The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

The Journal of Immunology, 2007, 179: 295-304.
Copyright © 2007 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Berndt, U.
Right arrow Articles by Sturm, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Berndt, U.
Right arrow Articles by Sturm, A.

Proteomic Analysis of the Inflamed Intestinal Mucosa Reveals Distinctive Immune Response Profiles in Crohn’s Disease and Ulcerative Colitis1

Uta Berndt2,*, Sebastian Bartsch2,{dagger}, Lars Philipsen{dagger}, Silvio Danese{ddagger}, Bertram Wiedenmann*, Axel U. Dignass*, Marcus Hämmerle{dagger} and Andreas Sturm3,*

* Division of Gastroenterology and Hepatology, Department of Medicine, Charité–Campus Virchow Clinic, Universitätsmedizin Berlin, Berlin, Germany; {dagger} MelTec, Magdeburg, Germany; and {ddagger} Division of Gastroenterology, Istituti di Ricovero e Cura a Carattere Scientifico Istituto Clinico Humanitas, Rozzano, Milan, Italy


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Although Crohn’s disease (CrD) and ulcerative colitis (UC) share several clinical features, the mechanisms of tissue injury differ. Because the global cellular function depends upon the protein network environment as a whole, we explored changes in the distribution and association of mucosal proteins to define key events involved in disease pathogenesis. Endoscopic biopsies were taken from CrD, UC, and control colonic mucosa, and Multi-Epitope-Ligand-Cartographie immunofluorescence microscopy with 32 different Abs was performed. Multi-Epitope-Ligand-Cartographie is a novel, highly multiplexed robotic imaging technology which allows integrating cell biology and biomathematical tools to visualize dozens of proteins simultaneously in a structurally intact cell or tissue. In CrD, the number of CD3+CD45RA+ naive T cells was markedly increased, but only activated memory, but not naive, T cells expressed decreased levels of Bax, active caspase-3 or -8. In UC, only CD4+ T cells coexpressing NF-{kappa}B were caspase-8 and poly(ADP-ribose)-polymerase positive. Furthermore, the number of CD4+CD25+ T cells was elevated only in UC, whereas in CrD and controls, the number of these cells was similar. By using hub analysis, we also identified that the colocalization pattern with NF-{kappa}B+ and poly(ADP-ribose)-polymerase+ as base motifs distinguished CrD from UC. High-content proteomic analysis of the intestinal mucosa demonstrated for the first time that different T cell populations within the intestinal mucosa express proteins translating distinct biological functions in each form of inflammatory bowel disease. Thus, topological proteomic analysis may help to unravel the pathogenesis of inflammatory bowel disease by defining distinct immunopathogenic profiles in CrD and UC.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Both Crohn’s disease (CrD)4 and ulcerative colitis (UC) represent an inflammatory reaction of the gastrointestinal tissue of still poorly understood etiology and are collectively named inflammatory bowel diseases (IBD). Both diseases present with distinct clinical and histopathological characteristics suggesting that in each condition the mechanism of tissue injury is distinct. Advances in disease pathogenesis have been numerous in the last decades when the combined power of cellular and molecular biology began to unveil the enigmas of IBD. It is now evident that no single agent or distinct mechanism alone explains all aspects of IBD and several factors are necessary to result in either CrD or UC (1). However, it is generally accepted that IBD results from a disturbed interplay of the intestinal microflora with the immune system of a genetically susceptible host. In this scenario, physiologic intestinal immune homeostasis is lost, with subsequent activation of a multiplicity of immunologic and nonimmunologic responses involving many different cell types.

Several susceptible genomic loci have been implicated in IBD; however, these candidates probably represent only a small subset of all genes involved in the pathogenesis of IBD (2, 3). An exact linear association between the genome, transcriptome, and proteome of a cell does not exist and translation and posttranslational modifications of proteins are not apparent from the DNA or mRNA sequence. Thus, a major challenge in modern biology is to understand the expression, function, and regulation of the entire set of proteins encoded by an organism—the aims of the emerging field of proteomics (4, 5, 6). This information will be invaluable to understanding how complex biological processes occur at a molecular level, how they differ in various cell types, and how they are altered in disease states.

To execute their biological functions, proteins form networks in the various cellular compartments. However, proteins and protein networks depend on their cellular environment and localization. Therefore, proteins are not stochastically distributed, but are timely and spatially organized, and each protein must be at the right time, at the right location, and at the right concentration in a cell to interact with other proteins. To exert its function, any given cell must sort and cocompartmentalize proteins to form specific protein patterns and networks (7). Thus, molecular networks, enabling specific cellular functions, obey a unique colocalization and anticolocalization code. Additionally, to change its function, a cell has to rearrange its code. To gain insight into these complex events, we need technologies to localize an extremely large number of molecular cell components in morphologically intact fixed cells in a given condition, on the single-cell level, in one single experiment which allow us to analyze the spatial arrangements of these components and therefore the cellular functions.

The so-termed Multi-Epitope-Ligand Cartographie (MELC) technology has perfected the use of fluorescent in situ protein detection by creating a highly flexible multiplex detection system (8). The unique robotic whole cell imaging technology visualizes dozens of proteins simultaneously in a structurally intact single cell as well as in the complexity of a specific tissue anatomy. The highly complex information generated by MELC is processed through advanced data analysis and visualization software, which integrates cell biology and biomathematical tools. This allows correlating cellular localization of proteins with functional identification of protein networks playing a crucial role in biological processes. The advantage that multidimensional microscopic robot technology for high-throughput protein recognition provides is to detect the considerable amount of 332 protein expression arrays and the opportunity to generate a protein colocalization map. Recently, it has been demonstrated that this approach can readily be used to identify an in situ protein expression pattern in inflammatory skin diseases (8, 9).

In the present study, we created for the first time a toponomic picture of the mucosal tissue of CrD and UC. The "toponome" is described as the entirety of all protein networks in a cell at a particular time. By using the novel and unique MELC technique, we hope to uncover changes in the expression, distribution, spatial location, and association of proteins critically involved in the pathogenesis in these diseases and thus predict key proteins critically involved in the pathogenesis of IBD, ultimately aiming to identify new diagnostic features and therapeutic targets in IBD.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Patient population

Endoscopic biopsies from patients with CrD (n = 10), UC (n = 10), and from the normal mucosa of control subjects (n = 10) were obtained. The median age of the CrD group was 33 years (range 19–48 years) with a median duration of disease of 7 years (range 1–22 years). The median age of the UC group was 39 years (range 22–50 years) with a median duration of disease of 10 years (range 4–29 years), and the median age of the healthy control group was 45 years (range 29–59 years). The gender distribution in each group was 50% male and 50% female.

In CrD and UC patients, the disease was active in all patients, as defined by the Crohn’s disease activity index of >150 and the colitis activity index of >9, respectively. At the time of colonoscopy, three CrD patients received azathioprine (2.5 mg/kg bodyweight (BW)) or 6-MP (1.5 mg/kg BW), four patients received 5-ASA, and four patients received corticosteroids. Because the patient groups were matched, three UC patients received azathioprine (2.5 mg/kg BW) or 6-MP (1.5 mg/kg BW), four patients received 5-ASA, and four patients received corticosteroids. From all patients, whether CrD or UC, samples were taken from macroscopically inflamed colonic lesions and inflammation was confirmed by histological evaluation. Control patients underwent colonoscopy for colon cancer prophylaxis and gastrointestinal symptoms such as diarrhea, abdominal pain, or changed stool habits were absent. Histological data of the control subjects showed no inflammation. Signed informed consent was obtained from each subject. Approval of the protocol and consent form was granted by the local ethics committee of the Charité (Berlin, Germany).

