|Home | About | Journals | Submit | Contact Us | Français|
Inflammatory bowel disease (IBD) is characterized by chronic inflammation of the gastrointestinal tract. Epstein-Barr virus (EBV) infection is associated with increased disease severity in therapeutically immunosuppressed IBD patients. The role of EBV infection in patients with IBD who are unresponsive to medical therapy is unclear. Anti-viral strategies may be a viable treatment option if severity of EBV infection, reflected in peripheral blood, contributes to IBD progression.
We investigated the role of EBV in IBD patients unresponsive to medical therapy by examining EBV reactivation and B-cell proliferation in colonic mucosa.
EBV DNA copy numbers were measured by real-time PCR in peripheral blood mononuclear cells (PBMC) of 84 patients with IBD and 115 non-IBD controls in a retrospective cross-sectional study. EBV-infected cells in colonic mucosa were identified by immunohistochemistry.
EBV load in PBMC was higher in patients with IBD than in non-IBD controls, especially in patients not responding to medication. Inflamed colonic mucosa of these patients had high levels of expression of lytic and latent EBV genes that localized to proliferating B-lymphocytes, which was not seen in patients responding to therapy.
EBV replication was associated with severe IBD and mucosal inflammation. Increased proliferation and EBV infection of B-lymphocytes in inflamed colonic mucosa highlight the potential role of EBV in mucosal inflammation. The immunomodulatory effects of EBV could delay the resolution of the IBD associated inflammation, thus contributing to disease progression. These results indicate that anti-viral therapeutic strategies for the resolution of IBD may be useful.
IBD is a complex polygenetic disease manifested by chronic inflammation of the intestinal mucosa.1–6 Patients with severe disease that cannot be managed by diet or other means receive intensive therapy consisting of immunomodulators and biologics, individually or in combination.7–11 Anti-tumor necrosis factor (TNF)α therapy (Infliximab) is effective in producing disease remission. 8, 12, 13 More than 95% of the general population is infected by EBV, which is a herpes virus. EBV can persist for the lifetime of a host and cause lymphoproliferative disorders such as lymphomas.14, 15,16 EBV infection has been implicated in the etiology of IBD. Reactivation and replication of EBV during immunosuppressive therapy may be important for the increased incidence of lymphoma in IBD patients.17
EBV is harbored in resting memory B-cells and induces B-cell proliferation.18 The immune system of an immunocompetent host can identify and eliminate EBV-infected cells, but immunosuppressed individuals are more susceptible to EBV-induced immunoblastic proliferation.16, 19 The increased magnitude of EBV-induced reactivation in B cells raises the risk for the development of lymphoma.19–22 Herpes virus infections (eg, cytomegalovirus) during anti-TNFα therapy are a major cause of IBD disease progression and are refractory to medical management.23–25 A high EBV load is a significant risk factor for developing post-transplant lymphoproliferative disorder and progression of IBD.19–22 The role of EBV in IBD that is refractory to conventional therapy has not been completely defined and warrants further investigation.16, 19 We hypothesize that EBV replication following reactivation in colonic mucosa could be reflected in peripheral blood and may be associated with the exacerbation of IBD.
The goal of this retrospective cross-sectional study was to examine the level of EBV in the peripheral blood mononuclear cells (PBMC) and in inflamed colonic mucosal sites in patients with IBD.
Patients with IBD (n=84) and individuals with no prior history of IBD (n=115) were enrolled in the study at the University of California, Davis. The IBD patient group (consisting of 69 patients on medical management and 15 requiring surgical management) were identified based on their clinical, endoscopic and histological evaluations (Table 1). Of the controls, 75 individuals had no known health concerns and 40 had non-IBD inflammatory conditions (4 with rheumatoid arthritis (RA) and 36 with primary biliary cirrhosis (PBC)). Peripheral blood samples from all participants were obtained at a single time-point. Concurrent peripheral blood and perioperative gut resection or colonic mucosal biopsy samples from IBD patients were obtained. DNA was extracted from PBMC (Ficoll-Hypaque density gradient centrifugation) using the Qiagen DNeasy extraction kit (Qiagen, Valencia).26 Mucosal samples were stored in CryoPrep (American Master Tech Scientific, Lodi). This study protocol was approved by the institutional review board at the University of California, Davis.
