Effects of pre‐treatments on bulk stable isotope ratios in fish samples : a cautionary note for studies comparisons

RATIONALE : Stable isotope analysis (SIA) has revolutionised ecological studies over the past thirty years. One of the major fields where SIA is applied in the marine environment is related to the definition of ecosystem structure and function. With marine top predators such as sharks, SIA is a method of choice because tissue samples can be collected without the sacrifice of the animal. In elasmobranch research, the influence of compounds such as urea, trimethylamine oxide and lipids must be considered when using stable isotopes as ecological markers. Currently, a range of pre‐treatments are used to chemically remove these molecules prior to SIA. METHODS : This study investigated the impact of eleven commonly used pre‐treatments on carbon and nitrogen contents and C:N atomic ratio, as well as carbon and nitrogen SI ratios in elasmobranch tissues and its prey, measured by isotope ratio mass spectrometry. Three tissues were tested: blood and muscle of the ragged‐tooth shark Carcharias taurus, and muscle of one teleost species, the Cape knifejaw Oplegnathus conwayi. RESULTS : Compared with untreated samples, no trend or generalisation could be highlighted with the influence of pre‐treatments being species‐, tissue‐ and chemical‐element‐dependent. For the δ13C and δ15N values, differences among pre‐treatments were as high as 3‰, therefore potentially leading to erroneous ecological interpretation. CONCLUSIONS : The chemical properties of compounds (e.g. urea, lipids) combined with the polarity of solutions (e.g. water, solvents) explained a large part of these observations. This study highlights that pre‐treatments need to be considered especially when comparing carbon and nitrogen stable isotope ratios between studies. The results of this study provide a call to all stable isotope researchers to make a concerted effort to standardise pre‐treatment methods. This is crucial as global reviews are becoming increasingly more informative.Supplementary material: Table S1. Post‐hoc pairwise comparisons after ANOVAs for repeated analyses conducted on ragged‐tooth shark muscle samples. Significant results are highlighted. Table S2. Post‐hoc pairwise comparisons after ANOVAs for repeated analyses conducted on Cape knifejaw muscle samples. Significant results are highlighted. Table S3. Post‐hoc pairwise comparisons after ANOVAs for repeated analyses conducted on ragged‐tooth shark blood samples. Significant results are highlighted.Field work was funded by the African Coelacanth Ecosystem Programme of the National Research Foundation (Grant No. 81879).http://wileyonlinelibrary.com/journal/rcm2020-02-15hj2019Mammal Research Institut