Bacteriophage EK99P-1 alleviates enterotoxigenic Escherichia coli K99-induced barrier dysfunction and inflammation

Bacteriophages, simply phages, have long been used as a potential alternative to antibiotics for livestock due to their ability to specifically kill enterotoxigenic Escherichia coli (ETEC), which is a major cause of diarrhea in piglets. However, the control of ETEC infection by phages within intestinal epithelial cells, and their relationship with host immune responses, remain poorly understood. In this study, we evaluated the effect of phage EK99P-1 against ETEC K99-infected porcine intestinal epithelial cell line (IPEC-J2). Phage EK99P-1 prevented ETEC K99-induced barrier disruption by attenuating the increased permeability mediated by the loss of tight junction proteins such as zonula occludens-1 (ZO-1), occludin, and claudin-3. ETEC K99-induced inflammatory responses, such as interleukin (IL)-8 secretion, were decreased by treatment with phage EK99P-1. We used a IPEC-J2/peripheral blood mononuclear cell (PBMC) transwell co-culture system to investigate whether the modulation of barrier disruption and chemokine secretion by phage EK99P-1 in ETEC K99-infected IPEC-J2 would influence immune cells at the site of basolateral. The results showed that phage EK99P-1 reduced the mRNA expression of ETEC K99-induced pro-inflammatory cytokines, IL-1β and IL-8, from PBMC collected on the basolateral side. Together, these results suggest that phage EK99P-1 prevented ETEC K99-induced barrier dysfunction in IPEC-J2 and alleviated inflammation caused by ETEC K99 infection. Reinforcement of the intestinal barrier, such as regulation of permeability and cytokines, by phage EK99P-1 also modulates the immune cell inflammatory response.

www.nature.com/scientificreports/ Bacteriophages simply phages have long been used as an alternative to antibiotics to treat bacterial infection 15 . Phage-in-feed have been shown to reduce the severity of ETEC-induced diarrhea in newborn 16 and post-weaning 17,18 piglets; they have also been shown to alleviate symptoms such as increased rectal temperature and E. coli adhesion concordant with reducing serum IL-8 and TNF-α 12 . However, limited information is available on the effect of phages on IECs in terms of regulating host immune responses upon ETEC infection.
Therefore, in this study, we investigated the beneficial effects of phage EK99P-1 on barrier protection in IECs, and the inflammatory response involving mucosal immune cells, under ETEC K99 infection.

Results
Phage EK99P-1 restrained intestinal barrier disruption. ETEC induces increased intestinal permeability, resulting in reduced barrier protection 8 . To determine whether phage EK99P-1 treatment alleviates intestinal barrier function damage caused by ETEC K99 infection, we treated differentiated IPEC-J2 with either ETEC K99 or phage EK99P-1, or both, on the apical side. Decreased transepithelial electrical resistance (TEER) values due to ETEC K99 infection were restored by phage EK99P-1 treatment in a dose-dependent manner (Fig. 1A). IPEC-J2 cell death did not occur at any concentration of phage EK99P-1 in this experiment (Fig. S1). TEER values of differentiated IPEC-J2 were maintained until 24 h after treatment with both ETEC K99 and phage EK99P-1, whereas treatment with ETEC K99 alone significantly reduced TEER (Fig. 1B). This result suggests that phage EK99P-1 treatment minimizes damage and inhibits the increased permeability induced by ETEC K99.
Soluble toxins, such as LT and ST, are able to permeate the IEC barrier. The molecular weights of LT and ST are ~ 80 and 2 kD, respectively; they may migrate to the lamina propria across the epithelial layer according to Each datum represents a percentage of initial TEER (n = 3). *P <0.05; **P < 0.01; ***P < 0.001. (C,D) Permeability was assessed by measuring 4-or 40-kD fluorescein isothiocyanate (FITC)-dextran transport after 24 h of infection (n = 3). Data are expressed as means ± standard deviation (SD). Means were compared by one-way analysis of variance (ANOVA), followed by the Friedman test and Tukey's multiple comparison test. Different letters in each group indicate significant differences at P < 0.05. www.nature.com/scientificreports/ their paracellular permeability 2 . To test this hypothesis, we used a dextran-FITC trans-epithelial permeability assay to quantify the effect of ETEC K99 and phage EK99P-1 on paracellular permeability: 4-or 40-kD dextran-FITC permeability was maintained in IPEC-J2 treated with ETEC K99 and phage EK99P-1, whereas ETEC K99 alone showed a significantly increased dextran permeability (Fig. 1C,D). These results demonstrate that phage EK99P-1 efficiently prevented the increased intestinal permeability associated with ETEC K99 infection. No significant changes in permeability were observed upon treatment with phage EK99P-1 alone, because no target bacteria of phage EK99P-1 were present (Fig. 1C,D). Taken together, these results show that phage EK99P-1 ameliorates the intestinal barrier disruption and permeability associated with ETEC K99 infection.