PCR primers and a fluorogenic probe were generated based on the conserved BALF5 region of the EBV genome as previosly described.27 DNA isolated from a known number of cells was analyzed by real-time PCR assay using an ABI Prism 7900 sequence detector (Applied Biosystems) 27; this assay has been previously validated.28 Cell numbers were confirmed by a standard curve constructed using the CCR5 single copy gene and a known number of cells.
Tissue sections were obtained from colon biopsies of five patients who were responsive to medical management and from colonic resections of three patients who were refractory to medical therapy. Sections were immunostained as previously described.29 Antibodies were used to detect EBV nuclear antigen (EBNA1, present in latent and lytic cycle), EBV viral capsid antigen (VCA, present in lytic cycle), Ki67, a cell proliferation marker, and CD19, a B cell specific cell surface marker (Santa Cruz Biotech. Inc., Santa Cruz). Tissue sections with no primary antibody were used as negative controls. Images were captured by confocal laser microscopy using LSM 5 and PASCAL software (Zeiss, New York). Cell counts were performed as previously published.26
The EBV viral loads were analyzed using unpaired t test, chi-square analysis and one-way analysis of variance (ANOVA) with appropriate post-tests. Statistical software included GraphPad Prism version 5.00 for Windows (GraphPad Software, San Diego) and JMP (JMP Genomics, Cary).
The IBD patient group (n=84) consisted of 39.3% male and 60.7% female, ranged in age from 20 to 91 years (mean, 46.5 years; Table 1). Most patients had Crohn’s Disease (CD) (90.5%, n=76), and the remaining had ulcerative colitis (UC) (9.5%, n=8). All patients were on combination therapy for IBD, with 52% receiving infliximab, 18% prednisone and other immunosuppressive medications including methotextrate (3.6%); and azathioprine, 6-mercaptopurine, and 6-thioguanine (45%) Sixty-nine patients (82%) responded well to medical management (the “medical group” MG); the other fifteen (18%) had severe colitis that was not responsive to immunosuppressive therapy and required surgical intervention (gut resection) for control of IBD symptoms (the “surgical group” SG). The decision to initiate treatment of IBD was based on clinical parameters, systemic manifestations, and quality of life.30
Pathologic evaluation of colonic resections and biopsies was performed. Three patients who required surgical resection had refractory ulcerative colitis (n=3), and pathologic evaluation of gut resections indicated severe disease. Microscopic examination identified well-defined areas of chronic and focal acute inflammation in colonic mucosal samples with increased inflammatory response and cellular infiltrate in the lamina propria, as well as lymphoid hyperplasia, distorted mucus-depleted crypts, cryptitis and crypt abscesses. Among CD patients (n=12) in the surgical group, disease severity was evident by the appearance of reactive lymphoid hyperplasia in the colonic mucosa with focal cryptitis. No enteric pathogens or neoplastic changes were detected in any cases.
Higher levels of EBV DNA copy numbers were detected in PBMC of IBD patients than in non-IBD controls (Figure 1 and Table 2). Patients were grouped based on the number of viral copies/105 PBMC and categorized as: none detected (below the limit of detection of 40 copies/105 PBMC), low (40–1000 copies/105 PBMC), and high (> 1000 copies/105 PBMC)31. EBV positivity, defined as having a detectible level of EBV DNA, occurred more frequently in IBD patients than in non-IBD controls (44% vs 31%). The patients who failed conventional therapies had the highest occurrence of EBV positivity in peripheral blood (53%). Of IBD patients in the surgical group, 47% had high EBV loads vs only 6% in patients responsive to medical management (Figure 1). Average viral load was 3.5 × 108 DNA copies/105 PBMC in IBD patients vs 3.8 × 102 in the combined control group, and 7.16 × 108 in the surgical group of IBD patients vs 4.1 × 103 in the medical group (Figure 2). These results indicate that EBV loads in peripheral blood may reflect increased viral activity in the colonic mucosa of IBD patients
To investigate the association between mucosal EBV replication and inflammation, immunohistochemical analysis was performed to detect the presence of EBV antigens associated with the lytic and latent phases of the EBV life cycle (Figure 3). High levels of EBV-infected cells (> 50/mm2 EBNA1+ and > 15/mm2 VCA+ cells) were detected in colonic samples from patients undergoing surgical resection (Figure 3). The level of latently infected cells, as indicated by the presence of EBV EBNA1 expression, was higher than that of cells positive for the lytic gene EBV VCA (Figure 3). Patients with high EBV viral loads in peripheral blood also had elevated levels of EBV in colonic biopsies. The majority of EBV infection was localized to the CD19+ B lymphocytes, as observed in double immuno-stained tissue sections (Figure 3F). Thus, peripheral blood EBV loads are reflective of mucosal EBV replication in IBD patients. Furthermore, elevated colonic EBV levels are associated with increased disease activity in patients not responding to medical management.
