Sil: A Streptococcus iniae Bacteriocin with Dual Role as an Antimicrobial and an Immunomodulator That Inhibits Innate Immune Response and Promotes S. iniae Infection

Streptococcus iniae is a Gram-positive bacterium and a severe pathogen to a wide range of economically important fish species. In addition, S. iniae is also a zoonotic pathogen and can cause serious infections in humans. In this study, we identified from a pathogenic S. iniae strain a putative bacteriocin, Sil, and examined its biological activity. Sil is composed of 101 amino acid residues and shares 35.6% overall sequence identity with the lactococcin 972 of Lactococcus lactis. Immunoblot analysis showed that Sil was secreted by S. iniae into the extracellular milieu. Purified recombinant Sil (rSil) exhibited a dose-dependent inhibitory effect on the growth of Bacillus subtilis but had no impact on the growths of other 16 Gram-positive bacteria and 10 Gram-negative bacteria representing 23 different bacterial species. Treatment of rSil by heating at 50°C abolished the activity of rSil. rSil bound to the surface of B. subtilis but induced no killing of the target cells. Cellular study revealed that rSil interacted with turbot (Scophthalmus maximus) head kidney monocytes and inhibited the innate immune response of the cells, which led to enhanced cellular infection of S. iniae. Antibody blocking of the extracellular Sil produced by S. iniae significantly attenuated the infectivity of S. iniae. Consistent with these in vitro observations, in vivo study showed that administration of turbot with rSil prior to S. iniae infection significantly increased bacterial dissemination and colonization in fish tissues. Taken together, these results indicate that Sil is a novel virulence-associated bacteriostatic and an immunoregulator that promotes S. iniae infection by impairing the immune defense of host fish.

Bacteriocins are ribosomally synthesized peptides or proteins with antibacterial activity. They are produced by a wide range of bacteria against bacteria of the same or closely related species [16]. Bacteriocins are generally grouped into two classes [17]. Class I bacteriocins (also called lantibiotics) are a class of small peptides (19-38 amino acids) characterized by the presence of unusual amino acids such as lanthionine, methyllanthionine, dehydroalanine, and 2-aminoisobutyric acid [18,19]. Class II bacteriocins are non-lantibiotic peptides that are further subdivided into several subclasses [20].
Bacteriocins employ a variety of action mechanisms to kill their target cells, notably blocking cell wall synthesis by binding to cell wall precursors and disrupting membrane potential by producing pores in the cell membrane [21,22]. Nisin, a lantibiotic bacteriocin, exhibits dual killing mechanisms. Through interaction with lipid II, an essential precursor for peptidoglycan biosynthesis, nisin can induce both pore formation and inhibition of cell wall biosynthesis [23]. In addition to targeting at cell membrane and cell wall, bacteriocins can also cause bacterial lysis by interfering with DNA, RNA, and protein metabolism [24].
Although bacteriocins were initially identified in Escherichia coli, most of the currently known bacteriocins are products of Grampositive bacteria, in particular lactic acid bacteria (LAB) [25]. LAB-derived bacteriocins, e.g. nisin, have been used widely as preservatives in food industries. Other Gram-positive bacteria, such as those belonging to the Bacillus and Staphylococcus genera, are also producers of bacteriocins [18,26]. In Streptococcus, bacteriocins have been discovered from S. mutans, S. pyogenes, S. uberis, and S. salivarius [27,28]. However, to our knowledge, no bacteriocins from S. iniae have been reported.
In a recent study, we sequenced the complete genome of a pathogenic S. iniae isolate, SF1 [29]. We found that the SF1 genome contains a gene (named sil) that encodes a homologue of lactococcin 972, a nonlan bacteriocin produced by Lactococcus lactis IPLA 972 [30]. In the present study, we examined the biological activity of Sil. Our results indicate that Sil is a novel bacteriocin that possesses not only antimicrobial activity but also immunoregulatory property and promotes S. iniae infection through suppressing the immune response of host fish.

