Nuclear factor kappa B plays a pivotal role in polyinosinic-polycytidylic acid-induced expression of human β-defensin 2 in intestinal epithelial cells

Authors


M. Asano, Nihon University School of Dentistry, Department of Pathology, 1-8-13 Kanda Surugadai, Chiyoda-ku, Tokyo, 101-8310, Japan
E-mail: asano-m@dent.nihon-u.ac.jp

Summary

Intestinal epithelial cells (IECs) play an important role in protecting the intestinal surface from invading pathogens by producing effector molecules. IECs are one of the major sources of human beta-defensin 2 (hBD-2), and can produce it in response to a variety of stimuli. Although IECs express Toll-like receptor 3 (TLR-3) and can respond to its ligand, double-stranded RNA (dsRNA), hBD-2 expression in response to dsRNA has not been elucidated. In the present study, using an artificial analogue of dsRNA, polyinosinic-polycytidylic acid (poly I:C), we investigated whether the human IEC line, HT-29, can produce hBD-2 in response to poly I:C. HT-29 cells can express hBD-2 mRNA only when stimulated with poly I:C. The induction of hBD-2 mRNA expression was observed at 3 h after stimulation and peaked at 12 h of post-stimulation. Pre-incubation of the cells with nuclear factor kappa B (NF-κB)-specific inhibitor, l-1–4′-tosylamino-phenylethyl-chloromethyl ketone (TPCK) and isohelenine abolished the expression of hBD-2. Detection of the poly I:C signal by TLR-3 on the surface of HT-29 cells was revealed by pre-incubating the cells with anti-TLR-3 antibody. The 5′-regulatory region of the hBD-2 gene contains two NF-κB binding sites. A luciferase assay revealed the importance of the proximal NF-κB binding site for poly I:C-induced expression of hBD-2. Among NF-κB subunits, p65 and p50 were activated by poly I:C stimulation and accumulated in the nucleus. Activation of the p65 subunit was investigated further by determining its phosphorylation status, which revealed that poly I:C stimulation resulted in prolonged phosphorylation of p65. These results indicate clearly that NF-κB plays an indispensable role in poly I:C induced hBD-2 expression in HT-29 cells.

Introduction

The intestinal mucosa is protected from invading pathogens by innate and adaptive immune responses [1,2]. Intestinal epithelial cells (IECs) play pivotal roles in innate immunity by forming a physical barrier and producing effector molecules, such as chemokines, proinflammatory or immunoregulatory cytokines, and anti-microbial peptides [2].

Defensins are cationic, cystein-rich peptides peptides with molecular masses ranging from 3 to 5 kDa [3–5]. They function as anti-microbial components of the innate immune system. Based on their molecular structures, human defensins have been divided into two major groups, α-defensins and β-defensins. While human α-defensins are produced predominantly in neutrophils and Paneth cells in the small intestine, human β-defensins (hBDs) are produced in epithelial cells [3–5]. The hBDs have been categorized further into several classes. hBD-1 is expressed in the respiratory tract, kidney, urogenital and oral epithelia [6,7], whereas hBD-2 is present in the skin, respiratory and gingival epithelia, the stomach and the small and large intestines [8–14]. Although hBD-1 is expressed constitutively, hBD-2 is produced in response to proinflammatory cytokines, microbial products [15,16] and viral infections.

In the course of viral infections, double-stranded RNA (dsRNA), a by-product of RNA virus infection, accumulates inside cells [17]. In response to dsRNA, IECs can induce the production of the polymeric immunoglobulin receptor, an important effector molecule in adaptive immunity in the intestinal tract [18]. dsRNA is sensed by Toll-like receptor 3 (TLR-3), which culminates in the production of adaptive immune effectors. These results demonstrate the importance of the functional link between innate and adaptive immunity. However, in that study, the induced expression of innate immune effectors, such as hBD-2, was not examined. IECs are a major sources of hBD-2 [19]. Although enhanced production of hBD-2 in response to lipopolysaccharide (LPS) or interleukin (IL)-1α was reported previously [14], the regulation of hBD-2 expression in response to dsRNA in IECs has never been examined. Moreover, in a previous report, we demonstrated that poly I:C, an artificial analogue of dsRNA, induced the expression of the intercellular adhesion molecule-1 in HT-29 cells through TLR-3 [20]. These results prompted us to question whether poly I:C stimulation of IECs could induce the production of the innate immune effector, hBD-2. We report here that the expression of hBD-2 was up-regulated strongly by poly I:C stimulation of HT-29 cells through TLR-3. The specific inhibitor of nuclear factor kappa (NF-κB) prevented the poly I:C-stimulated up-regulation of hBD-2, indicating the indispensable role of NF-κB in this signalling pathway.

