Department of Internal Medicine, Division of Gastroenterology, Seoul National University Boramae Hospital, Seoul National University College of Medicine, Seoul, Korea.
Department of Otorhinolaryngology-Head and Neck Surgery, Seoul National University Boramae Hospital, Seoul National University College of Medicine, Seoul, Korea.
Department of Pathology, Seoul National University Boramae Hospital, Seoul National University College of Medicine, Seoul, Korea.
Department of Internal Medicine and Liver Research Institute, Seoul National University College of Medicine, Seoul, Korea.
Abstract
Periostin is a matricellular protein that interacts with various integrin molecules on the cell surface. Although periostin is expressed in inflamed colonic mucosa, its role in the regulation of intestinal inflammation remains unclear. We investigated the role of periostin in intestinal inflammation using Postn-deficient (Postn-/-) mice. Intestinal epithelial cells (IECs) were transfected by Postn small interfering RNAs. Periostin expression was determined in colon tissue samples from ulcerative colitis (UC) patients. Oral administration of dextran sulfate sodium (DSS) or rectal administration of trinitrobenzene sulfonic acid, induced severe colitis in wild-type mice, but not in Postn-/- mice. Administration of recombinant periostin induced colitis in Postn-/- mice. The periostin neutralizing-antibody ameliorated the severity of colitis in DSS-treated wild-type mice. Silencing of Postn inhibited inteleukin (IL)-8 mRNA expression and NF-κB DNA-binding activity in IECs. Tumor necrosis factor (TNF)-α upregulated mRNA expression of Postn in IECs, and recombinant periostin strongly enhanced IL-8 expression in combination with TNF-α, which was suppressed by an antibody against integrin αv (CD51). Periostin and CD51 were expressed at significantly higher levels in UC patients than in controls. Periostin mediates intestinal inflammation through the activation of NF-κB signaling, which suggests that periostin is a potential therapeutic target for inflammatory bowel disease.
Figures
Fig 1 Periostin mediates intestinal inflammation in murine models of colitis. (A and B) Mice were divided into four groups in the DSS-induced colitis model (n = 5): wild-type control, Postn-/- control, wild-type treated with DSS, or Postn-/- treated with DSS. Postn-/- mice showed significant attenuation in the body weight reduction and the disease activity index three days after DSS administration. (C) Histological evaluations of the colons in the DSS-induced acute murine colitis model. The total histological scores were derived from the severity and extent of inflammation and crypt damage (mean ± SD). Results are representative of at least three separately examined sites. (D) Histological evaluations of colons in the TNBS-induced colitis (mean ± SD). Mice were divided into four groups in the TNBS-induced colitis model: wild-type control (n = 4), Postn-/- control (n = 4), wild-type treated with TNBS (n = 5), or Postn-/- treated with TNBS (n = 7). Results are representative of at least three separately examined sites. (E) Immunoreactivity index for phosphorylated IκB kinase (pIKK)-α/β (mean ± SD) of the colonic epithelium in DSS-induced murine colitis. DSS exposure in wild-type mice induced pIKK activity in the colonic epithelium. Periostin deficiency significantly attenuated phosphorylated pIKK-α/β activity in the colonic mucosa. (F) Representative H&E staining and immunostained images of pIKK-α/β and periostin in the colon tissues (Magnification x 200). These results are representative of three independent experiments. WT, wild-type; Postn-/-, Postn-deficient; DSS, dextran sulfate sodium; TNBS, trinitrobenzene sulfonic acid; pIKK, phosphorylated-IκB kinase *P < 0.05 compared with wild-type mice without DSS or TNBS, #P < 0.05 compared with wild-type mice treated with DSS or TNBS.Fig 2 Administration of recombinant periostin induces colitis in Postn-/- Mice. Mice were divided into three groups (n = 5): wild-type treated with PBS, Postn-/- treated with PBS, or Postn-/- treated with recombinant periostin (10 μg). Recombinant mouse periostin was dissolved in 100 μL of phosphate-buffered saline (PBS) and intraperitoneally administered every second day, commencing on the day of DSS administration (day 0). Control mice were intraperitoneally injected with 100 μL of vehicle according to the same schedule. (A) Body weight change after DSS administration. (B) Changes in the disease activity index after DSS administration. Administration of recombinant periostin resulted in a significant reduction in body weight, along with increased DAI when compared with that seen for control Postn-/- mice treated with PBS. (C) Extracted colon length at day 6 after DSS administration. (D) Histological evaluations of colons in the DSS-induced acute murine