Sample preparation

Endoscopic biopsies were snap-frozen in liquid nitrogen and stored at –80°C. After embedding the biopsy in Tissue-Tek (Sakura Finetek), cryosections of 5 µm thickness were sliced using the cryotome Frigotom 2800 (–30°C; Leica) and applied on silane (Sigma-Aldrich) coated coverslips. After fixing the tissue with acetone for 10 s at room temperature, the coverslips were stored at –20°C. In preparation for the MELC procedure, the tissue was fixed once again with acetone for 10 min at –20°C. Afterward, the sample was rehydrated with Dulbecco’s PBS (PAA Laboratories) and nonspecific signals were blocked with normal goat serum (PAA Laboratories, diluted 1/15 with PBS) for 30 min at room temperature. After rinsing the sample five times with PBS, analysis of the tissue with the MELC technique was started. After finalizing the MELC runs, H&E staining was performed to allow colocalization with the respective tissue structures.

MELC library

We used a MELC library of 31 fluorescence tags comprising Abs, lectins, and propidium iodide as a nucleic acid dye (Table I). The appropriate working dilutions, fluorophore labels, incubation time (15 min), and positions within the MELC run had been established and validated in the course of systematic experiments based on conventional immunohistochemistry (IHC) and MELC calibration runs (8, 9).


View this table:
[in this window]
[in a new window]

  		Table I. MELC Ab library

 
Data acquisition by Toponome Imaging Cycler Multi-Epitope readout

Aspects of MELC technology (Europapatent Patent 0 810 428, United States Patent 6,150,173) have been described previously (8, 9). Briefly, a slide with a given specimen was placed on the stage of an inverted wide-field fluorescence microscope (Leica DM IRE2; x20 air lens NA 0.7). Fluorochrome-labeled Abs (Table I) and wash solutions were added and removed robotically under temperature control, correcting any displacement of the sample and lens. By a robotic process, first the specimens were incubated with predetermined fluorescence tags and rinsed with wash solution. Afterward, the phase contrast and fluorescence signals were imaged by a cooled charge-coupled device camera (Apogee KX4; Apogee Instruments, 2048 x 2048 pixels; 2x binning results in images of 1024 x 1024 pixels). To delete the specific signal of the given Ab before pipetting the consecutive one, a soft bleaching step was performed sequentially for each tag as incubation-imaging-bleaching cycles. Performance of the first three MELC cycles with PBS and fluorochrome-labeled mouse IgG intrinsic fluorescence and unspecific tag binding was controlled. Pipetting, recording all image data, and coordinating all system components were controlled by software developed by MelTec. These prerequisites allowed a fully automated cyclic process (Ab incubation/fluorescence detection/bleaching) to be repeatedly performed at any appropriate epitope. Taking advantage of a motor-controlled XY stage of the microscope, several visual fields were recorded simultaneously in a given MELC run. The computer platform of the Toponome Imaging Cycler (MelTec) stored the phase contrast and raw fluorescence images for all tag-binding sites and chosen visual fields.

Data analysis

By using the corresponding phase contrast images, fluorescence images produced by each tag were aligned pixel-wise. Images were corrected for illumination faults using flat-field correction. Finally, pixels not belonging to the biological specimen’s information, e.g., in cases of section artifacts, were excluded as invalid by a mask-setting process. Preprocessed image data were subjected to binarization. The thresholds generated by the system automatically were validated and adjusted manually for each protein. The expression of a protein was set to the value of zero for a signal below the threshold and to 1 for a signal above the threshold in projection to a pixel. Superimposed binarized images composed a matrix of combinatorial molecular phenotypes (CMPs) which represented a binary (yes/no) code of n epitope expression in relation to each pixel (900 x 900 nm2 area) of a visual field (1024 x 1024 pixels). Thus, this MELC approach detected a theoretical maximum of as high as 2n different CMPs. Further analysis dealt with CMP motifs characterizing corresponding pixels. These CMP motifs are defined as pixel-related code of one/zero/wildcard ciphering. We used TopoMiner software packages (MelTec) to search for CMP motifs, whose overall frequency differs significantly in two different sample groups using the Wilcoxon rank-sum test (p < 0.01) or Student’s t test. In detail, TopoMiner calculated the relative frequency of CMP motifs in relation to the number of all valid pixels of the observed visual field or to the frequency of predefined CMP motifs (base motifs). The search through the space of motifs was performed in a sequentially ordered strategy: all single epitopes, combinations of two, three, four, and so on (n = epitopes) were searched. Due to the large number of possible motifs, and due to restricted computational time, the search depth was limited to n = 5. TopoMiner analysis was used to initially compare CrD and UC. Significant motifs were further validated by a comparison of both diseases with healthy control specimens, respectively, using Wilcoxon rank-sum, giving a total level of significance without overlapping distributions (p < 0.01).

Data of interest visualization

Colocalizations of one marker of interest with all the other markers are displayed in a hub diagram. The relative frequencies of the colocalizations are plotted radially around the marker of interest in the middle, like spokes around a hub in a wheel. The data are plotted logarithmically. This diagram helps comparing different colocalizations related to the base marker. As the three groups (UC, CrD, and controls) are all plotted in the same diagram, the frequencies can easily be compared. If there is a significant difference in the frequencies it will be shown by a symbol on the outside of the diagram.

For visualization of CMP motifs of interest, the Topolyzer (MelTec) was used. This software package allows visualization of CMP motifs of interest as tables, boxplots (10, 11, 12), and in a given tissue sample, by superimposing these CMP motifs on the corresponding biological structures to create toponome maps (13). In addition, the Topolyzer allows visualization of the fluorescence and binarized images for a given CMP motif.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Evaluation of the reproducibility and robustness of the MELC technology within the intestinal mucosa

The vast majority of studies investigating the pathogenesis of CrD and UC have been performed by analyzing protein or RNA extracted from cells isolated from CrD or UC tissues, often highly purified and therefore removed from other cell types normally present in the in vivo microenvironment. This artificial experimental situation neglects that proteins are assembled within on the surface of cells as the result of complex communication networks generating a myriad of different cell functionalities. Thus, before exploring the molecular networks involved in the pathogenesis of IBD and eventually performing a toponomic analysis of the disease, we first tested the reproducibility and robustness of the MELC technology in our system.

It has already been demonstrated that in different MELC runs, single marker-labeled cells obtain comparable fluorescent counts, resulting in a stability index of 0.97 (9). Because the architecture of mucosa far exceeds the complexity of a single cell, we first needed to confirm the reproducibility of this method at the level of the intestinal tissue. Analyzing a single biopsy in several independent runs, the relative frequencies of single markers in the separate runs resulted in a stability index of 0.94, confirming the reproducibility and robustness of the MELC technology (Fig. 1A), and the reliability of its application. Another issue before exploring the toponome of IBD is how representative are the alterations of protein patterns within the mucosa of an individual patient. To answer this important question, several biopsies from a single patient were analyzed in several runs. As depicted in Fig. 1B, the relative frequencies of the fluorescent signals of the Ab panel are highly conserved (correlation coefficient = 0.97), demonstrating that the described combinatorial molecular patterns of the biopsies are reproducible and thus representative for an individual sample. At last, H&E staining which was performed to allow colocalization with the respective tissue structures after finalizing the MELC runs demonstrated an excellent maintenance of the tissue integrity over the whole MELC run (data not shown).


