Fish innate and adaptive immune responses have humoral and cellular components (1), which are produced/mature and activated/proliferate in primary lymphoid organs (thymus and head kidney) and secondary (spleen and mucosa-associated lymphoid tissue, MALT), respectively (2). The innate immune system involves non-clonal pattern recognition receptors (PRRs) such as C-type lectin-like receptors, Toll-like receptors, and NOD-like receptors, compared to the adaptive immune system that uses highly specific clonal receptors (T- and B-cell receptors) that are able to recognize antigens and their derived peptides (3). Pathogen Associated Molecular Patterns (PAMPs) bind to these PRRs and trigger signaling cascades that activate defensive mechanisms such as phagocytosis, proteolysis, synthesis of antimicrobial molecules and secretion of pro-inflammatory cytokines like il-1β, tnf-α, il-18 and il-6 (4).

Under inflammatory conditions, the innate immune system can induce several antimicrobial mechanisms, including depleting the micronutrients available to pathogens at the systemic and cellular level (5). This defense mechanism is called nutritional immunity and involves depleting micronutrients such as iron, manganese and zinc from the circulation by sequestering them within the cells (5). Nutritional immunity in fish has been described in Eleginops maclovinus (6, 7), Notothenia coriiceps (8), Notothenia rossii (8) and Salmo salar (9), being this latter the most important species in the Chilean aquaculture (10). Outbreaks of infectious diseases in S. salar farms cause substantial economic losses in the Chilean aquaculture industry every year. The most prevalent etiological agent of these outbreaks in Chile is the Gram-negative bacterium Piscirickettsia salmonis, which causes Piscirickettsiosis (11). P. salmonis cells are pleomorphic, coccoid in shape, usually found in pairs and have a diameter of approximately 0.5-1.5 μm (12). This organism is classified as a facultative intracellular bacteria because it can also grow in agar medium enriched with L-cysteine and iron (13–17), and can synthesize siderophores under limited iron conditions (18) requiring/using different sources of iron (9, 18, 19).

In marine fish, micronutrients such as iron, manganese and zinc are absorbed by the gills and the gastrointestinal tract, having catalytic, structural, physiological and regulatory functions (20). Therefore, it is not surprising that microorganisms have developed sophisticated mechanisms to sequester these micronutrients and use them to survive (5). Restriction of iron availability has been described as a resistance mechanism to infection against P. salmonis in fish such as E. maclovinus (6, 7) and S. salar (9, 21). Additionally, Pulgar et al. (9) indicate that families of S. salar resistant to infection with P. salmonis can decrease the iron content in the head kidney at 14 dpi, without changes in intracellular zinc levels. This research suggests that components of the innate immune system such as hamp and il-6 may be regulating the nutritional immunity (22, 23), making it effective and efficient in families of S. salar with low susceptibility to P. salmonis

Proteins involved in iron uptake in P. salmonis have already been identified (9, 18, 24), and those involved in zinc and manganese uptake have been identified for other bacteria (25–28). On the other hand, Martínez et al. (8) reported that LPS modulates the expression of iron-related immune genes in N. coriiceps and N. rossi but does not affect the plasma iron concentrations (8). This latter research suggests that nutritional immunity may not differentiate between live pathogens (that need micronutrients to establish an infection) and PAMPs (that do not). Even, if nutritional immunity is activated by PAMPs, it is unknown which PAMPs from P. salmonis would trigger this immune response. Therefore, the objective of this study was to evaluate the transcriptional activation of markers involved in nutritional immunity using the SHK-1 cell line stimulated with different PAMPs isolated from P. salmonis.

Before this study, it was not known whether nutritional immune responses discriminate between living pathogens (which require nutrients) and non-living pathogens (which do not) or what types of P. salmonis PAMPs trigger nutritional immune responses in SHK-1 cells. We observed changes in the transcription of several nutritional immunity associated genes, suggesting that OMVs, TP, and LPS isolated from P. salmonis activate nutritional immune responses in SHK-1 cells. The SHK-1 cell line was created using head kidney cells from Atlantic salmon and has macrophage-like characteristics that make it a good model to evaluate the immune response of fish in vitro, especially if we consider that P. salmonis can modulates the intracellular environment in SHK-1 in the early (vacuolization) and late (propagation) stage of infection to facilitate its survival and propagation (37).