Ethics statement
Experiments involving live animals were conducted in accordance with the "Regulations for the Administration of Affairs Concerning Experimental Animals" promulgated by the State Science and Technology Commission of Shandong Province. The study was approved by the ethics committee of Institute of Oceanology, Chinese Academy of Sciences.

Fish
Clinically healthy turbot (Scophthalmus maximus) (average 9.7 g) were purchased from a local fish farm and acclimatized in the laboratory for two weeks before experimental manipulation. Fish were fed daily with commercial dry pellets and maintained at 20uC in aerated seawater that was changed once daily. For tissue collection, fish were euthanized with tricaine methanesulfonate (Sigma, St. Louis, MO, USA) as described previously [31].

Bacterial strains and culture conditions
The bacterial strains used in this study are listed in Table 1

Sequence analysis
The sequence of Sil was analyzed using the BLAST program at the National Center for Biotechnology Information (NCBI) and the Expert Protein Analysis System. The theoretical molecular mass and isoelectric point were predicted using the DNAStar software package (Madison, WI). Domain search was performed with the conserved domain search program of NCBI. Signal peptide search was performed with the SignalP (v3.0) program. Subcellular localization prediction was performed with the PSORTb v.3.0 server.

Plasmid construction and purification of recombinant proteins
To construct pEtSil, which expresses C-terminally His-tagged Sil, the coding sequence of Sil was amplified by PCR with primers F1 (5'-GATATCATGGAATCAATTAGTGTAGCCG-3', un-derlined sequence, EcoRV site) and R1 (5'-GATATCGTTA-GAATAAAATCCAGTTGG-3', underlined sequence, EcoRV site); the PCR products were ligated with the T-A cloning vector T-Simple (TransGen Biotech, Beijing, China), and the recombinant plasmid was digested with EcoRV to retrieve the silcontaining fragment, which was inserted into pET259 [37] at the EcoRV site.

Purification of recombinant proteins and preparation of antisera
For purification of recombinant Sil (rSil), E. coli BL21(DE3) was transformed with pEtSil. For purification of recombinant Trx (rTrx), E. coli BL21(DE3) was transformed with pET32a (Novagen, San Diego, CA, USA). rTrx was used as a control protein in this study, because (i) it has been used widely in many studies as a protein tag and in general has no effect on the biological operations of eukaryotic and prokaryotic systems; (ii) it can be purified from E. coli under exactly the same conditions as those used for the purification of rSil, and therefore is a good control for rSil as far as the purification procedure is concerned. In addition to rSil and rTrx, the recombinant E. tarda proteins rEta2 [38] and rEt18 [39], which were used in protein-cell binding assay (see below sections entitled ''Interaction of rSil with bacterial cells'' and ''Interaction of rSil with turbot HKM''), were also purified using the E. coli transformants BL21(DE3)/pETEta2 and BL21(DE3)/ pEt18 respectively, that express these two proteins [38,39]. For all purifications, the transformants were cultured in LB medium at 37uC to an OD 600 of 0.5, and isopropyl-b-D-thiogalactopyranoside was added to the culture to a final concentration of 1 mM. The growth was continued at 18uC for 12 h, and recombinant proteins were purified using nickel-nitrilotriacetic acid columns (GE Healthcare, USA) as recommended by the manufacturer. The purified proteins were treated with Triton X-114 to remove endotoxin as reported previously [40]. After treatment, the proteins were dialyzed for 24 h against phosphate-buffered saline (PBS) and concentrated with PEG20000 (Solarbio, Beijing, China). The proteins were analyzed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) and visualized after staining with Coomassie brilliant blue R-250. The concentrations of the proteins were determined using Protein Assay Kit (Sangon, Shanghai, China). Rat antisera against rSil and rTrx were prepared as reported previously [41].
Immunoblot to determine the subcellular localization of Sil S. iniae SF1 was cultured in LB medium to an OD 600 of 0.5. The supernatant was collected by centrifugation and concentrated 20 times with PEG20000 (Solarbio, Beijing, China). The proteins were resolved by SDS-PAGE and transferred to a nitrocellulose membrane (Amersham, Cambridge, UK). Western blot was performed as described previously [42] with rSil antiserum (1/ 1000 dilution) or rTrx antiserum (1/1000 dilution).