Figure 1
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 1. Evaluation of the MELC technology. A, Reproducibility and robustness of the MELC technology. Sections of the same biopsies were measured in various MELC runs as described in Materials and Methods. Afterward, the relative frequencies of the fluorescent signals from every applied marker were plotted against each other from a different run. The equal distribution of the dots near the line representing identical measurements and the calculation of a coefficient of determination R2 near 1.0 demonstrates the constancy of the output of several MELC runs. The given plot is representative of three individual series of experiments. B, Proof of intraindividually CMP stability. Biopsies from different inflamed regions from one patient were measured in a single MELC run as described in Materials and Methods. Afterward, the relative frequencies of the fluorescent signals from every applied marker were plotted against each other. The equal distribution of the dots near the line representing identical measurements, and the calculation of a coefficient of determination R2 near 1.0, demonstrates that the identified protein patterns are characteristic within the whole region of the mucosa in an individual. The given plot is representative of three individual series of experiments.

 
Multidimensional analysis of protein locations in inflamed CrD and UC tissue

Aiming at uncovering differences that distinguish CrD from UC and normal mucosa tissue on a toponomic level, we mapped 23 surface and 9 intracellular proteins in single tissue sample from 10 control, 10 CrD, and 10 UC patients. Using the multidimensional MELC process, we first generated conventional RGB panels combining three different Abs.

It is well-known that in the intestinal mucosa the number of memory T cells outmatches the number of naive T cells (14). Having confirmed this observation in our RGB panel (Fig. 2), we demonstrated that in CrD, the number of CD3+CD45RA+ naive T cells is markedly increased compared with control and UC tissue (Fig. 2A). Analyzing the CD45R0+ population in more detail, we then discovered that, compared with CrD and controls, the number of cells coexpressing CD25 as activation marker and poly(ADP-ribose)-polymerase (PARP) as apoptotic marker is increased in memory T cells in UC tissue (Fig. 2B).


Figure 2
View larger version (111K):
[in this window]
[in a new window]

  		FIGURE 2. Visualization of protein expression patterns in the RGB mode (upper and middle panel) and as a multidimensional protein localization map (lower panel). A, As demonstrated by IHC, the number of naive T cells is increased in CrD (CD) compared with UC and control tissue. Biopsies were taken from control, CrD, and UC mucosa and IHC was performed as described in Materials and Methods. Color assignments: red, CD3+; green, CD45RA+; blue, CD45R0+; magenta, CD3+CD45R0+; yellow, CD3+CD45RA+; cyan, CD45RA+CD45R0+. B, As demonstrated by IHC, the number of activated memory T cells coexpressing PARP is increased in UC, compared with CrD and control tissue. Biopsies were taken from control, CrD, and UC mucosa and IHC was performed as described in Materials and Methods. Color assignments: red, CD45R0+; green, CD25+; blue, PARP+; magenta, CD45R0+PARP+; yellow, CD25+CD45R0+; cyan, CD25+PARP+. C, The different protein expression patterns demonstrated in control, CrD, and UC tissue by the multidimensional protein localization map substantiate the distinct molecular events observed in CrD and UC. Biopsies were taken from control, CrD, and UC mucosa and IHC was performed as described in Materials and Methods. The respective color assignments of the multidimensional protein localization map are depicted in the upper end of the plot. Cy, Cytoceratin; RA, CD45RA. The given plots are representative of three individual series of experiments.

 
One of the major limitations regarding the RGB panel is the inability to display more than three different Abs. This limitation can now be overcome by the unique MELC technology, which allows combining multiple IHC pictures from one sample in any order (9). Thus, after having demonstrated that the number of naive T cells is increased in CrD mucosa and distinguishes CrD from UC, we set forth and performed a multidimensional analysis of protein locations using the unique advantages of the MELC technology (Fig. 2C). Although interesting and novel, these pictures make it obvious that the visual presentation of more than three proteins in one sample results in a picture which is difficult to analyze. In addition, the identification of proteins by conventional IHC cannot be quantified, a further limitation of this conventional approach, which can now be overcome by the MELC data mining process. Even more valuable and unique, the MELC technology also allows for the identification of molecular networks not only by protein colocalization motifs, but also by the absence of proteins, the so-called anticolocalization code (13).

Protein expression profiles of UC and CrD

Although CrD and UC share several clinical features, it is assumed that the mechanisms of tissue injury differ in each condition. Of the 332 possible CMP motifs we analyzed, when we reduced the number of distinct CMP motifs to a significance level of p < 0.01, we identified 1337 CMP motifs which are statistically significantly different in CrD tissue compared with controls, 2930 CMP motifs which are different between UC and control tissue, and 2599 CMP motifs which distinguish CrD from UC. When calculating the protein expression profiles being 3- and 10-fold different between the different conditions, the high numbers of differentially expressed CMP motifs highlight both shared unique features in the two forms of IBD (Table II).


View this table:
[in this window]
[in a new window]

  		Table II. Numeric distribution of different CMP expressions in control, CrD, and UC mucosaa

 
Distinct protein expression patterns in CrD, UC, and control tissue

Having demonstrated the validity of the technology to determine protein expression of cells critically involved in the inflammatory process in CrD and UC according to their protein colocalization and anticolocalization motifs, we next aimed to characterize distinct T cell subpopulations by combinatorial molecular phenotypes.

Distinct regulation of apoptosis in IBD tissue

Of the panel of 2599 significantly different CMP motifs which significantly distinguish CrD from UC, we focused first on the CD3-positive cell population. We previously described that in CrD the caspase-3 activity of mucosal T cells is decreased (15), helping to explain the impaired T cell apoptosis observed in CrD (16). In contrast, in CrD T cells are more activated and have an increased expression of the nuclear transcription factor NF-{kappa}B (17). These two findings were functionally linked by our analysis demonstrating that in CrD, only those cells which have decreased PARP activity coexpress NF-{kappa}B (Fig. 3, A and B), indicating that the decreased activation of PARP in CrD is associated with an increased activation status. In UC, the number of apoptotic T cells is increased probably due to an increased caspase-8 activity (18). When we further dissected the CD3-positive T cell population, we uncovered that in UC, only CD4-positive cells coexpressing NF-{kappa}B are active caspase-8 and PARP positive, and thus prone to die via caspase-8 signaling (Fig. 3C), whereas the number of CD4+caspase-8+PARP+ cells alone or the number of CD8+caspase-8+PARP+NF-{kappa}B+ was not different between the groups (data not shown). In this context, we also demonstrate that in CrD and UC, the number of cells which colocalize CD3 and NF-{kappa}B but lack p53 is significantly increased, highlighting the advantages of the MELC technology to uncover anticolocalization and underscoring the important role of p53 as negative regulator of intestinal immunity (19, 20) (Fig. 3D).


