Silybin Interferes with the Intracellular Replication of Piscirickettsia salmonis in SHK-1 Cells and Confers Protection in Salmo salar

TL;DR

Silybin reduces the replication of Piscirickettsia salmonis in fish cells and protects salmon from infection, potentially offering a new treatment option.

Contribution

Silybin's novel ability to interfere with P. salmonis replication in SHK-1 cells and confer protection in salmon is demonstrated.

Findings

Silybin modulates immune-related genes and reduces ROS in SHK-1 cells infected with P. salmonis.

Silybin interferes with intracellular replication of P. salmonis after 72 hours, not adherence or internalization.

Silybin provides protection in Salmo salar against P. salmonis without immune response stimulation.

Abstract

Background/Objectives: The salmon industry plays an important role in the Chilean economy, positioning the country as the second-largest producer of salmonids worldwide after Norway. However, this rapid growth has led to an increase in outbreaks of infectious diseases, which cause significant economic losses to the industry. The pathogen that most affects the salmon industry is the bacterium Piscirickettsia salmonis, accounting for 43.1% of infection-related deaths. In the search for new treatment alternatives against P. salmonis, we have previously reported that the effect of co-incubating silybin at sub-IC50 concentrations decreases the intracellular presence of P. salmonis in SHK-1 cells. Methods: This article evaluates the effect of silybin on the immune response and oxidative stress of SHK-1 cells infected with P. salmonis, as well as the reduction in intracellular bacterial replication during the first 72 h of infection. Furthermore, we assess the ability of silybin administration to modulate the immune response in S. salar and protect against P. salmonis infection. Results: The results show that co-incubation of silybin during infection in SHK-1 cells modulates the expression levels of the genes gsh-px, cat, tnf-α, and il-1β and also decreases the levels of intracellular ROS generated by the infection. Furthermore, the mechanism of action of silybin in SHK-1 cells is related to interference with the intracellular replication of P. salmonis after 72 h of infection and not to adherence or internalization of the bacteria. Finally, silybin is able to generate protection in S. salar infected with P. salmonis independently of stimulation of the immune response. Conclusions: In conclusion, silybin administration may be an effective treatment against P. salmonis in salmonids; however, further studies are needed to clarify the mechanism of action.

Figures

Full text

1. Introduction

Salmon farming is one of Chile’s most important productive sectors [1], positioning the country as the world’s second-largest producer of salmon [2]. The most produced species is Atlantic salmon (Salmo salar), with an accumulated harvest of 498,600 tons by August 2025, representing a 9.1% increase compared to the same period in 2024 and accounting for 52.3% of total aquaculture production [3]. Despite this growth, production is significantly affected by high mortality rates caused by multiple factors. In S. salar, infectious diseases represent the most relevant cause of mortality (22.1%), with piscirickettsiosis (SRS), caused by Piscirickettsia salmonis, accounting for 43.1% of infectious-related deaths [4]. This bacterium has affected the Chilean salmon farming industry for more than 30 years and remains its most relevant pathogen [5].

Currently, two main strategies are used in Chile to control P. salmonis: vaccination and antibiotic administration. Several vaccines have been available for years, including monovalent attenuated vaccines, pentavalent inactivated vaccines, and bacterins [6,7]. However, despite their widespread use, these vaccines have not been effective in eradicating the pathogen and have only delayed the onset of outbreaks [8,9]. Although the reasons for this limited effectiveness remain unclear, both intrinsic and extrinsic factors have been proposed, including vaccine formulation, vaccination procedures, coinfection with other pathogens such as Caligus rogercresseyi, and host genetic variability [6,8,10,11]. Consequently, antibiotics continue to be widely used to control SRS. In 2024, approximately 350 tons of antibiotics were administered in Chilean salmon farming, of which 322 tons were used exclusively to control P. salmonis [12]. This usage is markedly higher than that reported in Norway, where approximately 709 kg were used during the same period [12,13]. The intensive use of antibiotics generates environmental concerns, affects fish health, and ultimately compromises product quality [14,15,16].

Given the economic impact of P. salmonis and the urgent need to reduce antibiotic use, alternative treatments have been increasingly explored. Among these, natural compounds and plant-derived extracts have been extensively studied due to their ability to reduce bacterial replication and mortality in salmonids. For example, fucoidans and labdane diterpenes have been shown to stimulate the immune response in vitro, increasing il-12 and ifn-i levels, while in vivo administration resulted in relative percent survival (RPS) values of 62.3% and 57.1% after 60 and 80 days post-infection, respectively [17]. Similarly, food additives formulated with labdane diterpenes demonstrated immunostimulatory effects both in vitro and in vivo, increasing il-12, tnf-α, mhc-i, mhc-ii, il-10, and ifn-γ levels and generating an RPS of 48% [18]. In addition, saponins extracted from Quillaja saponaria have been reported to reduce P. salmonis replication in vitro [19], increase il-12 levels, and promote phagosome–lysosome fusion, facilitating bacterial degradation [20]. In vivo studies further demonstrated reduced mortality in fish fed with these extracts [21].