Antibacterial effect of rSil
The bacterial stains listed in Table 1 were cultured as described above to an OD 600 of 0.8. The cells were washed with PBS and resuspended in fresh medium to 10 5 CFU/ml. rSil was added to the cell culture to a final concentration of 20 mM, 40 mM, or 80 mM. PBS was added to the control cells. The cells were then cultured at 28uC or 37uC (as described above for different bacteria) for 0 h to 10 h, and cell density was measured using a spectrophotometer (Amersham, Cambridge, UK). To examine the bactericidal potential of rSil, B. subtilis 1.460 was suspended in PBS to 10 5 CFU/ml, and rSil was added to the bacterial suspension to a final concentration of 80 mM. The cells were incubated at 28uC for 4 h or 8 h. After incubation, the cells were diluted in PBS and plated on LB agar plates. The plates were incubated at 28uC for 24 h, and the colonies emerged on the plates were counted. To examine heat sensitivity of rSil, the protein was treated at 50uC, 60uC, or 100uC for 15 min, and the antibacterial effect of the treated rSil was determined as above.
Antibacterial effect of the culture supernatant of S. iniae SF1 S. iniae SF1 was cultured in LB medium to OD 600 of 0.5, and the supernatant was concentrated 20 times as described above. The concentrated supernatant was passed through 0.22 mM microfilters to remove any residual bacterial cells. B. subtilis 1.460 was cultured as above to OD 600 of 0.8 and collected by centrifugation. The cells were resuspended in 1 ml LB medium supplemented with or without 50 ml of (i) SF1 supernatant, (ii) SF1 supernatant plus rSil antiserum (1/200 dilution), or (iii) SF1 supernatant plus preimmune serum (1/200 dilution). The cells were incubated at 28uC for various hours and plated on LB agar plates. After incubation at 28uC for 24 h, the number of colonies was counted.

Interaction of rSil with bacterial cells
B. subtilis 1.460 cultured as above was resuspended in PBS to 10 8 CFU/ml. rSil, rTrx, rEta2, or rEt18 was added to the suspension to a final concentration of 80 mM. The cells were incubated at 28uC for 2 h and then centrifuged at 800 g for 5 min. The cells were collected, washed three times with PBS, and resuspended in PBS. Rat antiserum against rSil or rTrx was added to the cells (1/1000 dilution), and the cells were incubated at 30uC for 1 h. The cells were centrifuged, washed, and resuspended in PBS as above. Fluorescein isothiocyanate (FITC)-labeled goat anti-rat IgG (Bioss, Beijing, China) was added to the cells (1/1000 dilution), and the cells were incubated at 37uC for 1 h in the dark. The cells were centrifuged, washed with PBS, and resuspended in PBS. The cells were observed with a fluorescence microscope (Nikon E800, Japan). Effect of rSil1 on bacterial infection of turbot head kidney monocytes (HKM) To prepare HKM, head kidney was taken from turbot (,750 g) under aseptic conditions. The tissues were ground and passed through a sterile 75 mm metal mesh with PBS containing 10% fetal bovine serum (FBS). The cell suspension was centrifuged at 300 g for 5 min, and the pelleted cells were washed and resuspended in PBS containing 10% FBS. HKM were extracted from the cell suspension with Fish Monocyte Separation Kit (Hao Yang Biological Manufacture Co., Tianjin, China) as instructed by the manufacturer. The cells were distributed into 96-well tissue culture plates (,10 5 cells/well) containing L-15 medium (Thermo Scientific HyClone, Beijing, China). To examine the effect of rSil1 on bacterial infection, rSil or rTrx was added to HKM in the culture plates to a final concentration of 30 mM. For the control cells, the same volume of PBS was added. The plates were incubated at 22uC for 1 h. S iniae SF1 was cultured in LB medium to an OD 600 of 0.8 and resuspended in L-15 to 10 7 CFU/ml. One hundred microliters of S. iniae were added to each well of HKM. The plates were incubated at 22uC for 2 h or 4 h. After incubation, the cells were washed three times with PBS to remove uninfected bacteria. Penicillin and streptomycin (Sangon, Shanghai, China) (100 U) were added to the cells, and the cells were incubated at 28uC for 1 h to kill extracellular bacteria. The cells were then lysed by adding 100 ml of 1% Triton X-100 to each well; the lysate was diluted in LB medium and plated in triplicate on LB agar plates. The plates were incubated at 28uC for 48 h, and the colonies that emerged on the plates were counted. The genetic identity of the colonies was verified by PCR and sequencing analysis of randomly selected PCR products.