Materials and methods

Reagents

Polyinosinic-polycytidylic acid (poly I:C), poly deoxyinosinic-deoxycytidylic acid (poly dI:dC) and l-1-4′-tosylamino-phenylethyl-chloromethyl ketone (TPCK) were purchased from Sigma (St Louis, MO, USA). Isohelenine was purchased from Merck (Darmstadt, Germany). The human TLR-3 antibody was purchased from Imgenex (San Diego, CA, USA). Anti-NF-κB p65 subunit was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-phospho-p65 antibody was purchased from Cell Signaling Technology (Tokyo, Japan). Anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody was purchased from Chemicon (Tokyo, Japan).

Cell culture

HT-29 cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS), 50 µg/ml streptomycin and 50 U/ml penicillin (10% FCS–DMEM).

Poly I:C stimulation and reverse transcriptase–polymerase chain reaction (RT–PCR)

HT-29 cells were detached from the culture dish with 0·05% trypsin-ethylenediamine tetraacetic acid (EDTA) on the day before stimulation and plated at a density of 5 × 105 cells/35-mm dish and cultured for 18 h in a 5% CO2 incubator. The cells were washed twice with 10% FCS–DMEM and stimulated with or without 1–1000 µg/ml of poly I:C or poly dI:dC for the indicated lengths of time. At the end of the culture period, the cells were washed once with phosphate-buffered saline (PBS) and total RNA was extracted using the RNeasy Mini Kit (Qiagen, Tokyo, Japan). One µg of total RNA was subjected to first-strand cDNA synthesis using Superscript III reverse transcriptase (Invitrogen, San Diego, CA, USA). RT–PCR was performed using the primers (5′-hBD-2 and 3′-hBD-2) listed in Table 1. For the antibody inhibition assay, HT-29 cells were pre-incubated with anti-TLR-3 antibody or control antibody for 1 h at 37°C and then stimulated with poly I:C for 3 h. For NF-κB inhibitor experiments, the cells were incubated at 37°C with 25 µM TPCK for 30 min or with 25 µM isohelenin for 2 h. Cells were washed three times with 10% FCS–DMEM and cultured further in the presence or absence of 100 µg/ml of poly I:C for 3 h.

Table 1.  Primers used in this study.
5′-hBD-2 primer5′-CCAGCCATCAGCCATGAGGGT-3′
3′-hBD-2 primer5′-GGAGCCCTTTCTGAATCCGCA-3′
hBD-2 genomic 5′-primer5′-AAGAAAGGCCCTCTCCTGAGT-3′
hBD-2 genomic 3′-primer5′-GAGGAGATACAAGACCCTCAT-3′
QCU5′-primer5′-TAAATGGCGCAAGATAGGAGGTCCCAAGAGCAGGAGG-3′
QCU 3′-primer5′-CCTCCTGCTCTTGGGACCTCCTATCTTGCGCCATTTA-3′
QCL 5′-primer5′-GGAGGAAGGGATTTTCGAGTCCAGATTTGCATAAGAT-3′
QCL 3′-primer5′-ATCTTATGCAAATCTGGACTCGAAAATCCCTTCCTCC-3′

