Non-targeted discovery of class-distinguishing metabolites in Argentinian pacu fish by comprehensive two-dimensional gas chromatography with principal component analysis

Abstract

Pacu fish are an important source of revenue and food supply in Argentina, however the nutritional effects of raising these fish in an integrated rice-fish farming environment on the fish metabolome are still unclear. To investigate this issue, the discovery of class-distinguishing compounds between farmed and control pacu fish samples was performed by comprehensive two-dimensional gas chromatography coupled with time-of-flight mass spectrometry (GC × GC-TOFMS) followed by principal component analysis (PCA). Pacu fish muscle tissue samples were extracted and derivatized for a total of 10 farmed and 10 control fish prior to GC × GC-TOFMS analysis. The total ion current (TIC) and a diagnostic ion for sugars, mass channel (m/z) 217, data for each fish were input to PCA for the non-targeted discovery of metabolites that differentiated between the control and farmed samples. The scores revealed that the classes (farmed and control fish) separated best on PC2, with most of the variance captured on PC1 accounting for the within class variance when both the TIC and m/z 217 were used. The loadings were examined to retrieve the 50 most highly loaded analytes on PC1 and PC2 using the TIC and m/z 217 (a compiled list of the 20 most highly loaded variables on each PC for the TIC and m/z 217), which were subsequently identified and concentration ratios, [Farmed]/[Control], were calculated to determine how farming fish with rice affects metabolite concentrations in pacu fish muscle tissue. A majority of analytes are downregulated in the farmed pacu fish (39/50) with 30 of these passing a p-value threshold at the 95% confidence level while of the remaining 11 analytes that were upregulated, 4 analytes passed a p-value threshold (34 analytes total passing the p-value threshold). Many of these significant analytes were highly loaded on PC2 and are involved in important metabolic processes including protein catabolism or nucleotide metabolism and tissue repair. These PCA results provide meaningful insight into the effects of farming Argentinian pacu fish with rice fields on the fish metabolome.