Figure 3
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 3. Boxplot analysis of different T cell populations demonstrating distinct regulation of apoptosis in IBD tissue. A, Boxplot analysis of CD3-positive, PARP-2-negative cells shows a similar distribution in control, CrD (CD), and UC mucosa. B, In CrD, only those T cells which coexpress NF-{kappa}B have decreased PARP-2 activity. C, Boxplot analysis of CD4-positive, active caspase-8-positive, PARP-2-positive, and NF-{kappa}B-positive cells. In UC only, CD4-positive cells coexpressing NF-{kappa}B are active caspase-8 and PARP positive. D, Boxplot analysis of CD3-positive, NF-{kappa}B-positive, p53-negative cells. In CrD and UC, the number of cells which colocalize CD3 and NF-{kappa}B but lack of p53 is significantly increased compared with controls. E, Boxplot analysis of apoptotic markers in naive T cells. The number of CD45RA-positive naive T cells, lacking Bax expression, active caspase-8, or active caspase-3, but expression of NF-{kappa}B is similar in each group. F, In contrast, in CrD, NF-{kappa}B-positive, CD45R0-positive memory T cells not expressing Bax, active caspase-8, or active caspase-3 are increased. Biopsies were taken from control, CrD, and UC mucosa and MELC-staining cycles and data processing were performed as described in Materials and Methods. The respective significance levels are depicted within the graphs.

 
Aiming at further defining the cell populations implicated in the disturbed T cell apoptosis observed in IBD, we next analyzed apoptotic signaling in naive and memory cells. We demonstrated above that the number of CD3+CD45RA+ naive T cells is markedly increased in CrD tissue compared with controls and UC, suggesting an increased influx of peripheral blood T cells in the mucosa during the course of inflammation. As depicted in Fig. 3, E and F, apoptosis seems to be regulated distinctively in naive and memory T cells because activated naive T cells do not express decreased levels of Bax, active caspase-8, or active caspase-3 (Fig. 3E), which is in contrast observed in their memory type counterparts in CrD (Fig. 3F).

As demonstrated by boxplot analysis, NF-{kappa}B+ expression and apoptosis are closely linked. Because both pathways are critically involved in the initiation and perpetuation of both CrD and UC (21, 22), we next performed hub analysis. Hub analysis can visualize the dependency of one marker of interest with all the other markers, uncovering significant and distinct protein correlations related to the base marker. When we first picked NF-{kappa}B+ as marker of interest, 16 different protein colocalizations with this base marker discriminated UC from controls, 10 motifs discriminated CrD from controls, and 2 combinations discriminated CrD from UC tissue (Fig. 4A). When we then choose Bax as marker of interest, 13 different protein colocalizations were able to distinguish CrD from UC and two combinations distinguish control from UC tissue (Fig. 4B). Finally, we used PARP-2 as a base marker and plotted the frequencies of all other Abs tested in the same diagram (Fig. 4C). Interestingly, with PARP-2 as a base motif, all further combinations distinguished CrD from UC tissue, bringing to light a potentially distinct and important role of this protein in the pathogenesis of CrD (Fig. 4C).


Figure 4
View larger version (33K):
[in this window]
[in a new window]

  		FIGURE 4. Hub analysis of key proteins involved in apoptotic pathways. Colocalizations of a single marker of interest with all the additional markers existing in the Ab library are plotted in the hub diagram as relative frequencies of colocalization. A, Using NF-{kappa}B as marker of interest, 16 different protein colocalizations with this base marker discriminated UC from controls, 10 motifs discriminated CrD (CD) from controls, and two combinations discriminated CrD from UC significantly. B, Using Bax as a marker of interest, 13 different protein colocalizations were able to distinguish CrD from UC and two combinations control from UC tissue. C, Using PARP-2 as base motif, all further combinations distinguished CrD from UC tissue. Biopsies were taken from control, CrD, and UC mucosa and MELC-staining cycles and data processing were performed as described in Materials and Methods. The respective significance level is p < 0.05. Cy, Cytoceratin; Cas3, Caspase-3; Cas8, Caspase-8; Coll IV, Collagen IV.

 
Activation of B cells and monocytes in UC

CD4+ T cells that coexpress the CD25 Ag act as T regulatory cells controlling the immune responses to self- and foreign Ags. It has been demonstrated by Duchmann and colleagues (23) that in IBD tissue, the number of CD4+CD25+ T cells is increased, however, CrD and UC tissue were not separately analyzed. We now confirmed this finding (Fig. 5A) but demonstrating that the number of CD4+CD25+ T cells is elevated only in UC, whereas in CrD and controls the number of this cell type is similar. CD25 represents the IL-2R {alpha}-chain which is expressed on activated T and B cells as well as monocytes. When we then analyzed the mononuclear non-T cell population (CD3–), the number of CD25+ cells was still increased in UC, but not CrD tissue, showing for the first time that in UC not only the number of activated CD4+CD25+ cells is increased but also the number of CD3–CD45RA+CD25+CD2+ cells (Fig. 5B).


Figure 5
View larger version (14K):
[in this window]
[in a new window]

  		FIGURE 5. Boxplot analysis of activated mononuclear lamina propria cells. A, Regulatory T cells are increased in UC compared with CrD (CD). B, Activated non-T cells are also increased in UC compared with CrD. Biopsies were taken from control, CrD, and UC mucosa and MELC-staining cycles and data processing were performed as described in Materials and Methods. The respective significance levels are depicted within the graphs.

 
In this context, our finding that in CrD and UC the number of CD8+CD11a+NF-{kappa}B+ cells is significantly increased compared with controls was interesting, confirming the previously described enhanced cytolytic activity of intestinal intraepithelial lymphocytes in patients with CrD (Fig. 6) (24).


Figure 6
View larger version (15K):
[in this window]
[in a new window]

  		FIGURE 6. Boxplot analysis of CD8-positive, CD11a-positive, and NF-{kappa}B-positive cells. The number of T cells with cytolytic activity expressing CD11a is increased not only in CrD (CD) but also in UC compared with controls. Biopsies were taken from control, CrD, and UC mucosa and MELC-staining cycles and data processing were performed as described in Materials and Methods. The respective significance levels are depicted within the graph.

 
Enhanced accumulation of CD4+CD7– T cells in IBD

Modulation of CD7 is associated with an inhibition of T cell proliferation (25) and CD7-negative T cells represent a constant proportion of CD4+CD45RA–CD45R0+ memory T cells that accumulate in inflammatory skin lesions (26, 27), cutaneous T cell lymphoma, and in the peripheral blood of patients with rheumatoid arthritis (28). Furthermore, CD4+ cells that lack the CD7 Ag express the homing receptor cutaneous lymphocyte Ag (29), preferentially attach to vascular endothelial cells, and accumulate in the skin during chronic inflammation (30). We now demonstrate that CD4+CD7–CD45RA–NF-{kappa}B+ T cells are also significantly increased in CrD and UC tissue compared with controls, demonstrating for the first time that this cell type also accumulates in the intestinal mucosa during inflammation (Fig. 7A). Interestingly, when we further investigated the integrins potentially involved in this recruitment, only CD4+CD7–CD45RA–NF-{kappa}B+ cells that coexpress integrin beta2 (CD18) are elevated in CrD and UC (Fig. 7B), whereas CD4+CD7–CD45RA–NF-{kappa}B+ cells coexpressing integrin {alpha}L (CD11a) as part of the LFA-1 are only elevated in UC tissue (Fig. 7C). In contrast, the number of CD4+CD7–CD45RA–NF-{kappa}B+ cells coexpressing integrin beta1 (CD29) is similar in control, CrD, and UC tissue (Fig. 7D).