colitis model (mean ± SD). The total histological scores were derived from the severity and extent of inflammation and crypt damage. (E) Results are representative H&E staining of at least three separately examined sites (Magnification x 100). These results are representative of two independent experiments. WT, wild-type; Postn-/-, Postn-deficient; PBS, phosphate-buffered saline; IP, intraperitoneal injection; DSS, dextran sulfate sodium *P < 0.05 compared with Postn-/- mice treated with recombinant periostin, #P < 0.05 compared with wild-type mice treated with DSS.Fig 3 Periostin-neutralizing antibody attenuates acute murine colitis in wild-type mice. Mice were divided into two groups (n = 5). A periostin-neutralizing antibody (nAb) was dissolved in 100 μL of PBS and intraperitoneally administered to mice each day for 5 days, commencing on the day of DSS administration (day 0). Control mice were administered a mouse IgG isotype control antibody each day for 5 days. (A) Change in the disease activity index after DSS administration. Periostin nAb significantly attenuated disease activity index, compared with mice treated with isotype IgG control antibody. (B and C) The total histological scores were derived from the severity and extent of inflammation and crypt damage (mean ± SD). Results are representative of at least three separately examined sites. These results are representative of two independent experiments. WT, wild-type; Postn-/-, Postn-deficient; PBS, phosphate-buffered saline; Isotype IgG, Isotype IgG control antibody; anti-periostin Ab, periostin-neutralizing antibody; DSS, dextran sulfate sodium; X100, magnification x100; X200, magnification x200 * P < 0.05 compared with isotype IgG control antibody.Fig 4Postn silencing reduces IL-8 expression by inhibiting NF-κB signaling in intestinal epithelial cells. COLO205 cells were transfected with control or Postn-specific siRNA for 24 h. (A) COLO205 cells were stimulated with TNF-α (10 ng/mL) for 4 h. IL-8 mRNA expression was measured by real-time RT-PCR. Levels were normalized to β-actin. Data are expressed as fold-change in mRNA transcript levels relative to the unstimulated control (mean ± SD, n = 3). (B) COLO205 cells were stimulated with TNF-α (10 ng/ml) for 24 h. Secretion of IL-8 was measured by ELISA (mean ± SD, n = 3). (C) COLO205 cells were stimulated with TNF-α (10 ng/ml) for 30 min. NF-κB DNA binding activity in the nuclear extracts was assessed by electrophoretic mobility shift assay. TNF-α, tumor necrosis factor-α; siRNA, small interfering ribonucleic acid * P < 0.05 compared with control siRNA without TNF-α stimulation, #P < 0.05 < compared with control siRNA stimulated with TNF-α.Fig 5 Periostin enhances proinflammatory cytokine production via interaction with integrin in intestinal epithelial cells. (A) COLO205 cells were stimulated with TNF-α (10 ng/mL or 50 ng/mL) for 4 h. Postn mRNA expression was measured by real-time RT-PCR. Levels were normalized to β-actin. Data are expressed as fold-change in mRNA transcript levels relative to the unstimulated control (mean ± SD, n = 3). (B) COLO205 cells were stimulated with recombinant periostin (100 μg/mL or 500 μg/mL) for 24 h. IL-8 mRNA expression was measured by real-time RT-PCR (mean ± SD, n = 3). (C) COLO205 cells were pretreated with recombinant periostin for 24 h and then stimulated with TNF-α (10 ng/mL) for 4 hr. IL-8 mRNA expression was measured by real-time RT-PCR (mean ± SD, n = 3). (D) COLO205 cells were pretreated with anti-integrin αv (CD51) antibody and recombinant periostin for 24 h and then stimulated with TNF-α (10 ng/mL) for 4 h (mean ± SD, n = 3). TNF-α, tumor necrosis factor-α; Anti-CD51, anti-integrin αv antibody * P < 0.05 compared with negative control (without stimulation), ** P < 0.05 compared with TNF-α stimulation alone, #P < 0.05 compared with co-stimulation with TNF-α and recombinant periostin.Fig 6 Periostin is highly expressed in the colon tissues of ulcerative colitis patients. Tissue specimens were stained immunohistochemically with periostin and integrin αv(CD51). (A) Representative immunostained images of periostin in normal subjects and UC patients. Patients with ulcerative colitis exhibited linear staining of periostin along the subepithelial border (arrow), and a simultaneous increase in the immunoreactivity of CD51, periostin’s functional receptor, in the colonic epithelium where periostin was expressed (arrowhead). (B) The scores of periostin immunoreactivity were significantly increased in colonic samples of ulcerative colitis patients (mean ± SD). CD51, integrin αv; UC, ulcerative colitis; X200, magnification x 200; X400, magnification x 400 * P<0.05 compared with controls.
그림 출처: PubMed Central 공개 논문 (PMC4758640)
Conflict of interest statement
Competing Interests:The authors have declared that no competing interests exist.