In this context, our laboratory has investigated the effect of natural compounds, particularly polyphenols, on reducing P. salmonis replication. Specifically, we evaluated the effect of co-incubating silybin at sub-IC50 concentrations for 24 h in SHK-1 cells during P. salmonis infection. Under these conditions, silybin reduced the intracellular bacterial load after 7 and 14 days of infection [22]. Notably, this effect was lost when silybin was pre-incubated or post-incubated relative to infection, suggesting that its mechanism of action may involve interactions between the bacterium, the host cell, and the compound itself. However, the underlying mechanism responsible for this effect, as well as whether the reduction in intracellular bacterial load translates into protection in S. salar challenged with P. salmonis, remains unclear.

The present study addresses these gaps by evaluating the effects of silybin during P. salmonis infection in SHK-1 cells and in S. salar. Our results show that co-incubation with 68 µg/mL silybin modulates the transcript levels of immune markers such as il-1β and tnf-α, as well as oxidative stress markers including glutathione peroxidase (gsh-px) and manganese superoxide dismutase (mnsod), and reduces intracellular reactive oxygen species (ROS) levels. In contrast, silybin does not affect bacterial adhesion or internalization, although a decrease in intracellular bacterial load is observed over time, suggesting an intracellular mechanism of action. Furthermore, dietary administration of silybin in S. salar increased survival following P. salmonis infection, reaching a survival rate of 50%, compared to 16.6% in the control group.

2. Results

2.1. Effect of Silybin on SHK-1 Cell Immune Response Markers During P. salmonis Infection

In the search for new treatments to combat infections caused by P. salmonis in salmonids, we have recently reported that co-incubation of silybin at sub-IC_50_ concentrations during P. salmonis infection in SHK-1 cells reduces the number of bacteria inside the cells after 7 and 14 days of infection [22]. This effect was only observed when the compound was incubated together with the bacteria for the 24 h in which the infection took place, whereas pre-incubation of the compound had no effect and post-incubation had a slight effect on decreasing the number of bacteria after 7 and 14 days of infection. However, the mechanisms by which this compound is able to reduce the number of bacteria inside SHK-1 cells have not been evaluated. One of the possible mechanisms is the stimulation of the immune response, which could enhance the defense of these cells during infection. To evaluate this hypothesis, the modulation of four immune markers associated with the response to bacterial infections was measured. Assessments were performed 2 h post-infection, as an early response; at 24 h, the maximum incubation time for silybin and P. salmonis; and at 48 h, corresponding to 24 h after the elimination of the compound and extracellular bacteria. The evaluation of il-1β transcript levels at 2 h of experimentation showed that cells infected with P. salmonis (PsV) increased approximately 16-fold compared to control cells. On the other hand, a much greater increase of approximately 512-fold was observed in cells incubated with heat-killed P. salmonis (PsM), whereas cells co-incubated with P. salmonis + 68 µg/mL of silybin (PsV + S) increased approximately 8-fold (Figure 1a). At 24 h after infection, il-1β transcription levels were observed to increase approximately 64-fold in cells infected with P. salmonis (PsV), approximately 8-fold in cells infected with heat-killed P. salmonis (PsM), and approximately 256-fold in cells co-incubated with P. salmonis and silybin (PsV + S) (Figure 1a). After 48 h of experimentation, only an increase of approximately 8-fold was observed in cells infected with dead P. salmonis (PsM) (Figure 1a). In the case of tnf-α, after 2 h of experimentation, an increase in transcript levels of approximately 32-fold was observed, only in the cells infected with heat-killed P. salmonis (PsM) (Figure 1b). On the other hand, after 24 h of experimentation, an increase in tnf-α transcript levels of approximately 8-fold was observed in cells infected with P. salmonis (PsV), whereas a much larger increase was observed in cells incubated with P. salmonis + 68 µg/mL silybin (PsV + S), approximately 32-fold (Figure 1b). After 48 h of experimentation, only the cells infected with PsM increased the levels of the gene transcript by about 8-fold (Figure 1b). In the case of il-12 transcript levels, only a slight statistically significant difference was observed in cells infected with PsM at 2 h of experimentation, whereas at other times and treatments, no differences were observed with respect to control cells (Ctrl) (Figure 1c). Among the treatments, statistically significant differences were observed only at 24 h between PsV and PsM compared with PsV + S, whereas at 48 h, statistically significant differences were detected only between PsM and PsV + S (Figure 1c). Regarding tgf-β transcript levels, it was only possible to observe a considerable difference at 48 h of experimentation, in the cells that were infected with PsM, increasing about 16-fold (Figure 1d).

