Prediction of Fatty Acids in the Rainbow Trout Oncorhynchus mykiss: A Raman Scattering Spectroscopy Approach

Poly-unsaturated fatty acids (PUFAs) are important to improve animal and Human development and health. A potential to improve relative PUFAs composition by genetic selection was previously reported in fishes. However, analytical methods, such as gas chromatography (GC), have some disadvantages such as the use of consumables and solvent and their too high cost preventing large scale phenotyping as needed in breeding programs. Thus, it is crucial to develop alternative technologies to overcome these drawbacks. In this study, Raman spectroscopy has been used in order to determine the chemical composition of rainbow trout adipocytes in a non-destructive manner, and to correlate with the results obtained by gas chromatography. Two groups of trout have been created based to their diets: marine-based and plant-based. Then, visceral adipose tissues were collected and analysed by GC and Raman micro-spectroscopy. Two regression methods were used to establish calibration models from the GC and spectral data: ridge and partial least square (PLS). GC results confirmed that α-linolenic acid (ALA) is more present in the plant-based diet group, whereas eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA) and linoleic acid (LA) are more present in the marine-based diet group. By using both ridge and PLS regression methods on GC and spectral data, R2 showed high values for ALA, EPA, DHA and LA. Thus, this methodology shows that good correlation coefficients can be obtained to predict PUFAs, and calibration models can be used to predict PUFAs contents for large scale and high throughput phenotyping in rainbow trout.