Interaction of rSil with turbot HKM
HKM were resuspended in PBS to 10 7 cells/ml. rSil, rTrx, rEta2, or rEt18 was added to the HKM suspension to a final concentration of 80 mM. The cells were incubated at 22uC for 2 h and then centrifuged at 300 g for 5 min. The cells were collected, washed three times with PBS, and resuspended in PBS. rSil antiserum was added to the cells (1/1000 dilution), and the cells were incubated at 30uC for 1 h. After incubation, the cells were centrifuged, washed, and resuspended in PBS as above. FITClabeled goat anti-Rat IgG (Bioss, Beijing, China) was added to the cells (1/1000 dilution), and the cells were incubated at 37uC for 1 h in the dark. After incubation, the cells were centrifuged and washed with PBS as above. The cells were resuspended in PBS and observed under a fluorescence microscope (Nikon E800, Japan).

Respiratory burst and acid phosphatase assays
rSil was added to HKM in a 96-well tissue culture plate (,10 5 cells/well) to a final concentration of 10 mM, 20 mM, or 30 mM. The same volume of PBS was added to the control cells. The plate was incubated at 22uC for 1 h. S iniae SF1 was cultured as above and resuspended in L-15 to 10 7 CFU/ml. One hundred microliters of S. iniae was added to each well of HKM. The plate was incubated at 22uC for 3 h. After incubation, the cells were determined for respiratory burst as reported previously [43] and for acid phosphatase activity as follows: the cells were lysed by adding 100 ml of 1% Triton X-100 to each well and incubation at 4uC for 20 min. After incubation, acid phosphatase activity was determined using Acid Phosphatase Assay Kit (Beyotime, Beijing, China) according to manufacturer's instruction.
Effect of rSil antibodies on S. iniae infection of turbot HKM S iniae SF1 was cultured as above and resuspended in L-15 to 10 7 CFU/ml. The bacterial suspension was mixed with antiserum against rSil (1/250, 1/500, and 1/1000 dilutions), rTrx (1/250 dilution), or preimmune serum (1/250 dilution). The control bacterial cells were mixed with PBS. After incubation at room temperature for 1 h, 100 ml of each of the mixes were added to HKM in 96-well culture plates. The plates were incubated at 28uC for 2 h. After incubation, uninfected bacteria were removed as above. The cells were lysed, and bacterial recovery was determined as above.

Effect of rSil on S. iniae infection under in vivo conditions
Turbot were randomly divided into three groups (20 fish per group) and injected intraperitoneally with 50 mg rSil1, rTrx, or PBS. At 4 h post-injection, turbot in each group were infected via intraperitoneal injection with 100 ml S. iniae SF1 that had been cultured as above and resuspended in PBS to 1610 7 CFU/ml. At 6 h, 12 h, and 24 h post-infection, kidney and spleen were aseptically taken from the fish (five per time point). The tissues were weighed and homogenized in a glass homogenizer containing PBS. The homogenates were serially diluted and plated in triplicate on LB agar plates. The plates were incubated as above, and bacterial recovery was determined as above.