Cells co-infected with P. salmonis and silybin showed differences in the transcription levels of the cytokines analyzed, primarily at 24 h of experimentation. To determine the effect of silybin, the transcription levels of these cytokines were evaluated in cells incubated only with silybin. The results showed that cells incubated with 68 µg/mL of silybin (S) only exhibited an almost 4-fold increase in tnf-α transcription levels compared to control cells. However, the transcript levels of il-1β, il-12, and tgf-β did not show significant differences (Figure 2).

2.2. Effect of Silybin on SHK-1 Cell Oxidative Stress Markers During P. salmonis Infection

The modulation of transcript levels in genes associated with an oxidative stress response was also evaluated in SHK-1 cells infected with P. salmonis. The results showed that, in the case of the gsh-px gene, after 2 h of experimentation, cells infected with PsV decreased transcript levels of the gene by nearly 2-fold compared to control cells (Ctrl), whereas cells infected with PsM increased transcript levels by nearly 2-fold (Figure 3a). After 24 h of experimentation, cells infected with PsV, PsM, and PsV + S increased transcript levels by nearly 4-fold compared to control cells (Ctrl) (Figure 3a). Finally, after 48 h of experimentation, the cells incubated with PsV increased transcript levels by nearly 2-fold compared to the control cells (Ctrl), while the cells incubated with PsM and PsV + S increased transcript levels by nearly 4-fold (Figure 3a). The results of the cat gene transcription levels, which encode catalase, showed that at 2 h of experimentation, only cells infected with PsV + S exhibited a statistically significant decrease in gene transcription levels compared to control cells (Ctrl). At 24 h, cells infected with PsV showed a 2- to 4-fold decrease in transcription levels compared to the control, whereas cells infected with PsM showed an approximately 4-fold increase in gene transcription levels. In the case of cells infected with PsV + S, the decrease in gene transcription levels was similar to that observed in cells infected with PsV (Figure 3b). In the case of cusod, which encodes copper/zinc superoxide dismutase, no differences were observed in transcription levels for any treatment, at any of the times analyzed (Figure 3c). In mnsod, it was only possible to observe a difference in the gene transcript levels of close to 2-fold, in cells infected with PsV + S with respect to control cells and cells infected with PsV and PsM at 24 h of experimentation (Figure 3d).

Similar to what was observed with immune response markers, a greater effect of co-incubation with P. salmonis and 68 µg/mL of silybin was observed on oxidative stress markers after 24 h of experimentation. Therefore, the effect of 24 h of silybin incubation in SHK-1 cells on the transcription levels of oxidative stress markers was also evaluated. The results showed that transcription levels of the gsh-px gene increased almost 2-fold in cells treated with silybin compared to control cells, while transcription levels of the cat gene decreased 3- to 4-fold. No changes were observed in the transcription levels of the cusod and mnsod genes (Figure 4).

Intracellular ROS levels in SHK-1 cells during P. salmonis infection and co-incubation with silybin were evaluated during the first 24 h of infection. At 2 h post-infection, a statistically significant difference in relative fluorescence units (RFUs) per mg of protein per mL was observed between cells infected with PsV (543,079) and cells infected with PsM (398,169), PsV + S (375,987), and S (312,730). On the other hand, in the case of the control cells (470,784), only a difference was observed with respect to the cells treated with S (312,730) (Figure 5a). At 24 h after infection, an increase in RFUs was observed in cells infected with PsV (707,073), compared to control cells (572,716) and cells infected with PsM (398,585), PsV + S (459,600), and S (513,106). Moreover, cells infected with PsM (398,585) and with PsV + S (459,600) showed a decrease in RFU levels compared to control cells (572,716) (Figure 5b).

2.3. Internalization and Intracellular Replication of P. salmonis in SHK-1 Cells Co-Incubated with Silybin

Bacterial adhesion, internalization, and replication within cells were evaluated by detecting the glyA gene of P. salmonis during different incubation periods. The bacterial load of P. salmonis detected 2 h post-infection showed no statistically significant differences in the number of copies of the glyA gene between cells infected with PsV (249 ± 188), PsV + S (172 ± 56), and PsM (116 ± 42). Similar results were observed 24 h after infection, where there were no statistically significant differences in the number of copies of the glyA gene detected within cells infected with the different treatments: PsV (2298 ± 1443), PsV + S (3099 ± 2892), and PsM (4328 ± 1714). After 48 h of infection, statistically significant differences were observed in the number of copies of the glyA gene inside cells infected with PsV (8699 ± 2385) and PsM (3803 ± 1386). Finally, at 72 h of infection, a statistically significant difference was observed in the number of copies of the glyA gene inside SHK-1 cells, among all treatments, PsV (28,051 ± 116,88), PsV + S (4828 ± 3831), and PsM (250 ± 234) (Figure 6).