Figure 7
View larger version (28K):
[in this window]
[in a new window]

  		FIGURE 7. Boxplot analysis of CD4-positive, CD7-negative, CD45RA-negative, and NF-{kappa}B-positive cells focused on expression of integrins. A, This activated CD4-positive, CD7-negative subset of T cells is increased in CrD (CD) as well as in UC compared with controls. B, CD4+CD7–CD45RA–NF-{kappa}B+ cells with coexpression of integrin beta2 (CD18) are elevated in UC and CrD compared with control. C, Furthermore, the number of CD4+CD7–CD45RA–NF-{kappa}B+ cells with expression of integrin {alpha}L (CD11a) as part of the LFA-1 is increased in UC. D, Concerning expression of integrin beta1 (CD29) the numbers of CD4+CD7–CD45RA–NF-{kappa}B+ cells are similar in each group. Biopsies were taken from control, CrD, and UC mucosa and MELC-staining cycles and data processing were performed as described in Materials and Methods. The respective significance levels are depicted within the graphs.

 
Distinct activation of Th cells in the context of their HLA status

The contribution of the intestinal microflora in the pathogenesis of IBD is widely accepted (1, 31, 32, 33) and microbial Ags may lead to activation of T cells and expression of activation markers such as HLA class II. We therefore investigated in situ the expression profile of the MHC molecules HLA-DR and DQ in activated Th cells. Whereas the number of CD4-positive cells expressing HLA-DR was comparable in CrD, UC, and controls (data not shown), the number of CD4+CD25+HLA-DR+ cells was significantly increased in CrD compared with controls, while UC showed a similar but not statistically significant trend (Fig. 8A). In contrast, the number of cells coexpressing CD4+CD25+HLA-DQ+ was comparable in all groups (Fig. 8B).


Figure 8
View larger version (13K):
[in this window]
[in a new window]

  		FIGURE 8. Boxplot analysis of CD4-positive, CD25-positive cells in context of their HLA status. A, In CrD (CD), the number of activated Th cells expressing the MHC molecule HLA-DR is significantly increased compared with controls. B, In contrast, the number of activated Th cells expressing the MHC molecule HLA-DQ is comparably low in CrD, UC, and controls. Biopsies were taken from control, CrD, and UC mucosa and MELC-staining cycles and data processing were performed as described in Materials and Methods. The respective significance level is depicted within the graph.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
The need to better understand the pathophysiology of IBD and the distinct immune-mediated mechanisms of mucosal injury in CrD and UC that are responsible for their different clinical features is obvious. Cell function is determined at many levels, and even having the complete genome sequence associated with a specific disease entity is not enough to understand the biological dysfunction responsible for tissue injury. Thus, proteomics, the large scale analysis of proteins, emerges as one of the most important tools to characterize gene function, functionally link different gene products, and provide a global insight into the mechanism of biological processes. Whereas until now, in most cases, the sensitivity of this functional recognition in vivo exceeds the possibilities of modern proteomic techniques and cells or tissues have to be lysed to perform protein analysis, MELC technology allows for the identification of three-dimensional protein associations in selected subcellular compartments and disclosure of where and when proteins are colocalized in cellular or tissue compartments (9). Recognizing that every cell generates different protein patterns encoding distinct functions, we used the unique MELC technique to map functional clusters of proteins and provide for the first time context-dependent protein information of the normal and inflamed intestinal mucosa.

The robustness of CMP detection by MELC has been demonstrated recently (8, 9), but we initially tested the reproducibility and representativeness of this novel technique in the human intestinal mucosa. By demonstrating that repetitive measurements of the same sample do not yield significant differences, the hypothesis of equality was not rejected, validating the reproducibility of the MELC technology in the intestinal mucosa. Additionally, by comparing the protein expression profiles from different inflamed sites from one patient, we showed that within inflamed segments of the large intestine, protein expression profiles are similar, an important and novel finding.

Based on a search depth of five Ab combinations, we tested 7,061,120 motifs of 4.3 x 109 possible CMP motifs and identified 6,866 CMP motifs which distinguish control from CrD, control from UC, or CrD from UC tissue. The huge amount of CMP motifs which are comparable between these different groups may be indicative that in the course of mucosal inflammation only relatively "few" protein-protein patterns drive the inflammatory process, that several pathogenic events are shared secondary to chronic inflammation, or that numerous proteins are not directly involved in IBD, underlying the need to assess the mature form of gene expression, the proteome. At the same time, 2,599 differentially expressed unique CMP motifs in each IBD subtype spell distinctive disease signatures for UC and CrD that underscore fundamental differences in their pathogenesis.

The aim of our study was to analyze proteome topology in the normal and inflamed mucosa and therefore contribute to a better understanding of the pathophysiology of IBD. Second, we want to test a novel and unique automated multidimensional fluorescence microscopy capable of mapping hundreds of different protein combinations at different cellular levels. Dysregulated activation and apoptosis of lamina propria mononuclear cells are key events which initiate and perpetuate CrD and UC (34, 35). Consequently, drugs which inhibit cell activation, such as steroids, azathioprine/6-mercaptopurin or calcineurin inhibitors, or drugs which induce apoptosis, e.g., Abs to TNF-{alpha}, are highly effective in the treatment of both entities (36, 37, 38). For this reason, we included in our analysis proteins known to be critically involved in cell activation and apoptosis, but also functionally related protein complexes which are only weakly or transiently expressed, a unique and so far unexplored feature of our study.

During inflammation, the composition of the cells resident in the intestinal mucosa changes dramatically. Accordingly, the central question arises of whether molecular events during the course of IBD occur in all cell types comparably or whether naive or recently activated cells behave differently than memory or resting cells. We demonstrated that in CrD, the number of naive T cells is markedly increased in the inflamed mucosa and, importantly, that the naive T cell population, contrary to its memory counterparts, is resistant to apoptotic signals, confirming our previous findings that naive and memory T cells regulate their caspase activity distinctively (18). We next confirmed the distinct regulation of apoptosis in several T cell subpopulations and reinforced the need to analyze protein expression in their spatial arrangement. In UC for example, caspase-8 activity is only increased in Th cells coexpressing PARP and NF-{kappa}B. In contrast, in CrD, only activated memory T cells coexpressing NF-{kappa}B, but not naive T cells coexpressing NF-{kappa}B, have decreased levels of Bax, active caspase-3, or -8. These findings show that NF-{kappa}B has different activities dependent on its protein colocalization and link its expression with caspase regulation.

One of the best characterized and most powerful types of regulatory cells express both CD4 and CD25. They are involved in oral tolerance, inhibit the response of activated T cells against bacterial Ags and their adoptive transfer ameliorates experimental colitis (39, 40, 41, 42). Recently, it was demonstrated that inflamed IBD mucosa contains an increased number of CD4+CD25+FOXP3+ T cells, however, the increase of T regulatory cells was significantly lower compared with inflammatory controls (23). The assumption that the inability to sufficiently increase the number of regulatory T cells in IBD parallels the inability to limit gut inflammation is still valid but speculative. By analyzing CrD and UC separately in their own mucosal microenvironment and preserving the cellular integrity, we found that the number of CD4+CD25+ T regulatory cells is elevated only in UC, but not CrD. This finding is in accordance with the recent demonstration that CD4+CD25+ T cells are enriched in the colonic mucosa of patients with active UC (43) whereas the inability to up-regulate suppressive, regulatory T cells demonstrated here might explain the unrestricted proliferation and expansion of T cells in CrD (44).