Pulgar et al. (9) showed that S. salar families resistant to P. salmonis infection had less iron in the head kidney at 14 dpi, but the intracellular zinc levels did not change. We observed increased transcription of zip8 and zip14, which are involved in the uptake of Zn²⁺, Mn²⁺, Fe²⁺ and HSeO_-3, in all three of our experimental conditions, strongly suggesting that P. salmonis OMVs, TP and LPS can alter homeostatic regulation of this micronutrients in SHK-1 cells and increase their uptake into cells. Aditionally, Nebert et al. (38) reported different functions associated to the intracellular increment of these micronutrients by zip8, which were related to the immune response, catabolism, oxidative stress, protein glycosylation and cell morphology/proliferation/migration (38). Therefore, the increased zip8 we observed could suggests an increment in these cellular functions; however, further studies are needed to investigate this directly.

Divalent metal transporter (dmt1) is a phagosomal membrane protein that transports iron from the phagosome to the cytosol (39). In our study, the transcription of this transporter was up-regulated in cell line SHK-1 exposed to OMVs and TP, but was not modulated in cells stimulated with LPS. We expected that the three PAMPs used in this study modulated the transcription of this transporter as an antimicrobial response mechanism, since a previous study reported that families of S. salar with high susceptibility to P. salmonis increases dmt1 mRNAs in the head kidney to 14 dpi (9). The lack of effects with LPS could be due to a microbial strategy to prevent divalent metals escaping from the phagosomal space to the cytosol, but this should be further investigated for P. salmonis in in vitro infection assays. When Francisella [bacterium similar in terms of pathogenesis to P. salmonis (40)] infects, the host cell can induce the synthesis of dmt1 and ferroportin (ireg1), exporting the iron from the phagosome to the cytosol and from the cytosol to the extracellular. However, Francisella counteracts this response by inducing a high synthesis of hepcidin (hamp) in the host cell, which binds to ireg-1 and causes its degradation, increasing the iron available in the cytosol for its replication (39).

The modulation of iron homeostasis in macrophages is dependent on the type of pathogen, since Francisella uses an active iron acquisition system that is critical for its intracellular proliferation. This system involves the tfr1 pathway with induction of steap3, irp1, irp2 and dmt1, while Salmonella enterica subsp. enterica serovar Typhimurium does not require the expression of these markers for successful intracellular survival (41). We observed increased irp1, irp2 and tfr1 transcription in all three experimental conditions, suggesting P. salmonis acquires iron like Francisella does. Our results are consistent with those of Martínez et al. (8), who observed increased tfr1, ireg1 and hamp transcription in the head kidney of N. coriiceps exposed to LPS. These authors suggest LPS triggers iron associated nutritional immune responses; however, they did not observe any changes in plasma iron concentrations or evaluate tissue iron concentrations.

Iron output is regulated by hepcidin and ferroportin, the union of both proteins causes internalization and degradation of the latter (22). As mentioned previously, in cells infected with Francisella the production of ireg1 is induced, while Francisella triggers the synthesis of hamp to prevent iron from leaving the extracellular space (39). In our study, the transcription of ireg1 and hamp was modulated in the three experimental conditions. This suggests that the host cell stimulated with OMVs, TP and LPS regulates the synthesis of ireg1 and that the PAMPs of P. salmonis, as in Francisella, are sufficient to trigger the transcription of hamp. The results are consistent with the observations of Pulgar et al. (9), who reported increased ireg1 transcription in the head kidney of S. salar families with low susceptibility to P. salmonis at 14 dpi (9) and for Martínez et al. (7, 8), who reported increased ireg1 and hamp expression in E. maclovinus and N. coriiceps injected intraperitoneally with two strains of P. salmonis and LPS, respectively.