Scientific Reports
Phage EK99P-1 inhibited the loss of tight junction integrity in IPEC-J2 infected with ETEC K99. Our previous study showed that differentiated IPEC-J2 form a polarized monolayer with increased TEER values, concordant with increased expression of tight junction proteins such as ZO-1, occludin, and claudin-3 19 . Because the increase in permeability is partly caused by reduced expression of these tight junction proteins, we examined the expression of ZO-1, occludin, and claudin-3 in IPEC-J2 infected with ETEC K99. Consistent with our permeability results (Fig. 1), disrupted tight junction proteins (ZO-1, occludin, and claudin-3) were restored in ETEC K99-infected IPEC-J2 through phage EK99P-1 treatment ( Fig. 2A), as evidenced by the preservation of the outer line of ZO-1 expression on ETEC K99 treated IPEC-J2 by EK99P-1 (Fig. 2B). These results suggest that phage EK99P-1 treatment effectively prevents intestinal barrier disruption through inducing increased tight junction proteins in IPEC-J2 infected with ETEC K99.
Phage EK99P-1 prevented ETEC K99 adherence to IPEC-J2. ETEC adhesion to IECs during infection plays an important role in inducing inflammatory responses and disrupting barrier function at the intestinal lining 8 . Therefore, we hypothesized that the restoration of permeability by phage EK99P-1 in ETEC K99-infected IPEC-J2 could be related to the initial adhesion of ETEC K99 to IPEC-J2. To examine the interference of phage EK99P-1 in the attachment of ETEC K99 to IPEC-J2, we measured the colony-forming units (cfu) of cell-adhesive ETEC K99. We found that phage EK99P-1 treatment reduced ETEC K99 adhesion to IPEC-J2 (Fig. 3). We observed no ETEC K99 translocation from the apical to the basolateral side, with or without phage EK99P-1 (Fig. 2). This result suggests that the intestinal barrier protection conferred by phage EK99P-1 treatment may have begun with reduced ETEC K99 adhesion to IPEC-J2 due to the interference of phage EK99P-1.