Small interfering RNA (siRNA) transfection

The siRNA against TLR-3 and control siRNA were purchased from Invitrogen. The cells were plated on the day before transfection at the density of 5 × 105/35-mm dish. The cells were washed three times with OPTI-MEM (Invitrogen) and then transfected by the lipofectamine RNAiMAX transfection method (Invitrogen), according to the manufacturer's instructions. Briefly, 100 pmol of RNAi duplex was diluted in 250 µl of OPTI-MEM and 6 µl of lipofectamine RNAiMAX was diluted in 250 µl of OPTI-MEM in each Eppendorf tube, respectively. These two solutions were mixed and incubated for 20 min at room temperature. The transfection mixture was applied to HT-29 cells and incubated for 7 h in a 37°C CO2 incubator. After incubation, the cells were washed with 10% FCS–DMEM three times and cultured further for 18 h. The cells were washed with 10% FCS–DMEM and stimulated with 100 µg/ml of poly I:C for 3 h. The expression of hBD-2 was examined by RT–PCR. The expression level of TLR-3 was confirmed by immunoprecipitation followed by Western blotting. For immunoprecipitation, 1 mg of total protein was mixed with 1·5 µl of anti-TLR-3 antibody and incubated for 18 h at 4°C. Ten µl of protein G-sepharose (GE Healthcare, Tokyo, Japan) was added to samples and incubated for a further 1 h at 4°C. The samples were washed and subjected to Western blotting. Anti-TLR-3 antibody was diluted to ×166 with 1% bovine serum albumin (BSA)–PBS and used as a primary antibody.

Detection of NF-κB activity

For detection of NF-κB activity, HT-29 cells were stimulated with 100 µg/ml of poly I:C for 3 h. After stimulation, cell lysates were prepared with cell lysis buffer (50 mM Tris-HCl, pH 7·5, 150 mM NaCl and 0·5% TritonX-100) and the protein concentrations were measured by protein assay kit (Bio-Rad, Tokyo, Japan). For Western blotting, 100 µg of protein was subjected to 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to an Immobilon membrane (Millipore, Tokyo, Japan). The p65 subunit of NF-κB and phosphorylated p65 were detected by anti-p65 antibody and anti-phospho-p65 antibody, respectively. The membranes were incubated further with horseradish peroxidase (HRP)-conjugated goat anti-mouse immunoglobulin (Ig)G (H + L) (1:5000 dilution with 1% BSA–PBS-Tween 20 (PBST) for 1 h. After washing, the bands were detected using an ECL kit (GE Healthcare).

For the detection of NF-κB subunits translocated to the nucleus, HT-29 cells were stimulated and cell lysates were prepared. The protein concentrations were measured as above. Nuclear extracts were prepared from 100 µg of total protein using the TransFactor Extraction Kit (Clontech, CA, USA) and was subjected to Transfactor kit (Clontech). For immunoprecipitation (IP), the nuclear extracts were mixed with anti-p65 antibody and rotated for 1 h at 4°C. The samples were incubated further with protein G-Sepharose for 1 h at 4°C. After washing, the samples were mixed with 5× SDS sample buffer [250 mM Tris-HCl (pH 8·0), 10% SDS, 50% glycerol, 0·1% bromophenol blue] and boiled for 3 min. Samples were separated by 10% SDS-PAGE and subjected to Western blotting using anti-p65 antibody and HRP-conjugated goat anti-mouse IgG (H + L).

DNA construction

For the luciferase assay, 1·2 kb of the hBD-2 5′-untranslated region was amplified by PCR using genomic DNA extracted from HT-29 cells as a template. The sequences of primers (hBD-2 genomic 5′-primer and hBD-2 genomic 3′-primer) are listed in Table 1. The amplified PCR product was subcloned into the pGL4 vector (Stratagene, La Jolla, CA, USA). This plasmid is designated pGL4-hBD2 wild-type. For NF-κB binding site deletion mutants (pGL4-hBD-2-upper, pGL4-hBD-2-lower and pGL4-hBD-2-double), Quickchange II site-directed mutagenesis kit (Stratagene) was used. The sequences of the primers used for this construct are also listed in Table 1 (QCU 5′, QCU 3′, QCL 5′ and QCL 3′). The accuracy of the mutation was confirmed by DNA sequencing.