Pro-inflammatory cytokines like il-6 stimulate hamp transcription, triggering and enhancing the hypoferremic response to inflammation (42). We observed increases and decreases in il-6 transcription in all three experimental conditions, but they did not follow any obvious pattern like those we saw in hepcidin transcription. This can be explained because the levels of mRNAs do not always reflect the levels of protein that are synthesized in the cell and it is likely that in future studies, we will be able to quantify il-6 in plasma and tissue to have a clearer idea of its relationship with the nutritional immunity. However, il-6 transcription increases in the kidney and spleen of zebrafish stimulated with OMVs from P. salmonis, suggesting OMVs could be candidates for the development of vaccines that provide a protective effect against P. salmonis (43). Additionally, il-6 receptor transcription increases in the head kidney of N. coriiceps stimulated with LPS, suggesting a relationship between il6rβ, il-6 and hamp (8). Furthermore, others authors have evaluated the nutritional immune responses in S. salar (9), and E. maclovinus (7) challenged with P. salmonis, but they did not measure il-6 transcription; therefore, it is difficult to compare our results.

irp1/2 proteins modulates iron metabolism in vertebrates by regulating the translation of genes involved in the homeostasis of this micronutrients (44). Under conditions of iron deficiency, irp1/2 bind to the IRE located at the 5’UTR of ft-h, ft-l and ireg1 mRNAs, repressing their translation, while their binding to the IREs of the 3’UTR stabilize the tfr and dmt1 transcripts, preventing their degradation and increasing iron uptake. On the other hand, under overload conditions irp1/2 decrease their binding activity to the IREs at the 3’UTR and 5’UTR of tfr/dmt1 and ft-h/ft-l/ireg1, respectively. This leads to a destabilization of the tfr/dmt1 mRNAs and an efficient translation of the ft-h, ft-l and ireg1 mRNAs, favoring iron sequestration during uptake (44). In our study, the expression profile of ft-h and ft-m was similar, with an up-regulation in the first minutes of the challenge (OMVs, TP and LPS) and a down-regulation at 120-min in the TP treatments and LPS, suggesting that PAMPs from P. salmonis could induce iron storage within the SHK-1 cell line. Our results are consistent with those of Naves et al. (45), who reported a decreased and increased ferritin expression in low and high iron conditions, correspondingly. Additionally, others have reported increased ft-h, and ft-l/m transcription in N. coriiceps (8) and E. maclovinus (6, 7) exposed to LPS and P. salmonis, respectively. However, Pulgar et al. (9) observed increased ft-l expression and decreased intracellular iron content in the head kidney of S. salar families with low susceptibility to P. salmonis at 14 dpi, suggesting increased ft-l transcription does not necessarily result in iron storage.

This study reveals for the first time the temporal expression profiles of markers involved in nutritional immunity in the SHK-1 cell line stimulated with different PAMPs from P. salmonis. The results strongly suggest that the three PAMPs of P. salmonis used in this study are capable of modulating nutritional immunity in SHK-1, inducing the transcription of immune markers involved in the transport (zip8, zip14, ireg1 and dmt1), uptake (tfr1), storage (ft-h and ft-m) and regulation (il-6, hamp, irp1 and irp2) of micronutrients such as iron, manganese and zinc.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: NCBI, accession IDs: XM_014173032.1, NM_001140849.1, BT045467.1, BT072056.1, XM_014188394.1, XM_014143440.1, XM_014143031.1, AJ427629.1.

DM: Writing – original draft, experimental design, sampling, sample analysis, wrote the initial MS version, revision of final MS. CO: Writing – original draft, experimental design, wrote the initial MS version, revision of the final MS. NS and JC: Sample analysis. RO-S: Samples analysis, writing – original draft, revision of the final MS. RE: Maintenance of P. salmonis LF-89. LV-C: Writing – original draft, experimental design, revision of the final MS. AR: Writing – original draft, experimental design, revision of the final MS. All authors contributed to the article and approved the submitted version.

This work was financially supported by Fondecyt-Postdoctoral N° 3200418, Fondap-Ideal Grant N° 15150003, Fondecyt-Iniciación N° 11180994, Fondap-Incar N° 15110027 and Vicerrectoría de Investigación, Desarrollo y Creación Artística (VIDCA) of the Universidad Austral de Chile.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.