2.4. Evaluation of the Immunostimulatory Effect of Silybin in S. salar

The effect of administering silybin mixed with food for 10, 20 and 30 days on the immune response of S. salar was evaluated. The results showed that during 10 days of feeding, no differences were observed in the transcript levels of any of the genes analyzed, either in the anterior kidney or in the intestine, between the fish that received silybin (S) and the control fish (Ctrl) (Figure 7a). After 20 days of administration, slight but statistically significant differences (less than 1) were observed in the transcript levels of some analyzed genes between fish that received silybin (S) and control fish (Ctrl). In the intestine, a slight increase in tgf-β levels and a slight decrease in lysozyme and perforin levels were observed. In the case of the anterior kidney, a slight decrease in the transcript levels of ifn-γ, il-1β, tgf-β, and lysozyme was observed (Figure 7b). Finally, after 30 days, no statistically significant differences were observed (Figure 7c).

2.5. Evaluation of the Protective Effect of Silybin Administration in S. salar Against P. salmonis Infection

The ability of silybin (68 μg/g) to confer protection in S. salar against a challenge with P. salmonis was evaluated. The results showed that silybin administration generated protection, with 50% of the fish surviving, compared to only 16.6% in the challenge control group. The survival curve showed a statistically significant difference (p < 0.0001) according to the Mantel–Cox statistical analysis (Figure 8a). The bacterial load of P. salmonis in dead and surviving fish was quantified by detecting the 16S rRNA gene. The results showed that dead fish from the control group and those treated with silybin had similar bacterial loads in the intestine and kidney, with a number of copies of the 16S rRNA gene between 1 × 10^4^ and 1 × 10^5^ in 50 ng of DNA (Figure 8b,c). In the surviving fish, it was not possible to detect bacterial load in the intestine of the fish in the control group, whereas in the group of fish fed with silybin, in two of the nine fish, it was possible to detect the bacterial load of P. salmonis, with an average number of copies of the 16S rRNA gene of 1 × 10^3^ in 50 ng of DNA. No bacterial load of P. salmonis was detected in the kidney of the control group fish, whereas one of the nine fish fed silybin showed a bacterial load of P. salmonis of approximately 5 × 10^2^ copies of the 16S rRNA gene in 50 ng of DNA (Figure 8b,c).

3. Discussion

In the search for new treatments to control P. salmonis infections, we have previously studied the effect of two polyphenols on reducing the intracellular load of P. salmonis in SHK-1 cells. These compounds were used at sub-IC50 concentrations and co-incubated for 24 h of infection in SHK-1 cells. However, the mechanism of action by which these compounds decrease intracellular bacterial presence, as well as their ability to generate protection in S. salar against P. salmonis infection, has not yet been evaluated [22]. Therefore, understanding how these compounds interact with host cellular responses becomes essential.

In this context, different studies have evaluated the effect of P. salmonis on the modulation of the immune response of salmonids as a mechanism to colonize, replicate inside cells, and evade the different control mechanisms of their target cells [9]. Based on this background, in this article, four immune markers were analyzed: il-1β, tnf-α, il-12, and tgf-β. The cytokine Il-1β is a mediator of inflammation and a chemoattractant for other cells of the immune response [23]. tnf-α is a cytokine expressed in the early stages of infection, increases the phagocytic activity of macrophages, and stimulates the survival of infected macrophages, decreasing bacterial growth [23]. Il-12 is a cytokine that promotes polarization toward a Th1-type response and increases the expression of ifn-γ and tnf-α [24]. Finally, tgf-β was analyzed as an anti-inflammatory marker [23]. Different studies have shown that infection with P. salmonis is able to modulate the expression of these genes, both in cells and in salmonids [9,25,26].

Consistent with these previous reports, the results obtained in this study showed differences in the modulation of these four immune response markers between cells infected with live bacteria (PsV) and those infected with temperature-killed bacteria (PsM). For pro-inflammatory markers and those related to a Th1-type response, such as il-1β, tnf-α, and il-12, infection with live bacteria (PsV) induced a response in the cells of lower magnitude and of shorter duration than the response elicited by temperature-killed bacteria (PsM). Interestingly, the response elicited by PsV was only observed during the first 24 h post-infection; however, when extracellular bacteria were eliminated and only the effect generated by the intracellular bacteria was evaluated (48 h), no modulation of immune response markers was observed. In contrast, for the anti-inflammatory marker tgf-β, an increase in expression was only observed in cells infected with PsM. Taken together, these findings suggest that once P. salmonis enters cells, it can modulate the cellular immune response, allowing it to persist and replicate within them.