Luciferase assay

HT-29 cells were plated on 48-well culture plates at a density of 1 × 105 cells/well on the day before transfection. The cells were washed twice with OPTI-MEM and transfected with 1 µg of reporter plasmid containing the 5′-regulatory region of the hBD-2 gene (pGL4-hBD-2 wild-type, pGL4-hBD-2-upper, pGL4-hBD-2-lower and pGL4-hBD-2-double) using the lipofectin transfection method (Bio-Rad, Tokyo, Japan). After 5 h of transfection, the cells were washed with 10% FCS–DMEM and cultured further for 24 h. The cells were washed and left unstimulated or stimulated with 100 µg/ml of poly I:C for 3 h. After stimulation, the cells were lysed with 1× passive lysis buffer (Promega, Tokyo, Japan) and cell lysates were collected. Transfection efficiency was normalized to renilla luciferase activity by co-transfection with the pRL/cytomegalovirus (CMV) vector (Promega). Both firefly and renilla luciferase activities were determined using the Dual-Luciferase Reporter Assay System (Promega). Luminescence was measured on a Lumat LB9507 luminometer (Berthold, Bad Wildbad, Germany).

Statistical analysis

Statistical analysis was performed using a two-tailed Student's t-test. P levels < 0·05 were considered as significant.

Results

Expression of hBD-2 is up-regulated by poly I:C stimulation

We first attempted to determine whether poly I:C stimulation can induce the expression of hBD-2 in HT-29 cells. HT-29 cells were plated at a density of 5 × 105/35-mm dish and stimulated with 100 µg/ml of poly I:C for 3 h. Total RNA was extracted and subjected to RT–PCR. The expression of hBD-2 was detected readily after 3 h of stimulation (Fig. 1a). Without poly I:C stimulation, hBD-2 mRNA was absent. Moreover, hBD-2 expression was not observed by poly dI:dC stimulation, indicating the poly I:C-dependent induction of hBD-2. Poly I:C dose-dependency was examined by incubating HT-29 cells with varying concentrations (0–1000 µg/ml) of poly I:C for 3 h. As shown in Fig. 1b, hBD-2 was induced dose-dependently.

Figure 1.

Polyinosinic-polycytidylic acid (poly I:C) stimulation induces the expression of human beta-defensin 2 (hBD-2) in HT-29 cell. (a) HT-29 cells were plated on a 35-mm culture dish at a density of 3 × 105. The cells were stimulated with 100 µg/ml of poly I:C or 1 unit/ml of poly dI:dC for 3 h. The total RNA was extracted and subjected to complementary DNA synthesis. The expression of hBD-2 was examined by reverse transcriptase–polymerase chain reaction (RT–PCR). As an internal control, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was amplified. (b) HT-29 cells were stimulated with varying concentrations (0–1000 µg/ml) of poly I:C for 3 h. (c) HT-29 cells were stimulated with 100 µg/ml of poly I:C for the indicated times. The expression of hBD-2 was examined by real-time PCR. The error bars (b, c) represent the mean ± standard deviation (n = 3–4).

Based on these results, we next examined the time–course of hBD-2 up-regulation. The expression of hBD-2 was detected after 3 h of stimulation and the expression level was maintained until 9 h of stimulation (Fig. 1c). Peak induction was observed at 12 h of stimulation, when the expression level was fivefold higher than at 3 h post-stimulation. Expression decreased rapidly thereafter. Even after 24 h of stimulation, the expression of hBD-2 mRNA was equivalent to the levels observed at 3 h of stimulation and the low-level expression was observed after 48 h of stimulation. These results indicated that hBD-2 expression was augmented clearly in HT-29 cells by poly I:C stimulation.

TLR-3-dependent induction of hBD-2

HT-29 cells express TLR-3, suggesting that the poly I:C signal was transduced by TLR-3. To explore this possibility, HT-29 cells were pre-incubated with anti-TLR-3 neutralizing antibody or an isotype-matched control antibody for 1 h at 37°C and then stimulated with poly I:C for 3 h. Pre-incubation of the cells with control antibody did not affect the expression of hBD-2. In contrast, anti-TLR-3 antibody pre-incubation reduced the poly I:C-stimulated expression of hBD-2 (Fig. 2a).

Figure 2.

