. Hickman and Reid (2005) found no rela- tionship between HPCD-extractable phenanthrene concentrations and phenanthrene accumulation by Lumbricus rubellus. Barthe and Pelletier (2007) found that HPCD-extractable concentrations were poorly correlated with low molecular weight polyaromatic hydro- carbon (PAH) accumulation but well correlated with high molecular weight PAH accumulation in Nereis virens and Lumbriculus variegatus. In contrast, Hartnik et al. (2008) found that HPCD extraction was able to estimate uptake of a-cypermethrin and chlorfenvinphos by Eisenia fetida. An additional complication associated with the use of soft- chemical extraction methods is that they are indirect methods that extract all of the contaminant not strongly bound to organic matter, rather than the bioavailable portion of the contaminant to which the organism is directly exposed (Semple et al., 2004). Methods that directly evaluate the organism itself rather than the soil can offer an improved assessment of contaminant bioavailability. One such method is metabolomics, which offers a promising new way to directly evaluate the bioavailability and toxicity of contami- nants in soil to earthworms (Bundy et al., 2007; Simpson and McKelvie, 2009; Brown et al., 2010). Metabolomics involves the measurement of changes in endogenous low molecular weight metabolites, including sugars and amino acids, in biological samples (tissues, cells, biofluids) due to an external stress, such as contami- nant exposure (Morrison et al., 2000; Bundy et al., 2009). This method can offer an advantage over measurements of contaminant concentrations in earthworm tissues, since tissue measurements can be impacted by differential excretion or biotransformation, which could underestimate total exposure (Guo et al., 2009). Additionally, metabolomics has the benefit of not only evaluating contaminant * Corresponding author. E-mail address: myrna.simpson@utoronto.ca (M.J. Simpson). Contents lists available at ScienceDirect Environmental Pollution journal homepage: www.elsevier.com/locate/envpol 0269-7491/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2010.02.022 Environmental Pollution 158 (2010) 2150e2157