Phage EK99P-1 alleviated inflammatory responses of IPEC-J2 infected with ETEC K99. IECs
have a barrier function that involves sensing and responding to microbial stimuli, and also participate in the coordination of inflammatory responses. For example, IECs secrete cytokines and chemokines upon recognition of ETEC adhesion 20 to alert the underlying mucosal immune cells 21 . To investigate whether phage EK99P-1 treatment modulates inflammatory responses following ETEC K99 infection, we examined the release of cytokines and chemokines in IPEC-J2 treated with ETEC K99 and phage EK99P-1. Compared to treatment with ETEC K99 alone, simultaneous treatment of ETEC K99 and phage EK99P-1 significantly reduced the mRNA expression of IL-8 and MCP-1 (Fig. 4A), and significantly decreased IL-8 secretion (Fig. 4B). These results suggest that phage EK99P-1 alleviates inflammatory responses in IPEC-J2 cells infected with ETEC K99.
Phage EK99P-1 reduced inflammatory cytokines in porcine peripheral blood mononuclear cell (pPBMC) co-cultured with ETEC K99-infected IPEC-J2. IECs are thought to participate in the recognition of bacterial components and transduce signals to resident mucosal immune cells 22 . To investigate the effect of phage EK99P-1 treatment on the modulation of underlying mucosal immune responses by IECs, differentiated IPEC-J2 was co-cultured with pPBMCs using a previously established transwell co-culture system 23 . We investigated the production of pro-inflammatory cytokines and chemokines, IL-1β, IL-8, and IFN-γ, in the cocultured pPBMCs. The mRNA expression levels of IL-1β, IL-8, and IFN-γ were significantly increased under ETEC K99 alone compared to the non-treated control ( Fig. 5A-C). In the co-cultured pPBMCs treated with ETEC K99 in the presence of EK99P-1, IL-1β and IL-8 levels were comparably low to control levels ( Fig. 5A,B). However, we observed no differences in the mRNA expression of IFN-γ when cells were infected with ETEC K99 in the presence of EK99P-1 compared to ETEC K99 alone (Fig. 5C). No significant change in the major immune cells composing pPBMCs (CD172a + myeloid cells; CD3 + total T cells; or CD4 + , CD8 + , or CD4 + CD8 + T cells) was observed following ETEC K99 infection and phage EK99P-1 treatment (Fig. 5D). Together, these data suggest that phage EK99P-1 treatment prevents inflammatory responses caused by ETEC K99 infection by reducing inflammatory cytokines in the immune cells composing pPBMCs.

Discussion
In this study, we examined the effects of phage EK99P-1 on barrier protection and cytokine production in porcine IECs infected with ETEC in the presence or absence of immune cells. Upon ETEC infection, barrier disruption in the host is thought to occur due to the action of enterotoxins, such as LT or ST, produced from the ETEC. It has been reported that endogenously produced or exogenously added LT substantially enhances the adherence of ETEC to the IPEC-J2 24 . We postulated that the decrease of enterotoxin production caused by phage EK99P-1 inhibits ETEC K99 adhesion to IPEC-J2. As we expected, the attachment of ETEC K99 to IPEC-J2 was decreased when the phage EK99P-1 was added at 24 h after treatment. Bacterial cell adhesion is likely starting at early time point, which might increase in a time-dependent manner in proportion to the bacteria growth. Thus, in our experimental setting, we selected the time that showed the best effect of phages inhibiting bacterial adherence at 24 h. Since enterotoxin gives rise to ETEC pathogenesis by facilitating the attachment of ETEC to host IECs, major barrier disruption is initiated by ETEC adhesion 7,8 . Therefore, the alleviation of barrier disruption by phage A disrupted barrier favors indiscriminate and easy penetration by pathogens, and impairment of normal physiological function and the local mucosal immune system, often causing an inflammatory reaction 25 . ETEC infection has been reported to induce increased serum IL-8, which was lowered by dietary phage supplementation in post-weaning pigs 12,13 . As previously reported, phage-in-feed reduced the severity of ETEC-induced diarrhea [16][17][18] , and decreased rectal temperature and E. coli adhesion together with a reduction of serum IL-1β, IL-8 and TNF-α 12 in piglets. ST of ETEC is a key factor inducing IL-8 production in IPEC-J2 cells 26 . We speculated that ST reduction of ETEC K99 by phage K99P-1 could inhibit IL-8 induction in IPEC-J2. Indeed, the current study revealed the induction of IL-8 in cells infected with ETEC K99 in both IECs and immune cells, and its reduction by phage EK99P-1 treatment. Neither IL-1β was significantly increased nor TNF-α was detected (data not shown) after ETEC infection at IPEC-J2 in our experimental setting.
Because IECs are in close contact with immune cells in the lamina propria of intestinal tract, we hypothesized that the modulation of chemokine secretion by phage EK99P-1, in the context of IECs infected with ETEC K99, could influence immune cells at the basolateral side. Therefore, we adopted the IPEC-J2/PBMC www.nature.com/scientificreports/ co-culture system 23 and examined the effect of phage EK99P-1 on immune response modulation. Since it is difficult to distinguish the cytokine response between epithelial cells and PBMCs in the co-culture model we have examined mRNA expression by directly sampling PBMCs after the treatment. After co-culture, IL-1β and IL-8 mRNA expression, upregulated by ETEC K99 infection, was diminished by phage EK99P-1 treatment in pPBMCs. One reason for the low levels of inflammatory cytokines in basolateral pPBMCs may be that soluble toxins, such as LT and ST, secreted from ETEC did not permeate the IEC barrier. Our results showed that both 4and 40-kD dextran-FITC permeability was restored to control levels by phage EK99P-1 treatment against ETEC K99 exposure to IPEC-J2. These results suggest that the basolateral translocation of soluble toxins, especially ST, may also be reduced due to phage EK99P-1 treatment. Indeed, previous reports using enterotoxin-deficient ETEC mutants demonstrated that STb specifically regulates immune-related genes, such as IL-17A, IL-1α, and IL-1β 27 . In addition, ST enhances the expression of pro-inflammatory cytokines and chemokines, such as IL-6 and IL-8, in the small intestine 28 . However, further studies are needed to precisely determine the factors involved in inflammation modulation by immune cells.
Our results also showed that mRNA expression of IFN-γ, produced mainly by T cells 29 , was not significantly different in pPBMCs co-cultured with IPEC-J2 infected with ETEC K99, with or without phage EK99P-1. Therefore, it is probable that the control of ETEC K99 by phage EK99P-1 is unrelated to IFN-γ expression by T cells. However, reduced mRNA expression of IL-1β and il8 was observed in the co-cultured system of treated ETEC K99 and phage EK99P-1, compared to ETEC K99 alone. Considering that the main cells producing IL-1β and IL-8 in pPBMCs are myeloid cells 30 , we infer a possible difference in cytokine production capacity, although no proportional change in the population of cells was observed. Our results show that phage EK99P-1 blocks intestinal barrier disruption and inhibits inflammatory responses in IPEC-J2 infected with ETEC K99, by reducing

