Secondary exposure to second-generation anticoagulant rodenticides in European polecats (Mustela putorius) in Great Britain 2013 to 2016

Polecat carcasses were collected as part of a national monitoring survey carried out by the Vincent Wildlife Trust between December 2013 and March 2016. Sixty-eight carcasses were selected for rodenticide analysis, based on stratification by sex, location and collection date. Of the animals selected 56 were road traffic casualties; the remainder were found dead in fields, killed by dogs, trapped or the cause of death was unknown.

Collection date and location were recorded for all carcasses, which were stored frozen until necropsy examination at the National Museum of Scotland. The poor condition of the majority of the carcasses precluded assessment of clinical signs of exposure to rodenticides. Where carcass condition allowed, gross necropsy examination included recording of sex, head and body length (nose to tip of tail), mass and internal fat, scored on a five-point scale. Teeth (for ageing), whiskers (for stable isotope analysis) and liver tissue (for rodenticide analysis) were collected. Liver samples were frozen and transferred to the Centre for Ecology & Hydrology (CEH) for rodenticide analysis. Whiskers were prepared for analysis at the University of Exeter and analysed at Elemtex, UK and teeth were sent to Matson’s Lab LLC, USA for aging by analysis of cementum layers.

Rodenticide analysis: Concentrations of the five SGARs licensed for use in Great Britain (bromadiolone, difenacoum, brodifacoum, flocoumafen and difethialone) were determined in the polecat livers. A 0.25 g sub-sample of each liver was thawed, weighed accurately, ground and dried with anhydrous sodium sulphate. Labelled standard (d5- Bromadiolone, QMx) was added to each sample for quality control purposes and determination of analyte recovery. Each liver sub-sample was solvent-extracted and then cleaned-up using size exclusion chromatography followed by elution through solid-phase cartridges. Extraction was carried out twice with clean solvent. Each extraction involved vortex mixing of the sample with 1:1 v/v chloroform:acetone, mechanical shaking and centrifugation. The resultant supernatants from the two extraction runs were combined, solvent-exchanged into (1:1; v/v) chloroform:acetone, filtered (0.2 mm PTFE filter), subjected to a further solvent exchange into (1:23; v/v) acetone:DCM, filtered again, and cleaned-up by size-exclusion chromatography (Agilent 1200 HPLC). The cleaned extract was solvent-exchanged into 1:1:8; v/v. chloroform:acetone:acetonitrile and underwent a second clean-up using solid phase, methanol-washed, acetonitrile-activated extraction cartridges (ISOLUTE® SI 500 mg, 6 ml). The cartridges were eluted with the same solvent and the eluate exchanged for the mobile phase.

Liver SGAR residues were quantified by HPLC linked to a triple quadrupole mass spectrometer interfaced with an ion max source in Atmospheric Pressure Chemical Ionisation mode (APCI) with negative polarity. Full details of the operational parameters used are as given by Walker et al. (2017). All rodenticide standards (Dr Ehrenstorfer) were matrix matched and linear calibration curves were defined such that R2>0.99. A blank was run with each batch of unknowns. The mean method limit of detection (LOD) across batches for each compound was 0.0014 µg/g, except for difethialone which was 0.0022 µg/g. The mean (± SE) recovery for the total procedure was calculated from the labelled bromadiolone standard applied to each sample and was 68.0 ±2.1%. Liver SGAR concentrations were not recovery corrected and are expressed on a wet weight basis. Summed (Σ) SGAR liver concentrations in individual animals were calculated by summing the concentrations of the five different SGARs, a zero concentration being assigned to individual compounds that were not detected.

Stable isotope analysis: Whiskers were gently rinsed in distilled water and then freeze dried for 24 hours. One whisker per animal was cut into ~1mm segments using a scalpel, starting at the proximal end of the whisker. Consecutive segments were pooled until the summed sample weight was ~0.7 mg. The sample was enclosed in a tin cup and put into a tray for analysis. The next segment was prepared in the same way and the process was further repeated until either the whole whisker was used, or less than 0.2 mg was remaining. Samples were analysed on a Thermoquest EA1110 elemental analyser linked to a Europa Scientific 2020 isotope ratio mass spectrometer at Elemtex Ltd (Cornwall, UK) for δ15N and δ13C. δ15N and δ13C abundance are reported as δ-values and expressed as a per mil (‰) deviation from the international reference standards (PDB for carbon and AIR for nitrogen). Replicate analysis of standards (USGS 40, USGS 41 and an in-house bovine liver standard) yielded standard deviations of 0.05 – 0.29 for δ15N and 0.05 – 0.22 for δ13C.

Cementum ageing was undertaken by Matson’s Lab LLC (Manhattan, MT, USA) following a standard protocol. After decalcification in a weak hydrochloric acid solution, teeth were sectioned sagittally and mounted on glass slides. The sections were stained to allow visual differentiation of annual cementum growth layers. These layers (annuli) were examined microscopically for age estimation at time of death. Birth date was set to 1 May for the purpose of estimating age in months.

Calibration with historical dataset: Combination of new and historical data involved applying the limits of detection (LOD) for each compound from Shore et al. (2003), which were higher than those in the present study, to eliminate biases caused by changes in analytical sensitivity.

For full references please refer to the supporting document.

This dataset is included in the following collections

The Predatory Bird Monitoring Scheme (PBMS)