Importantly, similar patterns have been described in other experimental models. A similar difference was reported in RTS cells, where cells infected with live P. salmonis increased levels of transcript of il-10 compared to cells infected with temperature-killed bacteria. In contrast, the inactivated bacteria increased the expression of other molecules such as hepcidin, demonstrating that P. salmonis modulates host gene expression to persist inside the cell [27]. Likewise, in salmonids, differences in the expression of immune genes such as tnf-α, ifn-γ, and tgf-β have also been reported between fish injected with live P. salmonis and those injected with temperature-inactivated P. salmonis [28].

In addition to evaluating the effect of bacterial viability, we analyzed the impact of silybin during infection. Co-incubation with silybin showed a considerable increase in il-1β and tnf-α transcript levels 24 h post-infection. However, when the effect of silybin alone was analyzed in the cells, only a slight increase in tnf-α was observed, but not in il-1β. These results suggest a possible synergistic effect between silybin and P. salmonis in the induction of gene expression, particularly for il-1β and tnf-α genes. Nevertheless, despite this early increase, the modulation profile of the immune response markers analyzed during the first 48 h of infection in cells treated with silybin (PsV + S) was similar to that observed in cells only infected with P. salmonis and different from that observed in cells infected with temperature-killed bacteria.

Beyond immune gene expression, this study also evaluated the effect of silybin on oxidative stress markers. It has been observed that cells respond to bacterial infections by generating an intracellular environment toxic to bacteria through the production of reactive oxygen species (ROS), while simultaneously protecting themselves from oxidative damage via the action of various antioxidant enzymes [29,30]. In this sense, superoxide dismutase (sod), catalase (cat), and glutathione peroxidase (gsh-px) are important enzymes for the conversion of radicals into non-reactive molecules [31]. This cellular defense mechanism has also been reported to be modulated by P. salmonis during both in vitro cell culture infections and in vivo salmonid infections [32,33,34].

In agreement with this, a difference in the expression of the gsh-px and cat genes was observed when cells were infected with live bacteria PsV and those infected with temperature-killed bacteria PsM, suggesting that P. salmonis modulates the expression of these antioxidant-related genes. On the other hand, cells co-incubated with silybin (PsV + S) showed a gsh-px expression profile similar to that of cells infected with PsM. The gsh-px encodes glutathione peroxidase, the enzyme responsible for the oxidation of reduced glutathione (gsh) to oxidized glutathione (gssg) while reducing peroxides. The increase in the levels of this gene may be related to its antioxidant capacity or also due to the increase in glutathione in the cells. Glutathione has been described as a carrier of nitric oxide (NO) in macrophage-like cells, fulfilling an antibacterial role against mycobacterium [30]; furthermore, the antibacterial activity of glutathione has been demonstrated against bacteria such Acinetobacter baumannii [35,36].

Supporting these transcriptional findings, measurements of intracellular ROS levels also showed differences between cells infected with PsV and PsM, demonstrating the ability of P. salmonis to modulate host cellular processes. The consistent increase in intracellular ROS observed in PsV-infected cells could be a mechanism of cell damage and death induced by P. salmonis [29]. Conversely, the decrease in intracellular ROS induced by silybin treatment can be attributed to its antioxidant capacity to scavenge hydroxyl radicals [37,38].

Considering the experimental design, silybin incubation was performed in conjunction with infection by P. salmonis for 24 h. Subsequently, both the bacteria and the compound were eliminated from the cells; therefore, P. salmonis replication corresponded only to the bacteria that were able to enter the cell for a maximum of 24 h [22]. Based on these observations, the potential mechanism of action of silybin could involve interference with bacterial adhesion, internalization, or replication of the bacteria within the cell. However, the bacterial load results showed that differences in the number of bacteria detected within cells infected with PsV and PsV + S are only noticeable at 72 h post-infection. This suggests that the effect of silybin does not occur in bacterial adhesion (2 h) or internalization (2 to 24 h), but rather in bacterial replication within the cell, since differences are observed 48 h after the compound is removed from the cells (72 h post-infection).

Taken together, the ability of silybin to reduce P. salmonis replication within cells could be due to its effect on modulating the cellular oxidative response to infection, rather than a direct effect on the immune response. However, further research on the effect of silybin on these oxidative stress markers during longer incubation periods is needed to confirm this hypothesis. Additionally, silybin could also modulate the capacity of P. salmonis to evade lysosomal activity in cells [39], as has been described for other natural extracts [20]; however, further experiments are required to verify this hypothesis.