Statistical analysis
Statistical analyses were performed using analysis of variance (ANOVA) in the SPSS 18.0 package (SPSS Inc., Chicago, IL, USA). All experiments were performed at least three times, and the results are shown as means plus or minus standard errors of the means (SEM). In all cases, significance was defined as P,0.05.

Sequence of Sil
Sil is composed of 101 amino acid residues. It possesses a putative signal peptide (formed by the N-terminal 25 residues) and was predicted to be an extracellular protein. Conserved domain search indicated that Sil is a member of the lactococcin 972 family ( Figure S1A). A search of the S. iniae SF1 genome sequence (GenBank accession no. CP005941) showed that, like lactococcin 972, which is genetically closely linked to putative ABC transporters, the sil gene (named lactococcin 972 in the genome) is followed immediately by the genes encoding a bacteriocinassociated integral membrane protein and an ABC transporter ATP-binding protein ( Figure S1B). Blast analysis identified one close homologue to Sil, i.e., lactococcin 972 of Lactococcus lactis, which shares 35.6% overall sequence identity with Sil ( Figure S2).

Extracellular production of Sil
To facilitate functional and subcellular localization studies, rSil was purified from E. coli as a His-tagged protein ( Figure S3) and used for antibody preparation. To examine whether, as predicted, Sil was localized in the extracellular milieu, S. iniae SF1 was cultured in LB medium to mid-logarithmic phase, and the proteins in the culture supernatant were subjected to Western blot with antibodies against rSil. The results showed that Sil was detected in the supernatant by anti-rSil antibodies but not by antibodies against rTrx, a His-tagged protein purified under the same condition as rSil (Fig. 1). The production level of Sil protein in the culture supernatant of SF1 was estimated by Western blot, which showed that at OD 600 of 0.5, the amount of Sil produced by 2.4610 7 SF1 was approximately 1 mg ( Figure S4).

Antibacterial effect of rSil
To examine whether rSil possessed any antibacterial activity, rSil was incubated with 27 different bacteria (17 Gram-positive and 10 Gram-negative), which represent 23 bacterial species (13 Gram-positive species and 10 Gram-negative species). Subsequent growth analysis showed that rSil inhibited the growth of B. subtilis in a manner that depended on the dose of the protein (Fig. 2). In contrast, rSil had no apparent effect on the growth of all other tested bacteria (data not shown). When rSil was treated at 50uC, 60uC, or 100uC for 15 min, no antibacterial effect on B. subtilis was detected. To examine whether rSil was bactericidal against B. subtilis, the bacterium was incubated with rSil at a high dose (80 mM) for 4 h or 8 h, and the survival rate of the cells was determined by plate count. The results showed that at both examined time points, the bacterial survival rates were ,100%, suggesting that rSil is not bactericidal.

Antibacterial effect of the culture supernatant of S. iniae SF1
Since, as shown above, Sil was secreted into the extracellular milieu, we examined whether the extracellular Sil had any antibacterial activity. For this purpose, the culture supernatant of S. iniae SF1 together with or without rSil antiserum was added to the cell culture of B. subtilis. The cells were then incubated under normal conditions for 4 h and 8 h. Subsequent plate count showed that although the numbers of B. subtilis increased with time in all cases, addition of S. iniae SF1 supernatant drastically decreased the number of B. subtilis, and that this decreasing effect of SF1 supernatant was abolished in the presence of rSil antiserum but not in the presence of preimmune serum (Fig. 3).