The adhesion of circulating lymphocytes to vascular endothelial cells is required for the migration of cells into the mucosa. We found that in CrD the number of naive T cells, probably migrating from the peripheral blood into the mucosa, is increased, confirming the importance of this event in the course of intestinal inflammation. Vascular endothelial cells within the gastrointestinal tract express E-selectin (CD62E) which is up-regulated during the active phase of IBD (45). E-selectin binds to the cutaneous lymphocyte Ag which is preferentially expressed on circulating CD4+ T cells that lack CD7 (29). Interestingly, the majority of CD7– T cells are of CD4+ helper and CD45RA–CD45RO+ phenotype (46). Within the skin, highly increased numbers of CD7– T cells are present among lymphocytes cultured from biopsies of inflammatory skin lesions (26) as well as the cutaneous T cell lymphoma, underlining the important role of CD7 during inflammatory processes and malignancies (47). We now expand the role of CD4+CD7– memory T cells by demonstrating that this regulatory T cell type is increased also in CrD and UC tissue, involving these new cell subsets into the complex pathophysiology of IBD. TNF-{alpha} stimulation increases CD4+CD7– T cell adherence to endothelial cells (48) and the enhanced TNF-{alpha} levels in IBD (49, 50) might facilitate this process. The interaction between LFA-1 and ICAM-1 is also crucial to mediate T cell adhesion to endothelial cells (51). In this scenario, we demonstrated that CD18 and CD11a as part of the LFA-1 receptor, but not integrin beta1 (CD29), seem to mediate this recruitment.

Both CrD and UC are associated with specific HLA-DR and -DQ phenotypes, although much of these data are controversial (52, 53). Having the ability to analyze distinct T cell subpopulations, we showed that the number of HLA-DR-positive cells was only increased in CrD when CD4 and CD25 were colocalized, underlying the important role of HLA-DR in the pathogenesis of CrD (54). Interestingly, the number of HLA-DQ was comparable in all groups regardless of the T cell type or status of activation examined.

By performing hub analysis, the second level of MELC data processing, which discloses dual coexpression of given epitopes, we finally tried to identify proteins that are translating molecular elements leading to the formation of molecular networks in IBD. We found that a colocalization pattern with NF-{kappa}B is distinct for several proteins in CrD and UC and hence underlines the importance of this transcription factor in the pathophysiology of IBD (55, 56). Hub analysis also demonstrated that protein colocalization with PARP distinguished CrD from UC tissue, bringing to light a potentially distinct and important role of this protein in the pathogenesis of CrD. PARP is activated by nicks and breaks in the DNA strand which can be induced by a variety of environmental stimuli and free radicals (57). It plays an important role in mediating the normal cellular response to DNA damage and is a target of the caspase protease activity associated with apoptosis (58). With regard to IBD, several studies revealed that the pharmacological inhibition of PARP ameliorates experimental colitis (59), indicating the clinical significance of our finding. In this context, p53 expression is also induced by DNA-strand breaks (60) and the demonstration that, in CrD and UC, the number of CD3- and NF-{kappa}B-positive cells which lack p53 is significantly increased not only confirms the link between PARP and NF-{kappa}B in the colon (61), but also brings to light the importance of this combinatorial protein pattern in the course of IBD.

In conclusion, using a novel, automated, multidimensional, fluorescence-based microscopy robot technology, we performed the first proteomic analysis of the human normal and inflamed intestinal mucosa. In particular, we demonstrated that the different T cell populations within the intestinal mucosa have a distinct susceptibility toward apoptosis, depending on their phenotype and status of activation. Furthermore, we identified an increased number of CD4+CD7– memory T cells in the mucosa of IBD patients, pointing to a yet unexplored role of this immune-regulatory cell type in the pathophysiology of IBD. Finally, we identified NF-{kappa}B and PARP as lead proteins that are representative molecular elements controlling the formation of molecular networks and the associated cellular functions in IBD. As a result, analysis of key immune function-related proteins and in situ detection of their modification may not only facilitate our understanding of the markedly different pathogenic events observed in CrD and UC, but also help assemble a toponomic picture of the intestinal mucosa which might aid the discovery of new targets or development of drugs specifically targeting those T cell populations whose function is disturbed in IBD.


   Acknowledgments
 
We thank Claudio Fiocchi (Lerner Research Institute, The Cleveland Clinic Foundation, Cleveland, OH) for critical reading of the manuscript and Daniela Paclik and Claudia Guzy for technical assistance.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
Sebastian Bartsch, Lars Philipsen, and Marcus Hämmerle are employed at MelTec.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was financially supported by the Charité Medical School (Berlin, Germany). Back

2 U.B. and S.B. contributed equally to this work. Back

3 Address correspondence and reprint requests to Dr. Andreas Sturm, Campus Virchow Clinic, Department of Hepatology and Gastroenterology, Charité–Universitätsmedizin Berlin, Augustenburger Platz 1, D-13353 Berlin, Germany. E-mail address: andreas.sturm{at}charite.de Back

4 Abbreviations used in this paper: CrD, Crohn’s disease; UC, ulcerative colitis; IBD, inflammatory bowel disease; MELC, Multi-Epitope-Ligand-Carthographie; CMP, combinatorial molecular phenotype; PARP, poly(ADP-ribose)-polymerase; IHC, immunohistochemistry; BW, bodyweight. Back