Materials and methods
The use of porcine blood and experimental protocols of the present study were approved by the Institutional Animal Care and Use Committee of Seoul National University and all methods were performed in accordance with the relevant guidelines and regulations.

Bacterial infection and adhesion assay. Differentiated IPEC-J2 cultured in a transwell were washed
with pre-warmed phosphate-buffered saline (PBS) and then placed in infection media (DMEM-F12 supplemented with 5% heat-inactivated FBS without antibiotics). The cells were infected with ETEC K99 at a density of 1 × 10 7 cfu/mL for 24 h at 37 °C and 5% CO 2 . For the adhesion assay, infected IPEC-J2 were washed with prewarmed PBS three times and then lysed in 0.1% Triton X-100 for 5 min. The lysates were serial diluted and each inoculum, plated onto TSB agar and stored overnight. Viable bacterial cells were quantified as cfu.

Dextran permeability measurement. Differentiated IPEC-J2 was cultured in a transwell and treated
with ETEC K99 and phage EK99P-1 for 24 h, and then washed with pre-warmed PBS three times. Cell culture medium containing 4 or 40 kDa dextran conjugated with fluorescein isothiocyanate (dextran-FITC) was added to the upper compartment of the transwell plate at 2.2 mg/mL. After 1 h of incubation, the fluorescence intensity in the lower compartment of the transwell plate was measured using a fluorescence multiple plate reader (Victor 3; Perkin Elmer, USA). The excitation and emission wavelengths were 490 and 520 nm, respectively. pPBMC isolation. Porcine blood samples were obtained from 4-to 6-month-old Landrace-Yorkshire-Duroc pigs (Hyupsin Food Co., Ltd., Korea). The use of porcine blood was approved by the Institutional Animal Care and Use Committee of Seoul National University (IACUC no. SNU-150327-2). Porcine whole blood was diluted with PBS at a ratio of 1:1, and pPBMCs were isolated by density gradient centrifugation (400×g for 20 min without brake) using Ficoll-Paque Plus (Amersham Bioscience, UK) 32 . pPBMCs were suspended in RPMI 1640 medium supplemented with 10% FBS and 1% penicillin/streptomycin.
Co-culture IPEC-J2/PBMCs. IPEC-J2 was seeded at a density of 1 × 10 5 cells/mL in 500 μL DMEM medium as described above on 1.12-cm 2 polyester membrane inserts (pore size, 0.4); the basolateral side was filled with 1 mL DMEM. During cell growth and differentiation, the medium in both compartments was replaced three times per week for 7-9 days. Then, pPBMCs (2 × 10 6 cells/mL) were seeded in the basolateral compartment of the transwell plate with 1 mL RPMI media, and ETEC K99 and phage EK99P-1 were added to the apical compartment.