Polyinosinic-polycytidylic acid (poly I:C) signal was sensed by Toll-like receptor (TLR)-3. (a) HT-29 cells were pre-incubated with 2 µg/ml of anti-TLR-3 antibody or with the class-matched control antibody for 1 h at 37°C. The cells were then stimulated with poly I:C for 3 h. The expression of human beta-defensin 2 (hBD-2) was measured by reverse transcriptase–polymerase chain reaction (RT–PCR). The expression of hBD-2 in the non-treated cell was set to 100% and the expression level of hBD-2 in each condition was expressed as the ratio to this value. (b) SiRNA transfection was performed as described in Materials and methods. The expression level of TLR-3 was examined by immunoprecipitation followed by Western blotting (upper panel). Ten µg of total protein of each cell extract was subjected for Western blotting as internal control (lower panel). (c) After siRNA transfection, the cells were stimulated with poly I:C for 3 h. HBD-2 expression levels were measured by RT–PCR and the expression level in non-treated cell was set as 100%. (d) The time-dependent induction of TLR-3 was examined by RT–PCR. The expression level of endogenous TLR-3 was set to 1. The induction of TLR-3 compared to the endogenous level was expressed as fold induction. For (a) and (c), the statistical analysis was performed. *P < 0·05. The error bars (d) represent the mean ± standard deviation (n = 4).

TLR-3-dependent induction of hBD-2 was examined further by siRNA experiments. HT-29 cells were transfected with either control or TLR-3 siRNA. The expression level of TLR-3 was examined by immunoprecipitation followed by Western blotting. Significant reduction of TLR-3 was observed by TLR-3 siRNA transfection (Fig. 2b). Under these conditions, the cells were stimulated further with poly I:C for 3 h and hBD-2 induction was observed with RT–PCR. As shown in Fig. 2c, TLR-3 siRNA transfection reduced significantly the induction of hBD-2 by poly I:C stimulation. In contrast, control siRNA transfection augmented hBD-2 expression. These results indicated that the transduction of the poly I:C signal was dependent on TLR-3 in HT-29 cells. To investigate further whether poly I:C stimulation could induce its cognate receptor, TLR-3, RT–PCR was performed. HT-29 cells express endogenous TLR-3 in the resting state (Fig. 2d). Poly I:C stimulation augmented the expression of TLR-3 mRNA gradually (Fig. 2d), and peak induction was observed at 12 h of stimulation. The extent of induction was sevenfold higher than the endogenous expression level. These results indicated that poly I:C up-regulated hBD-2 via TLR-3 on the surface of HT-29 cells, and that poly I:C stimulation could augment the expression of TLR-3 on HT-29 cells.

Poly I:C signalling is dependent on NF-κB

The binding of poly I:C to TLR-3 has been shown to activate several transcription factors [21]. We focused on NF-κB because hBD-2 is a well-known target gene for NF-κB [16]. To test the dependency of poly I:C signalling on the NF-κB pathway, we used specific NF-κB inhibitors, TPCK and isohelenine. HT-29 cells were pre-incubated with either TPCK or isohelenin for 30 min and 2 h, respectively. The cells were washed and stimulated with poly I:C for 3 h. After stimulation, total RNA was extracted and subjected to RT–PCR. Both TPCK and isohelenin pretreatment completely blocked poly I:C stimulation of hBD-2 mRNA expression (Fig. 3a).

Figure 3.

Polyinosinic-polycytidylic acid (poly I:C) signal was transduced by nuclear factor kappa B (NF-κB). (a) HT-29 cells were incubated with L-1-4′-tosylamino-phenylethyl-chloromethyl ketone (TPCK) or isohelenine for 30 min or 2 h, respectively. After washing, the cells were stimulated with poly I:C for 3 h. The expression of human beta-defensin 2 (hBD-2) was measured by reverse transcriptase–polymerase chain reaction (RT–PCR). (b) Schematic illustration of the reporter construct for the luciferase assay. pGL4-hBD-2 wild-type contains 1·2 kb of the hBD-2 5′-regulatory region. This region contains two NF-κB binding motifs. Each site was deleted by site-directed mutagenesis and designated as pGL4-hBD-2-upper, pGL4-hBD-2-lower and pGL4-hBD-2-double, respectively. Nucleotide numbering is relative to the transcription initiation site TGA, where G is +1. (c) HT-29 cells were transfected with the above plasmids along with renilla luciferase plasmid, pRL/cytomegalovirus (CMV). After 24 h of transfection, the cells were stimulated with poly I:C for 3 h. The luciferase activity was determined as the ratio of the activities of firefly and renilla luciferase. For (a) and (c), the statistical analysis was performed. *P < 0·05.