Binding of rSil to target bacterial cells
To examine potential interaction between rSil and the target bacterial cells, B. subtilis, which is rSil-susceptible, and Staphylococcus aureus, which is rSil-resistant, were incubated with rSil or the control protein rTrx. The cell-bound protein was detected with FITC-labeled antibodies and observed under a fluorescence microscopy. In addition, to rule out the possibility that binding was due to nonspecific cationic interactions between the protein and the cells, the binding assay was also conducted with two more recombinant proteins, i.e. the E. tarda proteins rEta2 [38] and rEt18 [39] expressed in and purified from E. coli as rSil. Like rSil, which has an isoelectric point (pI) of 8.4, rEta2 and rEt18 also have high pI values (9.5 and 9.6 respectively) and are positively charged at physiological pH. The results showed that fluorescence was detected in B. subtilis treated with rSil but not in B. subtilis treated with rTrx, rEta2, or Et18 ( Fig. 4 and data not shown). In contrast, no binding between rSil and S. aureus was observed ( Figure S5).

Interaction of rSil with turbot head kidney monocytes (HKM) and its effect on cellular immune defense
In an in vitro cellular infection study, we found that when S. iniae was incubated with turbot HKM that had been pre-treated with rSil, the number of bacterial cells succeeded in infection was significantly increased (Fig. 5). In contrast, S. iniae infection of HKM treated with rTrx was comparable to that of the untreated control cells. Since, as observed above, rSil had no direct effect on S. iniae, we wondered whether rSil could affect the activity of   HKM and thereby cause enhanced bacterial infection. To investigate this question, we first examined whether there was any direct physical interaction between rSil and HKM. We found that, as revealed by immunofluorescence microscopy, rSil was indeed able to bind HKM, whereas the Sil-irrelevant proteins rEta2 and rEt18 failed to do so ( Fig. 6 and data not shown). We next examined the effect of rSil-HKM interaction on the immune activity of the cells. For this purpose, HKM were treated with different concentrations of rSil and examined for respiratory burst and acid phosphatase activity. The results showed that compared to untreated cells, cells treated with rSil exhibited significantly reduced respiratory burst and acid phosphatase activity in a manner that was dependent on the dose of rSil (Fig. 7). In contrast, the levels of respiratory burst and phosphatase activity in cells treated with rTrx were comparable to those of untreated control cells (data not shown).

Effect of rSil antibodies on S. iniae infection
With the above results, which showed that Sil was secreted by S. iniae and that rSil inhibited activation of host HKM, we reasoned that antibody blocking of the extracellular Sil produced by S. iniae may have a negative impact on S. iniae infection. To examine whether or not this was true, a cellular infection study was conducted, in which HKM were incubated with S. iniae in the presence or absence of rSil antiserum in different dilutions. The results showed that compared to untreated cells, cells treated with rSil antiserum exhibited significantly decreased bacterial infection, in particular when the antiserum dilutions were small (Fig. 8). In

Effect of rSil on S. iniae dissemination and colonization in fish tissues
So far all the biological effects of rSil were observed under in vitro conditions. To examine the potential in vivo effect of rSil, turbot were administered with rSil or rTrx before being infected with S. iniae. Bacterial loads in kidney and spleen were determined at 6 h, 12 h, and 24 h post-infection. The results showed that compared to untreated fish and fish treated with rTrx, fish treated with rSil exhibited significantly increased numbers of bacterial recoveries from both examined tissues at 6 h and 12 h postinfection (Fig. 9).

