Final Progress Report
Proposal No. IBD-0105R
Principal Investigator: Klaus Michel, Ph.D. (replacement PI); Anne Rühl, M.D., Ph.D. (original PI)
Applicant Organization: Technische Universität München (Freising-Weihenstephan, Germany)
Project Title: Enteric glia in intestinal inflammation
Period of Award: July 1, 2004 – June 30, 2006
1. Summary of project aims:
The hypotheses of the research plan are as follows:
(1) Enteric glia (EGC) are involved in the regulation of mucosal defense. In vitro, co-culturing with enteric glia alters transepithelial resistance and macromolecular permeability of epithelial cell layers.
(2) EGC support epithelial cells. This is mediated by glial expression of neurotrophins, neurotrophic factors and/or epithelial growth factor and can be regulated by proinflammatory cytokines.
(3) EGC is affected in human IBD.
(4) Glial expression of neurotrophins, neurotrophic factors and/or epithelial growth factor is altered in human IBD.
2. Accomplishments towards meeting those aims:
(1) To analyze interactions between enteric glia and the intestinal epithelial barrier in un-inflamed and inflamed conditions, intestinal epithelial cells (IEC, T84) were grown on filter supports in the apical compartment of a Transwell co-culture system in the presence or absence of EGC in the basolateral compartment. Mature IEC were stimulated for 24 hrs with human recombinant tumor necrosis factor (TNF)α (1, 10, 100ng/mL), interleukin (IL)-1β (1, 10, 100ng/mL), IL-13 (10ng/mL) or interferon (IFN)γ ( (100ng/mL). Subsequently, the integrity of IEC monolayers was determined by assessing the transepithelial electrical resistance (TER) over a period of 48hrs. To detect cytokine-induced defects in the epithelial monolayers, macromolecular permeability was measured.
(2) To address the second hypothesis, we extracted RNA from EGC treated with TNFα (1, 10, 100ng/mL) or IL-1β (1, 10, 100ng/mL) or untreated controls in the Transwell co-culture system (see above). Quantitative real-time polymerase chain reaction (qRT-PCR) was performed in order to detect the glial expression of neurotrophins, neurotrophic factors, epithelial growth factor and their receptors. In addition, a potential regulation of these factors was analyzed.
(3) To answer the question whether enteric glia is affected in IBD, we first had to provide a basis on human EGC numbers and density in uninflamed control tissues, which comprised uninvolved resection margins from patients operated for gastrointestinal cancer diseases. To do this, we have determined the immunohistochemical marker profile of EGC and thereafter identified and established a monoclonal antibody against the transcription factors Sox8/9/10 as a reliable pan-glial marker in the ENS. The anti-Sox8/9/10 antibody in conjunction with the neuronal marker NSE was subsequently used to quantify EGC in whole-mount preparations of the myenteric (MP) and the submucous plexus (SMP; 3 layers in man) from normal human ileum, proximal and distal colon. Additionally, to better understand the peculiarities of the human ENS, we have also quantified the guinea pig ENS with this method, because it has traditionally been the most widely used laboratory animal in neurogastroenterological research.
For the analysis of inflamed tissues, we collected IBD samples and stained them with the Sox8/9/10 antibody. If no whole-mount preparation were technically available, we used cryosections for the quantification. This is still in progress.
(4) We have collected a considerable number of IBD samples from which RNA has been extracted for the verification of this hypothesis, but the work is still in progress.
3. Significant results:
(1) Coculturing IEC with EGC significantly enhanced TER within 48hrs (p<0.001). Immediately after TNFα-stimulation, a dose-dependent drop in TER was observed in the absence of EGC, most likely reflecting tight–junctional alterations because there were no detectable TNFα-induced increases in macromolecular permeability at any concentration or time-point. Epithelial integrity was fully preserved when IEC were treated with 10ng/ml TNFα in the presence of EGC (p<0.001 vs. stimulated IEC w/o EGC). However, EGC could only partially prevent TNFα-effects on epithelial tightness when IEC were exposed to 1 or 100ng/ml TNFα. Forty-eight hrs after TNFa exposure, all TNFα-treated IEC monolayers displayed a marked recovery with a significant rise in TER-values (p<0.005). This increase was potentiated if TNFα-stimulated EGC were present in the basolateral compartment (p<0.001). Similarly, adding EGC to IEC after TNFα-exposure significantly augmented TER values after at least 48hrs of coculture (p<0.001). Adding IL-1β to the coculture system effected less pronounced TER-alterations, whereas both IL-13 and IFNγ induced a fulminant decrease of epithelial tightness which only slowly recovered during the 48hrs observation period. If EGC were added to the IL-13- or the IFNg-treated IEC the integrity improved. In summary, the intestinal epithelial barrier appears to be supported and enhanced by EGC.
