SEARCH

SEARCH BY CITATION

Keywords:

  • inflammatory bowel disease;
  • ulcerative colitis;
  • CD25+ Treg;
  • T cell suppression;
  • TCR;
  • Vβ repertoire

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. References

Background: Factors determining the extension and degree of inflammation in the colonic mucosa of patients with ulcerative colitis (UC) are largely unknown, but CD4+CD25high regulatory T cells (Tregs) have been implicated to play an important role in suppressing inflammation. Therefore, the aims of this study were to determine whether colonic Tregs have suppressive effects on colonic effector T cells in UC and to analyze the association between segmental colonic Treg distribution and disease activity. Materials and Methods: The suppressive activity of colonic CD4+CD25high Tregs from patients with active UC was determined in coculture assays measuring proliferation and cytokine production. The frequency of Tregs and the expression of the Treg marker FOXP3 were analyzed with flow cytometry and RT-PCR in isolated cells and the whole mucosa from patients with active and inactive disease, as well as healthy mucosa. Results: Colonic CD4+CD25high T cells from patients with UC suppressed the proliferation and cytokine secretion of colonic effector CD4+ T cells. Healthy controls but not patients with UC had lower Treg frequencies in the sigmoid than in the ascending colon. Patients with UC with active disease had increased frequency of colonic Tregs. The frequency of Tregs was positively correlated with colonic disease activity and serum C-reactive protein. Conclusions: Colonic CD4+CD25high Tregs are able to suppress colonic effector T cell activity in vitro, and the Treg frequency in the inflamed intestine increases with disease activity in patients with active UC. This suggests that Tregs may be outnumbered by other inflammatory cells or that their suppressive activity may be influenced by the in vivo environment.

The extension of ulcerative colitis (UC) is variable; it can be confined to the rectum (proctitis), can involve the colonic mucosa up to the left flexure (proctosigmoiditis), or can extend above the left flexure (extensive colitis). Colitis confined to the distal colon only is characterized by a milder clinical course, and the patients generally are more responsive to therapy than patients with extensive colitis.1 The factors determining the extension and degree of activity of the mucosal inflammation are largely unknown.

Studies in mouse models of experimental colitis have demonstrated that intestinal inflammation not only can be prevented but also can be cured by the presence of CD4+ CD25+ T cells with regulatory properties in the gut mucosa.2–4 Recent reports also have described the occurrence of CD4+ T cells with the phenotype of regulatory cells expressing high levels of CD25 (CD4+CD25high T cells) in the colonic mucosa of healthy individuals and patients with inflammatory bowel disease (IBD).5,6 The CD4+CD25high T cells from inflamed human mucosa maintain the regulatory phenotype after propagation in vitro.7 Freshly isolated and propagated colonic CD4+CD25high T cells from patients with IBD also have been shown to suppress the activity of autologous blood effector cells.5,7 However, although previous studies indicate that colonic effector T cells from patients with IBD are nonresponsive to T cell-mediated suppression,8–10 it is also important to investigate whether colonic effector T cells can be suppressed by CD4+CD25high T cells.

The suppressive effects of CD4+CD25high regulatory T cells (Tregs) are most likely dependent on cell-cell contact, although the mechanisms by which Tregs suppress naïve and effector T cells activity are so far inadequately understood. It has been shown that Tregs need antigen-specific or polyclonal activation via the T cell receptor to exert suppressive activity, at least in vitro.11,12 Recent studies have shown that the FOXP3 gene is important in the development and function13 of CD4+CD25high T cells in both humans and mice14 and that a defective FOXP3 expression generates strong activation of the immune system, resulting in multiorgan autoimmune diseases, including chronic intestinal inflammation.15,16 Furthermore, CD4+CD25high T cells have increased intracellular expression of cytotoxic T lymphocyte-associated antigen 4 (CTLA-4),17,18 which is a negative regulator of T cell activity.

