Effects of eicosapentaneoic acid on innate immune responses in an Atlantic salmon kidney cell line in vitro

Abstract

Studies of the interplay between metabolism and immunity, known as immunometabolism, is steadily transforming immunological research into new understandings of how environmental cues like diet are affecting innate and adaptive immune responses. The aim of this study was to explore antiviral transcriptomic responses under various levels of polyunsaturated fatty acid. Atlantic salmon kidney cells (ASK cell line) were incubated for one week in different levels of the unsaturated n-3 eicosapentaneoic acid (EPA) resulting in cellular levels ranging from 2-20% of total fatty acid. These cells were then stimulated with the viral mimic and interferon inducer poly I:C (30 ug/ml) for 24 hours before total RNA was isolated and sequenced for transcriptomic analyses. Up to 200 uM EPA had no detrimental effects on cell viability and induced very few transcriptional changes in these cells. However, in combination with poly I:C, our results shows that the level of EPA in the cellular membranes exert profound dose dependent effects of the transcriptional profiles induced by this treatment. Metabolic pathways like autophagy, apelin and VEGF signaling were attenuated by EPA whereas transcripts related to fatty acid metabolism, ferroptosis and the PPAR signaling pathways were upregulated. These results suggests that innate antiviral responses are heavily influenced by the fatty acid profile of salmonid cells and constitute another example of the strong linkage between general metabolic pathways and inflammatory responses.

Fig 1. Total levels of EPA (in % of total FA) in medium and ASK cells at various levels of added BSA-EPA (0–200 uM) in the cell medium after one week of incubation.

Fig 2. QPCR analysis of relative expression of interferon inducible transcripts in ASK cells treated with 200 uM EPA for 7 days (Control_EPA) or 24 hours with 30 ug/ml polyI:C (Control_pIC).

Expression was calculated relative to Ef1a and 18S as described. Boxplot shows median, max and min values (n = 6).

Fig 3. Principal component loadings for PC1 and PC2 of the most variable transcripts in the RNAseq dataset from Atlantic salmon kidney cells stimulated with poly I:C.

Fig 4. The 50 most up- and downregulated transcripts (meanlog2FC across 4 doses of EPA relative to no EPA added) in ASK cells after poly I:C stimulation.

Fig 5. Venndiagram showing overlap of significant DEGs in the various experimental groups (control are just expressed genes).

Fig 6. Enriched KEGG pathways in ASK cells incubated in EPA and stimulated with poly I:C.

Fig 7. Plot of raw counts and best fit line for the transcripts most positively correlated to EPA concentration in poly I:C stimulated ASK cells.

Fig 8. Plot of raw counts and best fit line for the transcripts most negatively correlated to EPA concentration in poly I:C stimulated ASK cells.