Western blot analysis.
To examine the effect of phage EK99P-1 on tight junction proteins, 1 × 10 5 cells were cultured on 12-well plates for 2 days until confluence was reached. Confluent IPEC-J2 was treated with ETEC K99 and/or phage EK99P-1, washed with PBS and then lysed in a RIPA lysis buffer (50 mM Tris-HCl, 150 mM NaCl, 1% NP_40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate containing protease inhibitor), followed by protein quantitation using a Micro BCA kit (Thermo, USA). The same amount of extract was loaded into 12% Tris-glycine polyacrylamide gel and electrophoresed. Then, the proteins were transferred onto a polyvinylidene fluoride microporous membrane for 90 min at 4 °C and blocked with 5% skim milk in TBS-T (1 M Tris-HCl, 5 M NaCl, 10% Tween-20) for 1 h. The blot was incubated with rabbit anti-claudin-3, -occludin, and -ZO-1 or mouse anti-β-actin antibodies (Invitrogen, USA) overnight. Then, the membrane was washed and incubated with goat anti-rabbit or anti-mouse IgG-HRP antibody (Santa Cruz Biotechnology, USA) for 1 h. The target protein was visualized using an enhanced chemiluminescence system (GE Healthcare, USA), followed by analysis using a ChemiDoc XRS system (Bio-Rad, USA).
Proportional changes of immune cells. We washed pPBMCs with PBS containing 1% FBS and per- Apoptosis analysis. Floating cells were collected and the attached cells were trypsinized for 5 min and washed with PBS. Finally, both trypsinized and floating cells were added together and stained with Annexin V-APC and propidium iodide (PI). Marker intensity was examined using flow cytometry (FACSCanto II; BD Biosciences). All flow cytometry data were analyzed using the FlowJo software (Tree Star).

Statistical analyses.
Means ± standard deviation were determined on the basis of at least three different samples. All experiments were performed at least three times. Means were compared between two groups using two-tailed paired Student's t-tests. The groups were compared by one-way analysis of variance (ANOVA), followed by the Friedman test and Tukey's multiple comparison test. All statistical analyses were performed using GraphPad Prism software (v5.01; GraphPad Software, USA). Significance was evaluated at a level of P < 0.05.

Ethics declarations.
The porcine blood samples used in this study were approved by the Institutional Animal Care and Use Committee of Seoul National University (IACUC No. SNU-150327-2).   TGG CAG TTT TCC TGC TTT CT  154  Reverse  CAG TGG GGT CCA CTC TCA AT   IL-1β  Forward GTG CAA ACT CCA GGA CAA AGA CCA  120  Reverse  CAC AAG CTC ATG CAG AAC ACCAC   IFN-γ  Forward TGG ATG TGA TCA AGC AAG AC  120  Reverse  TGG CTT TGC GCT GGA TCT   MCP-1  Forward AAG TGG GCA CAC CCG TTT C  120  Reverse  CGC CAT TAT GCG TGA TTG TT   GAPDH  Forward GTC GGT TGT GGA TCT GAC CT  210  Reverse  AGC TTG ACG AAG TGG TCG  www.nature.com/scientificreports/ Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.