Immunohistochemical analysis demonstrated that mucosal damage with loss of tissue architecture in inflamed colonic mucosa of IBD patients undergoing gut resection was associated with the presence of inflammatory infiltrate including B lymphocytes (Figure 4A and F) compared to patients responding to medical therapy (Figure 4C and F). Elevated B-cell levels in the inflamed region of the colonic biopsies were attributed to increased B-cell proliferation, as demonstrated by the increased expression of the cell proliferation marker, Ki67 (Figure 4B and E).32, 33 In comparison, colonic biopsies from patients with UC and CD without severe disease or with undetectable peripheral viral loads had lower levels of Ki67 expression (Figure 4D and F). Our data suggest that increased disease activity in IBD patients that is not responsive to medical therapy may be associated with elevated EBV levels resulting directly or indirectly from B cell infiltration and inflammation.
Patients with IBD have a poor quality of life due to persistent inflammation, diarrhea, bleeding, surgery, and extra-intestinal complications (abscesses, strictures, and neoplastic transformation).34 IBD is treated with disease-modifying agents that down-regulate the immune response. Treatment suppresses inflammation but may also induce loss of immune surveillance.8–11 Increased B-cell proliferation may result in worsening of IBD symptoms, suggesting that this process may be critical for inducing lymphoproliferative disease associated with IBD.35
The importance of EBV proliferation in patients with unresponsive IBD may be an integral component in its clinical management. Following EBV infection, severity of IBD may increase with or without immunosuppressive therapy.36 In this study, EBV replication, as measured by viral loads in PBMC, was higher in patients with IBD than in non-IBD controls. Increased levels of viral replication in the gut were associated with B-cell infiltration and proliferation in inflamed colonic mucosa of patients with severe IBD. Therefore, active EBV replication may be one of the factors increasing mucosal inflammation in the colonic microenvironment, which in turn could delay the resolution of inflammation in the tissue and exacerbate IBD pathology. Furthermore, EBV may drive its own replication through de novo infection of proliferating B cells in the gut mucosa in the absence of immune surveillance due to immunosuppressive therapy.19, 37 Although our retrospective pilot study did not have sufficient statistical power to identify the effects of gender and other variables, the findings suggest that EBV may play a role in mucosal inflammation in IBD and warrant a longitudinal assessment of mucosal samples. The role of the cytokine milieu, contribution of different cell types and their activation status, and the effects of immunosuppressive therapy could also contribute to disease progression.
Our findings highlight the potential role of reactivation of EBV infection in patients with progressive IBD. Inflammation plays an important role in the etiology of IBD as well as in increasing EBV loads in the gut mucosa and in peripheral blood. Our data suggest that EBV infection may contribute to further elevate B-cell proliferation, thus increasing viral load.38 Proactive viral surveillance and intervention with antiviral agents may improve patient response to therapy, but this needs to be investigated prospectively. Additionally, the magnitude of immunosuppression in patients need to considered during the clinical management. In summary, the cycle of EBV replication and B-cell proliferation in the presence of immunosuppression may augment the persistence of chronic inflammation in the colonic mucosa and reduce the responsiveness of the patient to immunosuppressive therapy.
We thank the patients for their participation in the study, the nursing staff at the UCDMC for their support and the Lucy Whittier Molecular Core Facility. We would also like to thank Dr M D George for his input in manuscript preparation. This study was supported by National Institutes of Health grants DK61297 and AI43274 and the California HIV AIDS research program grant CH05-D-606. Dr Sankaran is supported by a Building Interdisciplinary Research Careers in Women’s Health award (K12 HD051958) funded by the NICHD, ORWH, ODS, and the NIA.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of Interest
The authors have no conflict of interest in this study.
This material has not been previously presented or published