To confirm further the specific contribution of NF-κB to the induction of hBD-2, the 5′ regulatory region of the hBD-2 gene was cloned and fused to the luciferase gene (pGL4-hBD-2 wild-type), and a luciferase assay was performed. Poly I:C stimulation increased the luciferase activity by five- to ninefold (Fig. 3c). The 5′ regulatory region encompasses two NF-κB-binding sites (Fig. 3b). To determine which binding site contributes mainly to the poly I:C stimulation of hBD-2 expression, each site was deleted by site-directed mutagenesis and then used in a luciferase assay. As shown in Fig. 3c, the luciferase activity was equivalent to that of the wild-type construct when the distal NF-κB binding site was deleted. Conversely, no luciferase activity was detected using the mutant with the proximal site deleted. Moreover, the deletion mutant lacking both distal and proximal NF-κB-binding sites showed the background level of luciferase activity. These results demonstrated that the proximal NF-κB binding site contributes mainly to poly I:C stimulation of hBD-2 expression.

Poly I:C stimulation leads to nuclear translocation of the NF-κB subunits p65 and p50

Poly I:C stimulation is known to activate the NF-κB signalling pathway. To investigate which subunits of NF-κB are activated mainly by poly I:C stimulation, we performed a TransFactor assay (Clontech). The cells were stimulated with or without poly I:C for 3 h. After stimulation, nuclear extracts were prepared and the DNA binding activity of each NF-κB subunit was estimated. In the absence of poly I:C stimulation, a very low amount of NF-κB binding was detected. Rel-B was the most abundant NF-κB subunit in the nuclei of resting HT-29 cells. In contrast, poly I:C stimulation drastically induced the translocation of p65 and p50 subunits to the nucleus (Fig. 4a). The optical density (OD) value of p65 binding was increased fivefold over the amount in resting cells, and the OD value of p50 binding increased fourfold after stimulation. To confirm these results, p65 was immunoprecipitated from the nuclear extracts with an anti-p65 specific antibody and detected by Western blotting. P65 was detected in the nuclear extract of the resting HT-29 cells; however, poly I:C stimulation increased the amount of nuclear p65 significantly (Fig. 4b). The p65 subunit was absent from the samples incubated only with protein G-Sepharose, indicating the specific detection of the p65 subunit.

Figure 4.

P65 and p50 subunits of nuclear factor kappa B (NF-κB) are translocated to the nucleus by polyinosinic-polycytidylic acid (poly I:C) stimulation. (a) HT-29 cells were stimulated with poly I:C for 3 h. After stimulation, the nuclear extracts were prepared and subjected to TransFactor kit to measure the amount of each NF-κB subunit in the nuclear extracts. (b) The nuclear extracts were incubated with protein G-sepharose alone or in combination with anti-p65 antibody. The sample were washed and subjected to Western blotting with anti-p65 antibody as the primary antibody. The error bars (a) represent the mean ± standard deviation (n = 3).

Upon activation, p65 must be phosphorylated and released from its binding partner IκB in the cytoplasm. We further examined the phosphorylation states of p65 after poly I:C stimulation. The total amount of p65 remained constant throughout the stimulation period (Fig. 5, middle). No phosphorylation of p65 at Ser536 was observed in the resting state. A very low amount of phosphorylated p65 could be detected after 10 min of stimulation. The extent of phosphorylation increased according to the stimulation time, and even after 6 h of stimulation, significant phosphorylation was observed. These results indicate that poly I:C stimulation of HT-29 cells leads to the time-dependent phosphorylation of p65, which was maintained for at least 6 h.

Figure 5.

P65 (serine 563) phosphorylation was induced by polyinosinic-polycytidylic acid (poly I:C) stimulation. (a) HT-29 cells were stimulated with poly I:C for the indicated lengths of time. After stimulation, the cell extracts were prepared and subjected to Western blotting. (b) The density of phospho-p65 and total p65 bands were measured by NIH Image. The phospho p65/p65 ratio was calculated and the ratio at 360 min was set as 100%.

Discussion

Innate immunity has captured much attention since it was discovered that TLRs play a pivotal role in the detection of invading pathogens by recognizing pathogen-associated molecular patterns. TLR activation transduces signals to the cellular nucleus, resulting in the mobilization of anti-microbial effectors [21–24]. Among the effector molecules of innate immunity, defensins were shown previously to exert inhibitory activities against bacterial and viral infections [3–5]. In the present study, we focused on the expression of hBD-2 in IECs.