Received for publication January 10, 2007. Accepted for publication April 25, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Fiocchi, C.. 1998. Inflammatory bowel disease: etiology and pathogenesis. Gastroenterology 115: 182-205. [Medline]
  2. Gaya, D. R., R. K. Russell, E. R. Nimmo, J. Satsangi. 2006. New genes in inflammatory bowel disease: lessons for complex diseases?. Lancet 367: 1271-1284. [Medline]
  3. Lawrance, I. C., C. Fiocchi, S. Chakravarti. 2001. Ulcerative colitis and Crohn’s disease: distinctive gene expression profiles and novel susceptibility candidate genes. Hum. Mol. Genet. 10: 445-456. [Abstract/Free Full Text]
  4. Zhu, H., M. Bilgin, M. Snyder. 2003. Proteomics. Annu. Rev. Biochem. 72: 783-784. [Medline]
  5. Kok, K., P. Stokkers, P. H. Reitsma. 2004. Genomics and proteomics: implications for inflammatory bowel diseases. Inflamm. Bowel Dis. 10: (Suppl. 1):S1-S6. [Medline]
  6. Felley-Bosco, E., M. Andre. 2004. Proteomics and chronic inflammatory bowel diseases. Pathol. Res. Pract. 200: 129-133. [Medline]
  7. Stelzl, U., U. Worm, M. Lalowski, C. Haenig, F. H. Brembeck, H. Goehler, M. Stroedicke, M. Zenkner, A. Schoenherr, S. Koeppen, et al 2005. A human protein-protein interaction network: a resource for annotating the proteome. Cell 122: 957-968. [Medline]
  8. Schubert, W.. 2003. Topological proteomics, toponomics, MELK-technology. Adv. Biochem. Eng. Biotechnol. 83: 189-209. [Medline]
  9. Schubert, W., B. Bonnekoh, A. J. Pommer, L. Philipsen, R. Bockelmann, Y. Malykh, H. Gollnick, M. Friedenberger, M. Bode, A. W. Dress. 2006. Analyzing proteome topology and function by automated multidimensional fluorescence microscopy. Nat. Biotechnol. 24: 1270-1278. [Medline]
  10. McGill, R.. 1978. Variations of boxplots. Am. Stat. 32: 12-16. [Medline]
  11. Velleman, P. F., D. C. Hoaglin. 1984. Applications, Basics, and Computing of Exploratory Data Analysis 5th Ed.1-354. PWS-Kent Publishing, Boston.
  12. Nelson, L. S.. 2006. Evaluating overlapping confidence intervals. J. Qual. Technol. 21: 140-141.
  13. Schubert, W.. 2006. Cytomics in characterizing toponomes: towards the biological code of the cell. Cytometry 69: 209-211. [Medline]
  14. Schieferdecker, H. L., R. Ullrich, H. Hirseland, M. Zeitz. 1992. T cell differentiation antigens on lymphocytes in the human intestinal lamina propria. J. Immunol. 149: 2816-2822. [Abstract]
  15. Sturm, A., A. Z. Leite, S. Danese, K. A. Krivacic, G. A. West, S. Mohr, J. W. Jacobberger, C. Fiocchi. 2004. Divergent cell cycle kinetics underlie the distinct functional capacity of mucosal T cells in Crohn’s disease and ulcerative colitis. Gut 53: 1624-1631. [Abstract/Free Full Text]
  16. Boirivant, M., M. Marini, F. G. Di, A. M. Pronio, C. Montesani, R. Tersigni, W. Strober. 1999. Lamina propria T cells in Crohn’s disease and other gastrointestinal inflammation show defective CD2 pathway-induced apoptosis. Gastroenterology 116: 557-565. [Medline]
  17. Ellis, R. D., J. R. Goodlad, G. A. Limb, J. J. Powell, R. P. Thompson, N. A. Punchard. 1998. Activation of nuclear factor {kappa}B in Crohn’s disease. Inflamm. Res. 47: 440-445. [Medline]
  18. Sturm, A., S. Mohr, C. Fiocchi. 2002. Critical role of caspases in the regulation of apoptosis and proliferation of mucosal T cells. Gastroenterology 122: 1334-1345.
  19. Sturm, A., J. Itoh, J. W. Jacobberger, C. Fiocchi. 2002. p53 negatively regulates intestinal immunity by delaying mucosal T cell cycling. J. Clin. Invest. 109: 1481-1492. [Medline]
  20. Amundson, S. A., T. G. Myers, A. J. Fornace, Jr. 1998. Roles for p53 in growth arrest and apoptosis: putting on the brakes after genotoxic stress. Oncogene 17: 3287-3299. [Medline]
  21. Podolsky, D. K.. 2002. Inflammatory bowel disease. N. Engl. J. Med. 347: 417-429. [Free Full Text]
  22. Schreiber, S., S. Nikolaus, J. Hampe. 1998. Activation of nuclear factor {kappa}B inflammatory bowel disease. Gut 42: 477-484. [Abstract/Free Full Text]
  23. Maul, J., C. Loddenkemper, P. Mundt, E. Berg, T. Giese, A. Stallmach, M. Zeitz, R. Duchmann. 2005. Peripheral and intestinal regulatory CD4+ CD25high T cells in inflammatory bowel disease. Gastroenterology 128: 1868-1878.
  24. Nussler, N. C., B. Stange, R. A. Hoffman, W. H. Schraut, A. J. Bauer, P. Neuhaus. 2000. Enhanced cytolytic activity of intestinal intraepithelial lymphocytes in patients with Crohn’s disease. Langenbecks Arch. Surg. 385: 218-224. [Medline]
  25. Lazarovits, A. I., J. Karsh. 1989. Modulation of CD7 is associated with inhibition of T cell proliferation in response to tetanus toxoid and phytohemagglutinin. Transplant. Proc. 21: 3325-3327. [Medline]
  26. Moll, M., U. Reinhold, S. Kukel, H. Abken, R. Muller, I. Oltermann, H. W. Kreysel. 1994. CD7-negative helper T cells accumulate in inflammatory skin lesions. J. Invest. Dermatol. 102: 328-332. [Medline]
  27. Reinhold, U., H. Abken, S. Kukel, M. Moll, R. Muller, I. Oltermann, H. W. Kreysel. 1993. CD7– T cells represent a subset of normal human blood lymphocytes. J. Immunol. 150: 2081-2089. [Abstract]
  28. Schmidt, D., J. J. Goronzy, C. M. Weyand. 1996. CD4+ CD7–. J. Clin. Invest. 97: 2027-2037. [Medline]
  29. Sokolowska-Wojdylo, M., J. Wenzel, E. Gaffal, J. Lenz, P. Speuser, S. Erdmann, F. Abuzahra, E. Bowman, J. Roszkiewicz, T. Bieber, T. Tuting. 2005. Circulating clonal CLA+ and CD4+ T cells in Sezary syndrome express the skin-homing chemokine receptors CCR4 and CCR10 as well as the lymph node-homing chemokine receptor CCR7. Br. J. Dermatol. 152: 258-264. [Medline]
  30. Liu, L., H. Abken, C. Pfohler, G. Rappl, W. Tilgen, U. Reinhold. 2000. Accumulation of CD4+CD7– T cells in inflammatory skin lesions: evidence for preferential adhesion to vascular endothelial cells. Clin. Exp. Immunol. 121: 94-99. [Medline]
  31. Danese, S., M. Sans, C. Fiocchi. 2004. Inflammatory bowel disease: the role of environmental factors. Autoimmun. Rev. 3: 394-400. [Medline]
  32. Sokol, H., P. Seksik, L. Rigottier-Gois, C. Lay, P. Lepage, I. Podglajen, P. Marteau, J. Dore. 2006. Specificities of the fecal microbiota in inflammatory bowel disease. Inflamm. Bowel Dis. 12: 106-111. [Medline]
  33. Macpherson, A., U. Y. Khoo, I. Forgacs, J. Philpott-Howard, I. Bjarnason. 1996. Mucosal antibodies in inflammatory bowel disease are directed against intestinal bacteria. Gut 38: 365-375. [Abstract/Free Full Text]
