Lipid extraction alters amino acid composition and bulk, but not amino acid, carbon and nitrogen isotope values

This manuscript mainly seeks to understand the underlying mechanisms driving observed changes in bulk d15N values in fish tissue samples following common chemical extraction methods used to remove lipids for the purpose of alleviating bias in bulk 13C analyses. How to deal with lipids has been a topic of discussion and debate in the scientific literature for several decades and is an important topic given the known potential for lipids to bias bulk d13C values as well as the less well understood effects of chemical extractions on bulk d15N, which is commonly derived from the same sample. This manuscript advances this discussion by seeking to identify the processes causing such observed bulk d15N changes. As the author notes, understanding the mechanism(s) is important because the 2 mechanisms hypothesized in the literature (removal of nitrogenous waste products and removal of amino acids) lead to opposing recommendations on how to handle the bulk d15N analyses; the former provides support for using the d15N value derived from the chemically-extracted sample while the latter is more in line with running a separate non-extracted sample to get the most accurate d15N value. This study performed a common chemical lipid extraction (Bligh-Dyer) on fish muscle representing a wide range of lipid contents for comparison with non-extracted samples with regard to d13C, d15N, %C, %N, C:N, proximate composition, and amino acid-specific stable isotope values. Results confirmed the expected increase in d13C and decrease in C:N following lipid extraction as well as a 15N-enrichment post-extraction for bulk d15N (but no alteration of AA-specific stable isotope values). The bulk d15N alteration was linked to both removal of nitrogenous waste as well as amino acids. Results will be of use to ecologists using both bulk and compound-specific stable isotope analysis approaches and help to advance our understanding of this important aspect of stable isotope analysis with relevance to many ecological applications. While the experimental design was appropriate for rigorously assessing a wide range of impacts of chemical extractions on fish tissue samples, I think the broader ecological interpretation and discussion are less well supported and so would recommend a minor revision to edit and condense that component of the manuscript (see comments below).

General Comments

Limitations of Ecological Interpretation of Fish Stable Isotope Results: This study relies on small sample sizes (n=5 per species) from fish samples that were all obtained from fish markets. As such, in addition to small sample sizes, the samples lack most of the relevant information that would be needed to make broader ecological interpretations of these isotope data. For example, fork length and/or total weight data are lacking, as are sex and harvest date as well as more detailed and/or validated collection location information. For example, “Northwest Atlantic” is a rather broad geographic region used to characterize Atlantic herring origin that could apply to a variety of habitats. I would suggest removing much of the ecological interpretation of these results from the Discussion and maintaining the focus on the methodological comparisons, which are the strength of the study. The discussion of the “organic” farmed salmon sample could be retained as it reflects the appropriate level of sample information that a researcher would have available in trying to verify the diet characteristics of a supposedly organic farmed fish species.

Comparison with Existing Literature: It would be helpful to further compare the observed bulk d15N changes to what has been reported previously in the literature for marine fishes to put these results in the broader context of existing literature.

Line-by-Line Comments

Line 68 – Reword “, and so on”

Line 126 – After “interpretations”, add “of” or “for” or equivalent

Line 127 – Remove “isotope” before “common”

Line 174 – Why is “reference” written out rather than simply putting “41” as a superscript? Same question applies elsewhere in the manuscript (e.g., line 185). I am not sure if this is RCM convention.

Line 219 – Change “areas” to “area”

Line 240 – Why was such a small SD value applied here? Is discrimination really this uniform among essential AAs? Please confirm that this reference supports the mean and SD used here.

Line 274 – Remove “to” before “prior”

Line 871 (Fig. 5) - I found the dominance of “macroalgae” as a primary production source to bluefish to be unexpected and not entirely believable given their documented foraging on mostly zooplantivorous fishes. I would suggest removing the ecological data interpretations and instead maintaining focus on the methodological comparisons.

I feel that the author has done a good job of addressing my concerns and comments related to the original submission of this article. I think putting most of the ecological discussion in a separate online supplement is a nice way of including this information without distracting from the main methodological focus of the study. I would suggest removing “dermersal” from Line 94 of the Supplemental Material as this description really only applies to cod. Bluefish would be considered a pelagic predator. Otherwise, I have no further comments. I think this will make a nice contribution to the field of stable isotope ecology.