Finally, the in vivo results provide an additional layer of interpretation. Silybin administration in S. salar did not modulate the immune response. This result differs from that reported in other species, such as Cyprinus carpio [40] or (Epinephelus fuscoguttatus ♀ × Epinephelus lanceolatus ♂) [41], where silybin administration modulated various antioxidant and immunostimulatory parameters. This difference may be due to several factors, such as the administration time, as well as the differences between the species to which silybin was administered. This in vivo result is similar to that observed in vitro experiments, where no considerable increase in gene transcript levels was observed in cells incubated with silybin alone, demonstrating that silybin, at the concentration used, probably does not generate an immunostimulatory effect per se on salmonid cells.

Nevertheless, although no increase in the levels of immunological gene transcripts was observed, a protective effect of silybin administration was observed in S. salar against a challenge with P. salmonis. These results also differ from those observed for other treatments with natural products in S. salar against P. salmonis, such as the administration of purified extracts of Quillaja saponina [21], and a phytogenic food additive composed mainly of diterpenes [18], where a relationship between the modulation of the immune response and the protective effect was observed. The differences observed between silybin and other treatments based on natural compounds demonstrate that not all compounds exert the same effects in the same or different species, highlighting the need to individually evaluate their effects and mechanisms of action. Furthermore, this potential difference in the mechanism of action of silybin opens the possibility of exploring its combination with immunostimulatory treatments to enhance the protective effect against P. salmonis. Finally, further study is needed to determine other possible mechanisms of action by which silybin generates protection, considering a potential effect independent of immune stimulation.

4. Materials and Methods

4.1. Compounds, Bacterial Growth, and Cell Culture

Silybin (98% purity) was commercially obtained (Sigma-Aldrich, St Louis, MO, USA). P. salmonis isolate-like LF89 was used for in vitro and in vivo experiments and was obtained from the Bioinformatics and Gene Expression Laboratory, Instituto de Nutrición y Tecnologia de los Alimentos (INTA), University of Chile. Bacterial isolation was confirmed by means of PCR using two primers specific for P. salmonis: the 16S rRNA gene [42] and the glyA gene [43]. The characterization of the P. salmonis genogroup was performed using specific primers for the identification of P. salmonis LF89 [44]. The bacteria were grown in Austral-SRS cell-free medium as previously reported [45]. The SHK-1 cell line used in this study was obtained from the Immunology Laboratory, Aquaculture Biotechnology Center, University of Santiago de Chile.

4.2. Infection Assay in the SHK-1 Cell Line

SHK-1 cells were cultured in T175 cell culture bottles (SPL) using L-15 medium (Cytiva, Hyclone, South Logan, UT, USA) supplemented with 10% fetal bovine serum (Cytiva, Hyclone, South Logan, UT, USA), 4 mM L-glutamine (Mediatech, Corning, Manassas, VA, USA), and 40 μM β-mercaptoethanol (Life Technologies, Gibco, New York, NY, USA). After routine maintenance, 6 × 10^5^ cells were seeded into 6-well plates (SPL) and incubated at 17 °C for 24 h prior to experimental treatments. Subsequently, the cells were incubated with the following treatments: PsV, cells infected with live P. salmonis at an MOI (multiplicity of infection) of 50; PsM, cells infected with temperature-inactivated P. salmonis (50 °C for 30 min); and PsV + S, which corresponds to cells infected with live P. salmonis and incubated with silybin at a concentration of 68 µg/mL, as previously reported [22]. The cells were incubated with the treatment for 24 h and then all cells, including the control group (Ctrl), were washed with PBS 1X and incubated for 1 h with 50 µg/mL of gentamicin, then washed with PBS 1X and incubated for another 24 h with fresh L-15 medium. Samples were collected at 2, 24, and 48 h (24 h post-treatment with gentamicin) of experimentation for further analysis of the transcription levels of immune genes and oxidative stress. To confirm the elimination of extracellular bacteria after gentamicin treatment, the last wash was collected and used for bacterial load detection. For intracellular ROS measurements, the experiment was performed for 2 and 24 h. For bacterial load analysis, the same experiment mentioned above was performed, but in 12-well plates, inoculating 2 × 10^5^ cells. Samples were collected at 2, 24, 48, and 72 h of experimentation. All experiments were performed in quintuplicate.

4.3. Silybin Incubation in the SHK-1 Cell Line

SHK-1 cells were cultured following the aforementioned protocol. Then, 6 × 10^5^ cells were seeded in 6-well plates (SPL) and incubated for 24 h at 17 °C. The cells were then incubated with 68 µg/mL of silybin (S) and maintained for 24 h at 17 °C. Subsequently, samples were collected for further analysis of immune response markers and oxidative stress markers. All experiments were performed in quintuplicate.