  34. Dignass, A. U., D. C. Baumgart, A. Sturm. 2004. Review article: the aetiopathogenesis of inflammatory bowel disease–immunology and repair mechanisms. Aliment. Pharmacol. Ther. 20: (Suppl. 4):9-17.
  35. Peppelenbosch, M. P., S. J. van Deventer. 2004. T cell apoptosis and inflammatory bowel disease. Gut 53: 1556-1558. [Free Full Text]
  36. Targan, S. R., S. B. Hanauer, S. J. van Deventer, L. Mayer, D. H. Present, T. Braakman, K. L. DeWoody, T. F. Schaible, P. J. Rutgeerts. 1997. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor {alpha} for Crohn’s disease. Crohn’s Disease cA2 Study Group. N. Engl. J. Med. 337: 1029-1035. [Abstract/Free Full Text]
  37. Lugering, A., M. Schmidt, N. Lugering, H. G. Pauels, W. Domschke, T. Kucharzik. 2001. Infliximab induces apoptosis in monocytes from patients with chronic active Crohn’s disease by using a caspase-dependent pathway. Gastroenterology 121: 1145-1157. [Medline]
  38. Tiede, I., G. Fritz, S. Strand, D. Poppe, R. Dvorsky, D. Strand, H. A. Lehr, S. Wirtz, C. Becker, R. Atreya, et al 2003. CD28-dependent Rac1 activation is the molecular target of azathioprine in primary human CD4+ T lymphocytes. J. Clin. Invest. 111: 1133-1145. [Medline]
  39. Zhang, X., L. Izikson, L. Liu, H. L. Weiner. 2001. Activation of CD25+CD4+ regulatory T cells by oral antigen administration. J. Immunol. 167: 4245-4253. [Abstract/Free Full Text]
  40. Gad, M., J. Brimnes, M. H. Claesson. 2003. CD4+ T regulatory cells from the colonic lamina propria of normal mice inhibit proliferation of enterobacteria-reactive, disease-inducing Th1-cells from scid mice with colitis. Clin. Exp. Immunol. 131: 34-40. [Medline]
  41. Read, S., V. Malmstrom, F. Powrie. 2000. Cytotoxic T lymphocyte-associated antigen 4 plays an essential role in the function of CD25+CD4+ regulatory cells that control intestinal inflammation. J. Exp. Med. 192: 295-302. [Abstract/Free Full Text]
  42. Mottet, C., H. H. Uhlig, F. Powrie. 2003. Cutting edge: cure of colitis by CD4+CD25+ regulatory T cells. J. Immunol. 170: 3939-3943. [Abstract/Free Full Text]
  43. Holmen, N., A. Lundgren, S. Lundin, A. M. Bergin, A. Rudin, H. Sjovall, L. Ohman. 2006. Functional CD4+CD25high regulatory T cells are enriched in the colonic mucosa of patients with active ulcerative colitis and increase with disease activity. Inflamm. Bowel Dis. 12: 447-456. [Medline]
  44. Sturm, A., A. Z. Leite, S. Danese, K. A. Krivacic, G. A. West, S. Mohr, J. W. Jacobberger, C. Fiocchi. 2004. Divergent cell cycle kinetics underlie the distinct functional capacity of mucosal T cells in Crohn’s disease and ulcerative colitis. Gut 53: 1624-1631. [Abstract/Free Full Text]
  45. Lazaris, A. C., C. Dicoglou, S. Tseleni-Balafouta, H. Paraskevakou, P. S. Davaris. 1999. In situ expression of E-selectin and intercellular adhesion molecule-1 in chronic inflammatory diseases of the gastrointestinal tract. APMIS 107: 819-827. [Medline]
  46. Reinhold, U., H. Abken. 1997. CD4+ CD7– T cells: a separate subpopulation of memory T cells?. J. Clin. Immunol. 17: 265-271. [Medline]
  47. Reinhold, U., H. Abken. 1995. Cutaneous T-cell lymphoma: molecular genetics, immunology and pathogenesis. Eur. J. Cancer 31A: 793-799.
  48. Groves, R. W., M. H. Allen, J. N. Barker, D. O. Haskard, D. M. MacDonald. 1991. Endothelial leucocyte adhesion molecule-1 (ELAM-1) expression in cutaneous inflammation. Br. J. Dermatol. 124: 117-123. [Medline]
  49. Reinecker, H. C., M. Steffen, T. Witthoeft, I. Pflueger, S. Schreiber, R. P. MacDermott, A. Raedler. 1993. Enhanced secretion of tumour necrosis factor-{alpha}, IL-6, and IL-1beta by isolated lamina propria mononuclear cells from patients with ulcerative colitis and Crohn’s disease. Clin. Exp. Immunol. 94: 174-181. [Medline]
  50. Breese, E. J., C. A. Michie, S. W. Nicholls, S. H. Murch, C. B. Williams, P. Domizio, J. A. Walker-Smith, T. T. MacDonald. 1994. Tumor necrosis factor {alpha}-producing cells in the intestinal mucosa of children with inflammatory bowel disease. Gastroenterology 106: 1455-1466. [Medline]
  51. Dustin, M. L., T. A. Springer. 1988. Lymphocyte function-associated Ag-1 (LFA-1) interaction with intercellular adhesion molecule-1 (ICAM-1) is one of at least three mechanisms for lymphocyte adhesion to cultured endothelial cells. J. Cell Biol. 107: 321-331. [Abstract/Free Full Text]
  52. Toyoda, H., S. J. Wang, H. Y. Yang, A. Redford, D. Magalong, D. Tyan, C. K. McElree, S. R. Pressman, F. Shanahan, S. R. Targan, et al 1993. Distinct associations of HLA class II genes with inflammatory bowel disease. Gastroenterology 104: 741-748. [Medline]
  53. Stokkers, P. C., P. H. Reitsma, G. N. Tytgat, S. J. van Deventer. 1999. HLA-DR and -DQ phenotypes in inflammatory bowel disease: a meta-analysis. Gut 45: 395-401. [Abstract/Free Full Text]
  54. Ahmad, T., S. E. Marshall, D. Jewell. 2006. Genetics of inflammatory bowel disease: the role of the HLA complex. World J. Gastroenterol. 12: 3628-3635. [Medline]
  55. Kucharczak, J., M. J. Simmons, Y. Fan, C. Gelinas. 2003. To be, or not to be: NF-{kappa}B is the answer–role of Rel/NF-{kappa}B in the regulation of apoptosis. Oncogene 22: 8961-8982. [Medline]
  56. Kumar, A., Y. Takada, A. M. Boriek, B. B. Aggarwal. 2004. Nuclear factor-{kappa}B: its role in health and disease. J. Mol. Med. 82: 434-448. [Medline]
  57. Vidakovic, M., G. Poznanovic, J. Bode. 2005. DNA break repair: refined rules of an already complicated game. Biochem. Cell Biol. 83: 365-373. [Medline]
  58. Patel, T., G. J. Gores, S. H. Kaufmann. 1996. The role of proteases during apoptosis. FASEB J. 10: 587-597. [Abstract]
  59. Cuzzocrea, S.. 2005. Shock, inflammation and PARP. Pharmacol. Res. 52: 72-82. [Medline]
  60. Green, D. R., M. Schuler. 2000. T cell development: some cells get all the breaks. Nat. Immunol. 1: 15-17. [Medline]
  61. Zingarelli, B., M. O’Connor, P. W. Hake. 2003. Inhibitors of poly (ADP-ribose) polymerase modulate signal transduction pathways in colitis. Eur. J. Pharmacol. 469: 183-194. [Medline]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
U. Berndt, L. Philipsen, S. Bartsch, B. Wiedenmann, D. C. Baumgart, M. Hammerle, and A. Sturm
Systematic High-Content Proteomic Analysis Reveals Substantial Immunologic Changes in Colorectal Cancer
Cancer Res., February 1, 2008; 68(3): 880 - 888.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available