Discussion
In this study, we examined the activity and biological effect of Sil, a bacteriocin from the fish pathogen S. iniae, under in vitro and in vivo conditions. Sil is a structural homologue of lactococcin 972 but shares low sequence identity with the latter, which suggests a molecular basis for potential functional difference between these two proteins. Consistent with the presence of a signal peptide and the predicted subcellular localization in the extracellular region, Sil was detected in the culture supernatant of S. iniae SF1, apparently secreted by the cultured cells. These results are in line with the structural classification of Sil being a bacteriocin.
Many bacteriocins exhibit broad-spectrum anti activities. For example, subtilomycin, a lantibiotic from B. subtilis, has a wide target range that includes Gram-positive and Gram-negative pathogens as well as several Candida species [44]. However, most LAB-derived bacteriocins display a target range that is limited to only closely related bacterial strains [44][45][46]. In our study, of the 27 bacteria examined, only B. subtilis was inhibited in growth by rSil, suggesting that rSil as a bacteriocin has a very narrow target specificity. Previous reports showed that bacteriocins with both heat-resistant and heat-sensitive properties exist, the former can maintain activity at extremely high temperatures (over 80uC) [47][48][49][50][51][52]. In the case of rSil, we found that it lost activity after being heated at 50uC and above, suggesting that rSil is a heat-labile protein. In addition to rSil, the culture supernatant of S. iniae SF1 also inhibited the growth of B. subtilis, which inhibition, however, was blocked by antibodies against rSil. These results indicate that the antimicrobial effect observed with rSil is most likely a reflection of the natural biological property of the Sil produced by S. iniae.
It is known that entry of bacteriocins into the target cell requires interaction with specific cell surface molecules, such as mannosephosphotransferase systems and lipid II, which serve as receptors or docking sites for bacteriocins [53]. In our study, we observed specific binding of rSil to B. subtilis, suggesting the existence of receptor-like molecules on B. subtilis. It is likely that the rSil-specific receptor present in B. subtilis is absent in the other examined bacteria that showed resistance to rSil. Unlike most bacteriocins, which are bactericides that kill their target cells by various mechanisms, rSil inhibited the growth of B. subtilis but did not affect the viability of the cells. This property of rSil is apparently different from that of lactococcin 972, which blocks cell division by inhibiting septum formation, resulting in structural changes in the cell that eventually lead to cell death [54][55][56].
To date, numerous studies have been carried out on the antimicrobial activity and action mechanism of bacteriocins against the target microbial cells, however, the potential effect of bacteriocins on the immunity of animal hosts has not been investigated. In our study, we found that rSil interacted not only with target bacterial cells but also with fish HKM, and the latter caused reduction of respiratory burst and acid phosphatase activity in the immune cells. These evidences indicate that Sil is a unique bacteriocin that possesses immunoregulatory property in a negative manner. Consistent with these observations, treatment of HKM with rSil significantly weakened cellular defense against S. iniae infection, which is most likely due to the inability of HKM to elicit a full-scale immune response in the presence of rSil. These results point to a role of Sil as a virulence factor that promotes infection of the bacteriocin-producer strain by repressing host fish immunity. This conclusion was corroborated by the observation that antibody blocking of the extracellular Sil produced by S. iniae significantly impaired S. iniae invasion into HKM.
Although most investigations of bacteriocin activity were performed in vitro, some in vivo studies suggest application potentials  of bacteriocins as antimicrobials in disease control [57]. For example, studies in murine models showed that in vivo administration of bacteriocins provided protection against Salmonella and Listeria monocytogenes infection [58][59][60]. In addition, bacteriocinproducing probiotic Streptococcus salivarius as well as other bacterial species have been found to offer beneficial modulator capabilities within local microbiome through inhibitory activity against potentially deleterious bacteria [61][62][63]. However, investigations on the in vivo effect of bacteriocins on the infection of the bacteriocin-producer strains are scarce. In our study, we found that pre-administration of turbot with rSil before S. iniae infection significantly increased bacterial dissemination into tissues of the infected fish, suggesting that, as observed in the cellular studies with HKM, rSil facilitated S. iniae infection under in vivo conditions. Given its in vitro effect, it is possible that the in vivo effect of rSil is mediated through negative modulation of the fish immune response that is essential to block bacterial invasion.
In conclusion, we demonstrate in this study that Sil is a novel bacteriocin that acts as an antimicrobial and an immunomodulator, and that in the latter capacity Sil represses the innate immunity of host fish in a manner that contributes to the infectivity of S. iniae. As such, Sil participates in the pathogenicity of the producer strain not only through inhibition of the growth of potentially competing bacteria but, more likely, through manipulating the immune defense of the host fish. These observations add new insights to the function and working mechanism of bacteriocins, as well as indicating for the first time that immune evasion is a likely virulence strategy of S. iniae.