4.4. RNA Extraction and cDNA Synthesis

RNA was extracted from the samples using 1 mL of PrimeZOL™ Reagent (Canvax Reagents SL, Valladolid, Spain), following the manufacturer’s instructions. Complementary DNA synthesis was then carried out using the All-In-One 5X RT MasterMix kit (ABM, Richmond, BC, Canada), with 2 µg of RNA, 4 µL of the master mix, and nuclease-free H_2_O added to reach a final reaction volume of 20 µL. Reverse transcription was performed using the following thermal profile: 30 min at 37 °C, 10 min at 60 °C, and 3 min at 95 °C [46].

4.5. DNA Extraction from Cell Culture

Genomic DNA was extracted following a previously published protocol, using 300 µL of sample combined with 70 µL of 5X A solution (TRIS 279 mM, EDTA 101 mM, SDS 45 mM, β-mercaptoethanol 1.3% v/v, and NaCl 684 mM) and 4 µL of proteinase K (20 mg/mL; US Biological, Salem, MA, USA) [22].

4.6. Quantification via qPCR

The quantification of the levels of transcripts of immune markers, il-1β, il-12, tnf-α, and tgf-β, and oxidative stress, mnsod, cusod, cat, and gsh-px (Table S1), and the bacterial load of P. salmonis present in the cells, through the detection of the glyA gene (Table S1), was evaluated by means of real-time PCR. The reaction was prepared with 5 µL of SsoAdvanced™ Universal SYBR^®^ Green Supermix (Bio-Rad, Hercules, CA, USA), 0.5 µL of each primer (10 µM) [42,43,44,47,48,49,50] and 3 µL of nuclease-free water, making a total volume of 9 µL. To this mixture, 1 µL of cDNA was added, reaching a final volume of 10 µL per reaction. The amplifications were performed on the MIC qPCR Cycler (Bio Molecular Systems, Gold Coast, QLD, Australia, using the following thermal profile for the quantification of transcript levels: an initial stage at 95 °C for 2 min, followed by 40 amplification cycles consisting of 95 °C for 5 s, 60 °C for 15 s, and 72 °C for 15 s. Meanwhile, the following thermal profile was used for the detection of the glyA gene: one cycle at 95 °C for 2 min, followed by 35 cycles of 95 °C for 5 s, 61 °C for 15 s, and 72 °C for 15 s. Gene expression data were normalized using elongation factor 1α (ef1a) as the reference gene, and relative transcript levels were calculated using the 2^−ΔΔCT^ method [51]. To facilitate visualization of increases and decreases in gene expression, the results were plotted as (2^−ΔΔCT^).

4.7. Determination of Intracellular ROS

Cells co-incubated with P. salmonis and silybin for 2 and 24 h were washed three times with PBS 1X and incubated with 10 µL of 2′,7′-Dichlorofluorescein diacetate probe (1 mM) (Cayman chemical, Ann Arbor, MI, USA) in 990 µL of IF buffer (1X PBS, 2% fetal bovine serum) for 30 min at 140 rpm in the dark. Subsequently, the cells were washed with PBS 1X and collected with 50 µL of TripLE express (Life technologies, Gibco, New York, NY, USA). Then, cells were centrifuged for 10 min at 1000× g and washed twice with PBS 1X. The cell pellet was resuspended in 400 µL of PBS 1X and sonicated for 5 cycles of 20 s with an amplitude of 80%. The cells were then centrifuged at 16,000× g for 30 min at 4 °C. The resulting supernatant was used to measure fluorescence (excitation 497 nm; emission 522 nm) and protein concentration at 595 nm using Synergy HT (BioTek, Winooski, VT, USA). A Bradford standard curve was used to determine protein concentration [52].

4.8. Fish and Maintenance

Pre-smolt Atlantic salmon (Salmo salar) supplied by Blumar (Talcahuano, Chile) was used in this study. Prior to experimentation, the fish were acclimated for one week in ponds maintained at 12 °C with a stocking density of 14 g/L. During this period, the fish were fed daily at 1% of their body weight using the commercial diet, ORBIT Intro (Composition: Protein 45–49%, Fat: 22–26%, Moisture: 10%, Ash: 12%, Biomar, Puerto Montt, Chile). The same feed was used during the experiments. Approximately 80% of the pond water volume was replaced daily, and water quality parameters, including pH, temperature, and salinity, were routinely monitored. All procedures were conducted in accordance with the ethical standards of the Institutional Ethics Committee of Universidad de Santiago de Chile and current applicable legislation.