While inflammatory mediators like TNFα, IL-1β, IFNγ, and IL-13 diminish the tightness of IEC, most likely via alterations of epithelial tight junctions, these effects are prevented or mitigated in the presence of EGC. These data suggest a protective potential of EGC for the intestinal epithelial barrier in intestinal inflammation which may depend on the immunological profile of the inflammatory response.
(2) We have characterized the expression of neurotrophins, neurotrophic factors, epithelial growth factor and their receptors in EGC. We also determined expression of receptor mRNA for TNFα, IL-1β, IFNγ, and IL-13. To analyze the regulation of factors involved in the interplay of IEC and EGC, we used an advanced statistical method which includes three reference genes for the relative quantification of target gene expression. Data analysis is still in progress.
(3) The immunohistochemical marker profile of EGC comprises S100b, GFAP, p75NGFR and Sox8/9/10. Abundant immunoreactivity (IR) for these markers was detected in EGC of all studied regions. While the cytoplasmic staining pattern of most markers did not permit glial quantification, the nuclear localization of Sox8/9/10-IR allowed to identify and count all EGC individually. In both man and guinea pig, myenteric ganglia were larger and contained more EGC and neurons than submucous ganglia. Further, there were more EGC in the human than in the guinea pig myenteric plexus (MP), glial density was consistently higher in the human ENS, and the glia index (glia:neuron ratio) ranged from 1.4-1.9 and 6.0-6.3 in the human submucous plexus (SMP) and MP, respectively, whereas in guinea pig, the glia index was 0.7-0.9 in the SMP and 1.7-1.8 in the MP. The glia index was the most robust quantitative descriptor of tissues within one species. In summary, this is a comprehensive set of quantitative EGC measures in man and guinea pig which provides a basis for pathological assessments of glial proliferation and/or degeneration in the diseased gut.
(4) This part of the work is still in progress.
Lay summary:
The underlying cause of the human inflammatory bowel diseases (IBD), Crohn’s disease and ulcerative colitis, remains unknown. While there is a plethora of studies addressing the mucosal immune system in IBD, little is known about the contribution of the enteric nervous system (ENS). The ENS can be conceived as a little brain in the gut and it regulates almost all vital functions of the gastrointestinal tract. It has been repeatedly demonstrated that the ENS is affected by intestinal inflammation, and consequently, it was suggested that the ENS may also be involved in the development of IBD. However, little is known about precise mechanisms by which the ENS may impact upon intestinal inflammation.
In order to study interactions between the ENS and intestinal inflammation, it is most promising to focus on a specific cell type of the ENS, namely the glial cells (enteric glia). There are indications derived from genetically modified animal models that the glial cells in the ENS may play a central role to preserve the intestinal mucosal barrier: If mice are born without enteric glia, the normal architecture of the intestinal mucosa disintegrates, resulting in serious intestinal inflammation and death of the animals. These observations suggested that enteric glia may be essential to maintain mucosal defense mechanisms, and that a loss of the glial support may be involved in the processes that underlie the changes seen in the inflamed bowel.
In this project we will address the factors mediating glial-mucosal interactions and their regulation. We will use a unique tissue culture system in which we combine isolated enteric glial cells from human intestine with human epithelial cells, and in this system we will evaluate how the presence or absence of enteric glia changes the functional properties of the epithelial cells. We will then analyze which glial factors are needed to effect the observed changes. These studies will be complemented by microscopic analyses of inflamed and non-inflamed specimens from Crohn’s disease patients to assess if the patterns identified in tissue culture can also be seen in tissue samples from patients.
We expect that improving insight into glial-epithelial interactions and identifying factors involved in these interactions will significantly further our understanding of the mechanisms underlying the multiple stages of gastrointestinal inflammation. We hope that such an understanding will ultimately allow for novel therapeutic approaches to IBD.