Evaluation of sample preparation methods for NMR-based metabolomics of Atlantic Salmon (Salmo salar) ovarian fluid

Authors: FABIO, CASU - National Institute Of Standards And Technology; BAYLESS, AMANDA - National Institute Of Standards And Technology; Peterson, Brian; HAMLIN, HEATHER - University Of Maine; BOGGS, ASHLEY - National Institute Of Standards And Technology; SCHOCK, TRACEY - National Institute Of Standards And Technology

Submitted to: Fish Physiology and Biochemistry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/2/2026
Publication Date: 4/15/2026
Citation: Fabio, C., Bayless, A.L., Peterson, B.C., Hamlin, H., Boggs, A.S., Schock, T.B. 2026. Evaluation of sample preparation methods for NMR-based metabolomics of Atlantic Salmon (Salmo salar) ovarian fluid. Fish Physiology and Biochemistry. 52. Article 63. https://doi.org/10.1007/s10695-026-01678-0.
DOI: https://doi.org/10.1007/s10695-026-01678-0

Interpretive Summary: Over the past 25 years, the hatch rate of farmed Atlantic salmon (Salmo salar) embryos in North America has seen a steady decline from approximately 80 to 90% in the early 2000's to less than 50% with the underlying cause(s) of this dramatic decline still uncertain (Legacki et al. 2023; Thayer and Hamlin 2016). Viable solutions are lacking for an issue that is affecting the production of one of the top aquaculture finfish species in states such as Maine, a leading producer of this species in the USA (Legacki et al. 2023; Thayer and Hamlin 2016). Selective Atlantic salmon breeding programs, such as the North American Atlantic salmon breeding program developed at the USDA-ARS National Cold Water Marine Aquaculture Center in Franklin, ME, USA, have historically assigned a “breeding value” to individual broodstock fish with the goal of making informed and effective decisions on the selection of breeding pairs for the next generation. Breeding values are estimates of an individual’s genetic merit for specific traits and are calculated using statistical models that incorporate information from the individual and its relatives (full-siblings, half-siblings), as well as pedigree data. By selecting individuals with high breeding values for commercially important traits, these programs aim to increase the overall output of the production system and improve the overall performance of farmed salmon populations. Although breeding values at different salmon producers are based on proprietary formulas, growth parameters (e.g., carcass weight) are heavily weighted factors in these calculations. There are trade-offs between fish growth and fish reproduction and the relationship between a fish’s growth rate and its reproductive success can be quite complex; as a result, the best fish for growth are not necessarily the best for reproduction (Roff 1983; Folkvord et al. 2014). By incorporating measurements related to reproductive success (e.g., egg quality, fertilization success, embryo hatch rates, biomarkers) in breeding value calculations, breeding individuals with traits contributing to higher embryo survival can be identified and prioritized in selective breeding programs.

Technical Abstract: Average embryo survival in farmed Atlantic salmon (Salmo salar) has declined sharply from about 80% to below 50% in North America over the past two decades. The underlying causes remain unclear, and no effective solutions currently exist for this problem, posing a major challenge to the production of a leading aquaculture species in the United States. Traditional selective breeding programs typically assign “breeding values” to broodstock fish based on growth-related traits; however, incorporating reproductive success traits (e.g., embryo hatch rates, metabolic biomarkers) into breeding value calculations may improve the efficiency of broodstock selection and help identify individuals more likely to produce viable offspring. Ovarian fluid (OF), collected non-lethally during egg harvesting offers a valuable biological matrix for metabolomics-based biomarker discovery for fish reproduction. This study tested seven sample preparation methods for Nuclear Magnetic Resonance (NMR) metabolomics analysis of OF: a) Ultrafiltration; b) Protein Precipitation; c) Lyophilization; d) Dilution; e) Lyophilization-Ultrafiltration; f) Dilution-Ultrafiltration; g) Filtration-Concentration. The objectives were to confirm that OF contains measurable metabolites, and to establish a repeatable and practical OF sample preparation method for NMR metabolomic analysis. Among the methods tested, Filtration-Concentration produced the most consistent quality spectra with lower variability, higher signal-to-noise ratio, and ease of execution compared with alternative methods tested. Additionally, this study demonstrates that blood contamination that may accidentally occur during ovarian fluid collection can significantly alter OF metabolite profiles, potentially affecting downstream analyses and data interpretation. These findings provide a foundation for standardized NMR metabolomics in salmon broodstock reproductive research.