4.9. Evaluation of Silybin in the Immune Response of S. salar

The immunostimulatory effect of silybin was evaluated using a total of 48 S. salar (pre-smolt) weighing approximately 30 g each. The fish were divided into two groups (group A and group B) with 24 fish each; in turn, each group was divided into two tanks with 12 fish each. Group A consisted of the control fish (Ctrl), which were fed commercial pellets mechanically mixed with commercial oil. Group B consisted of fish fed commercial pellets mixed with 68 μg of silybin per gram of fish via mechanical oiling (S). The fish were fed their respective treatments for 30 days. Every 10 days, and 24 h post-feeding, 4 fish per tank (8 per group) were weighed, and head kidney and midgut samples were extracted and stored in RNAlater (Invitrogen, Carlsbad, CA, USA) for subsequent RNA extraction and synthesis of cDNA, following the protocol mentioned above in Section 4.4. The quantification of the levels of immune marker transcripts il-1β, il-12, tnf-α, ifn-γ, and tgf-β (Table S1) was performed by means of real-time PCR, according to the protocol mentioned above in Section 4.6.

4.10. Evaluation of the Protective Effect of Silybin in S. salar

The evaluation was carried out with 54 S. salar (pre-smolt) weighing approximately 50 g each. The fish were divided into three groups (group A, B, and C) with 18 fish each; in turn, each group was divided into two ponds with 9 fish each. Group A was used as a control, while group B was used as a control challenge; both groups were fed with commercial pellets mechanically mixed with commercial oil. Group C consisted of fish fed with commercial pellets mixed with 68 μg of silybin per gram of fish (silybin), using mechanical oiling. The fish were fed their respective diets for five days. On day six, the fish were challenged by intraperitoneal injection. Group A, used as the control, received 100 μL of physiological saline solution. Group B (control challenge) and group C (silybin) received 100 μL of physiological saline solution containing 3 × 10^5^ P. salmonis bacteria per gram of fish, which was previously grown for 3 days in the culture medium mentioned in Section 4.1 and quantified using the LIVE/DEAD BacLight bacterial viability and counting kit (Life technologies, Carlsbad, CA, USA) according to the manufacturer’s instructions [46]. Subsequently, the fish were fed their respective treatments for 30 days, and daily mortality was recorded. The anterior kidney and intestine were collected from both dead and surviving fish and stored at −20 °C for later DNA extraction.

4.11. DNA Extraction from Tissue and Quantification of the Bacterial Load of P. salmonis

DNA extraction was performed using the Wizard genomic DNA purification kit (Promega, Madison, WI, USA) following the manufacturer’s recommendations, using approximately 30 mg of each tissue (kidney and intestine), which were homogenized using the tissue disruptor Tissue Master 125 (OMNI International, GA, USA). DNA was quantified using a Nanoquant Infinite M200 Pro (Tecan, Zurich, Switzerland), and the DNA concentration was adjusted to 50 ng/μL. The bacterial load present in the kidney and intestine was quantified by detecting the 16S rRNA gene of P. salmonis (Table S1). The amplifications were performed in the MIC qPCR thermocycler (Bio Molecular Systems, Gold Coast, QLD, Australia), according to the protocol mentioned above in Section 4.6, adding 50 ng of DNA. The thermal profile used was 1 cycle at 95 °C for 2 min, followed by 35 cycles at 95 °C for 5 s, 60 °C for 15 s, and 72 °C for 15 s. A previously prepared calibration curve was used to calculate the number of gene copies.

4.12. Statistical Analysis

All statistical analyses were conducted using GraphPad Prism software (version 8.0). Data were first evaluated for normality and homogeneity of variance. For comparisons between two groups (treatment versus control), Welch’s t-test was applied when parametric assumptions were met, whereas the non-parametric Mann–Whitney U test was used when these assumptions were not satisfied. Differences considered statistically significant are indicated by asterisks, while non-significant differences are denoted as ns. For analyses involving multiple comparisons, ANOVA tests were performed, and statistically significant differences were indicated by different letters. Survival curves were analyzed using the log-rank (Mantel–Cox) test.

5. Conclusions

The results obtained in this study show that the administration of silybin at sub-IC50 concentrations (68 µg/mL) during P. salmonis infection in SHK-1 cells interferes with the intracellular replication process of the bacteria. Although silybin administration in infected cells modulates the transcript levels of pro-inflammatory genes such as il-1β and tnf-α, the overall gene expression profile observed in PsV + S was similar to that of PsV and different from PsM.

Additionally, silybin treatment during infection modulates oxidative stress markers, particularly gsh-px transcript levels and intracellular ROS, generating a profile similar to that observed in PsM and different from PsV. Taken together, these findings suggest that the decrease in intracellular replication of P. salmonis could be related to modulation of the oxidative stress response rather than a direct immunostimulatory effect; however, further experiments are needed to determine whether the effect on oxidative stress is related to the observed increase in pro-inflammatory markers.

Finally, the administration of 68 µg/g of silybin to S. salar generates protection against P. salmonis independently of modulation of the immune response. Further experiments are required to clarify the mechanism of action by which this compound generates protection in S. salar.

6. Patents

The research presented in this study is part of a patent application entitled “Food additive to combat infectious diseases caused by marine bacterial pathogens” (Application number 202402559).