Expression of hBD-2 in HT-29 cells was only detectable after poly I:C stimulation. These data are consistent with other reports demonstrating the stimulation-dependent expression of hBD-2. A variety of microbial compounds and proinflammatory factors have been reported to induce the expression of hBD-2 in various epithelial cells [3–5]. In IECs, hBD-2 expression was induced by IL-1α, Salmonella infection [14], Escherichia coli K232-derived lipopolysaccharide, Staphylococcus aureus-derived peptidoglycan [19] and a probiotic strain of Escherichia coli[25]. Poly I:C is an artificial analogue of dsRNA, a by-product of viral replication. Although the poly I:C-induced expression of hBD-2 has been reported in epithelial cells of the female reproductive tract [26] and fallopian tube [27], there have been no studies concerning poly I:C-induced expression of hBD-2 in IECs. In most cell types, the expression of hBD-2 was induced only by proinflammatory stimuli. The only exception to this is oral epithelial cells, in which hBD-2 is expressed constitutively [13,28]. Intriguingly, anti-viral activity of hBD-2 against human immunodeficiency virus (HIV) was demonstrated in oral mucosa [28,29]. To date, some viruses have been reported to induce the expression of hBD-2, such as human papilloma virus [30], rhinovirus [31] and HIV [13,28,29]. Generally, many viruses have developed strategies to evade and subvert the host immune system [32]. For instance, rotaviruses, which can infect IECs of the small intestine, prevent the production of interferons by degrading the transcription factors responsible for interferon expression. This evasion reaction is mediated by rotavirus-derived non-structural protein 1, which targets the transcription factors for proteasomal degradation [33]. The viruses that induce the expression of hBD-2 in IECs and the role of IEC-derived hBD-2 in the anti-viral response should be examined in future studies.

There are several different receptors for dsRNA–TLR-3, and retinoic acid-inducible gene I-like receptors (RLRs), such as RIG-I and MDA5. TLR-3 localization differs depending in cell type [34]. By using a specific monoclonal antibody, TLR-3 was shown to localize both to the endosomal membrane inside the cells as well as to the plasma membrane [34,35]. In fibroblasts, cell surface-expressed TLR-3 is responsible for the production of type I IFNs in response to dsRNA, and pre-incubation with an anti-TLR-3 antibody was found to inhibit dsRNA-stimulated IFN production [34]. In addition, endosomal TLR-3 is responsible for dendritic cell maturation in response to dsRNA [35].

In the present study, anti-TLR-3 antibody and TLR-3 siRNA transfection partially blocked poly I:C-induced expression of hBD-2 after 3 h of stimulation. TLR-3 is thought to be responsible for the early recognition of the poly I:C signal; therefore, the mechanisms responsible for the relatively late induction of hBD-2 (12 h after stimulation) are unclear. One possibility might be the recognition of poly I:C by cytoplasmic RLRs. Poly I:C has been shown to be recognized preferentially by MDA5 [36]. In fact, transfection with poly I:C can induce the anti-viral response much more rapidly and more strongly than the simple addition of poly I:C to the culture media [37]. In this study, we added poly I:C to the culture media; thus the peak in hBD-2 induction at 12 h could be attributed to the recognition of permeabilized poly I:C by intracellular receptors within the cytoplasmic space.

In contrast to the biphasic induction of hBD-2, the expression of TLR-3 mRNA increased gradually after poly I:C stimulation and peaked at 12 h after stimulation. Although the signalling cascade underlying this phenomenon is not known, several cytokines were demonstrated to control the expression of TLR-3 [38,39]. In active Crohn's disease, the lower expression level of TLR-3 was reported to be due to an imbalance in the production of T helper type 1 (Th1) and Th2 cytokines [40,41]. The strict regulation of TLR expression is indispensable for the maintenance of the integrity of the intestinal environment. It will be important to examine the mechanisms for dsRNA-stimulated TLR-3 expression in IECs.