Even though it has previously been shown that naturally occurring CD4+CD25high T cells are present in the colonic mucosa of patients with IBD, many questions remain concerning the function of the CD4+CD25high T cells and the colonic distribution of the cells in these patients. We therefore investigated the ability of colonic CD4+CD25high T cells from patients with UC to suppress colonic and blood effector/naïve T cells. Furthermore, we examined the segmental distribution of CD4+CD25high T cells and studied the relationship between colonic CD4+CD25high T cell frequency and colonic/systemic disease activity in patients with UC in the exacerbation phase.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. References

Volunteers and Collection of Specimens

Study subjects were recruited among patients referred for colonoscopy at Sahlgrenka University Hospital, Göteborg, Sweden. Venous blood and biopsies from the ascending colon and sigmoid colon, respectively, were assessed in 37 patients with UC, all with well-defined disease (29 women; mean age 44.3 ± 13.3 years; body mass index, 24.3 ± 3.5 kg/m2). Disease activity in each patient was analyzed with regard to both colonoscopic and histopathological data. Eighteen patients with UC had active extensive colitis (UCa); 22 patients with UC were in remission (UCr). Furthermore, tissue from resected sigmoid colon and blood were obtained from 3 patients with UC undergoing surgery for severe colitis (2 men; mean age 46.3 ± 9.2 years; body mass index, 22.3 ± 2.2 kg/m2). Biopsies and blood also were assessed from 18 asymptomatic control subjects (8 women; mean age 56.9 ± 10.5 years; body mass index, 25.9 ± 4.7 kg/m2) who underwent colonoscopy for investigation of anemia, rectal bleeding, or polyp surveillance. These subjects had a normal colonic mucosa as determined macroscopically and microscopically. The study was performed after written informed consent was obtained from all of the subjects and approval was granted by the Human Research Ethical Committee of the Medical Faculty, Göteborg University.

Quantification of Morphological Inflammatory Activity and Systemic Disease Activity

At least 3 standard formalin-embedded biopsies were taken from each studied segment and evaluated by a trained pathologist according to normal hospital standards. The degree of inflammation (lymphocyte infiltration), extent of crypt affection (crypt abscesses), and extent of superficial damage (ulcers, hemorrhage) were evaluated blindly by an experienced clinician (H.S.) and converted into a quantitative score with a maximum of 10 points. Immunological parameters were then related to this score. As a marker for systemic inflammation, we used the C-reactive protein (CRP) value measured in serum according to standard hospital routines.

Isolation of Intestinal and Peripheral Blood Lymphocytes

Lamina propria lymphocytes (LPLs) were isolated with collagenase/DNase enzymatic digestion as previously described.19 Briefly, the epithelium was removed by incubation in Hank's balanced salt solution, without calcium or magnesium, containing 1 mmol/L EDTA and 1 mmol/L DTT (Sigma, St Louis, Mo) 4 times for 15 min. After the biopsies were washed twice for 15 min in Hank's balanced salt solution, LPLs were isolated by stirring remaining tissue in Iscove's complete medium containing 100 U/mL collagenase (Sigma) and 0.1 mg/mL DNAse (Sigma) at 37°C for 2 h. The obtained cell suspension was filtered through a nylon mesh, and the number of lymphocytes was counted under the microscope. The resection tissue was first stripped of all muscle and fat layers before being cut into small pieces and then treated the same way as the biopsies. Peripheral blood mononuclear cells (PBMCs) were isolated from venous blood by density gradient centrifugation on Ficoll-Paque (Pharmacia, Uppsala, Sweden).

Flow Cytometric Staining, Analysis, and Sorting

Freshly isolated cells, 2 × 105cells per sample, were stained for flow cytometric analysis of various surface markers using combinations of the following antibodies: anti-CD8-FITC, anti-CD4-PerCP, anti-CD25-APC, and anti-CD69-PE (BD Pharmingen, San Diego, Calif). Before staining with PE-conjugated anti-CTLA-4, the cells were fixed and permeabilized with Cytofix/Cytoperm (BD Pharmingen). Intracellular staining and washing were performed in Perm/Wash solution (BD Pharmingen). All cells were fixed in Cellfix (BD Pharmingen) before flow cytometric analysis, which was performed with an LSR II (BD Pharmingen). The data were analyzed with Flow Jo software (Treestar, Ashland, Ore), and lymphocytes were gated according to their forward- and side-scatter properties. T cells were identified on the basis of their expression of CD4 or CD8. To discriminate between CD25high Tregs and CD25low activated effector T cells, we included into the gate for CD25high T cells only CD4+ cells expressing CD25 with higher intensities than the CD8+ cells, which express only moderate levels of CD25. For all markers, unstained cells and cells stained with isotype-matched control antibodies served as controls.