The signalling of TLR-3 is unique among TLRs. Other TLRs utilize MyD88 as an adaptor molecule; however, TLR-3 can signal through the Toll/IL-1 receptor domain-containing adaptor inducing IFN-β (TRIF), and can activate the transcription factors NF-κB and interferon regulatory factor 3 (IRF3) [21]. In the female reproductive tract, dsRNA stimulation was shown to induce hBD-2 production via TLR-3 and NF-κB, which resulted in the production of anti-viral agents such as IFNs, hBDs and chemokines [26]. Based on these findings, we examined the contribution of NF-κB to poly I:C-induced expression of hBD-2 in IECs. Both of the specific NF-κB inhibitors, TPCK and isohelenine, completely abrogated the induction of hBD-2, indicating the importance of NF-κB in this pathway. The significance of NF-κB for hBD-2 induction has been demonstrated in other systems. The 5′-regulatory region of the hBD-2 gene contains two NF-κB binding sites [25]. The critical importance of the proximal NF-κB binding site was demonstrated in the probiotic-induced hBD-2 expression system [25]. Our results confirm the pivotal function of the proximal NF-κB binding site in poly I:C-induced hBD-2 expression. Another transcription factor downstream of TLR-3 activation is IRF3. However, to date, the role of IRF3 in the expression of hBD-2 has not been demonstrated, because there is no IRF-3 binding site in the 5′-regulatory region of the hBD-2 gene.

We further investigated the subunit composition of NF-κB activated by poly I:C stimulation. The enzyme-linked immunosorbent assay (ELISA)-based TransFactor assay demonstrated the clear accumulation of both p65 and p50 in the nuclei of poly I:C stimulated HT-29 cells. As demonstrated by immunoprecipitation, some p65 subunit accumulation was detected even in resting HT-29 cells. The frequently observed high background of NF-κB activity in HT-29 cells might therefore be attributed to the accumulation of the p65 subunit in the resting nucleus.

The activation of p65 was determined by examining its phosphorylation status. The phosphorylation of p65 increased according to the duration of poly I:C stimulation. The most intense phosphorylation was detected at 6 h post-stimulation. These results are consistent with results of other experimental systems. A relatively long duration of p65 phosphorylation was demonstrated in HT-29 cells stimulated with the peroxisome proliferator activated receptor γ ligand [42]. In our experiment, HT-29 cells were stimulated continuously with a relatively high concentration of poly I:C, which may have caused the prolonged activation of p65.

The results obtained in this study demonstrate clearly that poly I:C induces the production of the innate immune effector, hBD-2, in IECs. The poly I:C signal was sensed by its cognate receptor, TLR-3, and transduced by transcription factor NF-κB. As demonstrated in other studies, IECs can induce both innate and adaptive immunity in response to poly I:C. The functional coupling of both systems might be indispensable for the protection of the intestinal environment from invading pathogens.

Conclusions

In the present study, poly I:C-induced expression of hBD-2 in an IEC line, HT-29, was demonstrated. Poly I:C was sensed by TLR-3 on the cell surface and the poly I:C-binding signal was transduced by the transcription factor, NF-κB. As the regulatory region of the hBD-2 gene contains two NF-κB binding sites, we performed a luciferase assay to examine which binding site contributes mainly to the induction of hBD-2 in HT-29 cells. Luciferase activity was reduced drastically when the proximal NF-κB-binding site was deleted. Moreover, poly I:C stimulation led to the accumulation of p65 and p50 subunits within the nucleus. Poly I:C also resulted in the prolonged phosphorylation of the p65 subunit. These results demonstrate that dsRNA, a by-product of viral replication, can be sensed by IECs, which then respond to viral infection by secreting the anti-microbial peptide hBD-2. The intrinsic importance of the transcription factor NF-κB in this process was demonstrated clearly.

Acknowledgements

This work was supported by the Academic Frontier Project for Private Universities; matching fund subsidy from MEXT 2006–10 (to Dr Hiroki Nagase); Nihon University Joint Research Grant for 2010–12 (to Dr Okayama); a grant from the Strategic Research Base Development Program for Private Universities from Ministry of Education, Culture, Sports, Science, and Technology, Japan (MEXT), 2010–14 (S1001024); Health and Labour Sciences research grants and research on international cooperation in medical science; and the Promotion and Mutual Aid Corporation for Private Schools of Japan (2011).

Disclosure

The authors declare no competing financial interests.

Ancillary