To sort cells into CD4+CD25high, CD4+CD25low, and CD4+CD25low/− fractions, LPLs and PBMCs were labeled with anti-CD25-PE, anti-CD4-FITC, and anti-CD8-APC (all from BD Pharmingen), and the cells were sorted with a FACSVantage SE (BD, San Jose, CA) operating at sheath pressure of 22 psi. After sorting, the CD25high fraction contained >96% CD4+CD25high cells, the CD25low fraction contained <16% CD4+CD25high cells, and the CD25low/− fraction contained <0.5% CD4+CD25high cells.

Proliferation and Suppression Assays

As antigen presenting cells (APCs), we used monocytes isolated from PBMCs by adherence of the remaining cell population after CD4 and CD8 cell depletion using Dynabead Positive Isolation Kit (Dynal Biotech ASA, Oslo, Norway). Then, 1 × 105 CD4CD8 mononuclear cells were incubated in round-bottomed 96-well plates for 2 h at 37°C in 5% CO2. Wells were washed, and the remaining adherent cells were used as APCs in the following experiment. The suppressive capacity of CD25high cells was determined by adding sorted CD4+CD25low/− and/or CD4+CD25high cells at various ratios and incubating the cells for 5 days in Iscove's medium supplemented with 5% AB+ serum, 1% Gentamicin (Sigma), and 1% l-glutamine (Sigma). Cultures were stimulated with 1 μg/mL soluble anti-CD3mAb (OKT-3, Ortho-McNeil Pharmaceutical, Raritan, NJ). Cell culture supernatant (100 μL) was removed from each well after 48 h and replaced with fresh medium. Replicated supernatants were pooled and stored at −70°C until analysis of cytokine content. After 72 more hours of culture, the T cell proliferation was measured by pulsing the cells with 0.5 μCi [3H] thymidine per well (Amersham, Arlington Heights, Ill) for 16 h; then, the incorporated radioactivity was analyzed with a scintillation counter.

Cytokine Assays

The amounts of released interleukin (IL)-2, IL-10, and INF-γ in culture supernatants were measured by cytometric bead array (Human TH1/ TH2 cytokine CBA kit, BD Pharmingen) according to manufacturer's instructions.

Analysis of FOXP3 mRNA Levels by Quantitative Real-time Polymerase Chain Reaction (RT-PCR)

Total RNA was extracted from mucosal biopsies with the RNeasy Mini kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Contaminating DNA was removed by treatment with DNA-free (Ambion Austin, Tex). cDNA was prepared in a random hexamer-primed superscript RT reaction (Invitrogen, Stockholm, Sweden) according to manufacturer's protocol. FOXP3 mRNA levels were measured in duplicate with the LightCycler (Roche Diagnostics, Mannheim, Germany) with reagents from the LightCycler FastStart DNA master SYBR Green I kit and the following primers: 5′-CAG CAC ATT CCC AGA GTT CCT-3′ (forward) and 5′ GCG TGT GAA CCA GTG GTA GAT-3′ (reverse). PBGD was used as the endogenous reference gene for relative quantification and was detected by these primers: 5′-GAA ACC CTG CCA GAG AAG A-3′ (forward) and 5′-CTG GCC CAC AGC ATA CAT-3′. All primers were designed not to amplify genomic DNA and were ordered from TIB MOLBIOL (Berlin, Germany). PCR cycling conditions consisted of 95°C for 10 min, followed by 45 cycles of 95°C for 15 s, 65°C for 7 s, and 72°C for 10 s. Cycle threshold values for each gene were compared against a standard curve to estimate starting amounts of RNA, and received values for the target gene were normalized to threshold values for the reference gene. A melting curve analysis was performed in each run to ensure the specificity of the primers, and data were obtained using with LightCycler data analysis software. LightCycler relative quantification software was used for calibrator-normalized relative quantification.

Isolated CD25high and CD25low/− PBMCs and LPLs also were analyzed regarding FOXP3 expression. RNA was extracted from each cell sample using the same method as above, although RNeasy Micro kit (Qiagen, Hilden, Germany) was used for maximum yield of RNA. GAPDH was used as a reference gene because PBGD expression was too low in RNA from isolated cells compared with RNA from biopsies. GAPDH was detected by these primers: 5′- GGC TGC TTT TAA CTC TGG-3′ (forward) and 5′-GGA GGG ATC TCG CTC C -3′ (reverse), also ordered from TIB MOLBIOL and designed not to amplify genomic DNA.

Analysis of TCR Vβ Repertoire

The TCR Vβ repertoire of PBMC and LPL cells was analyzed by flow cytometry with FACSCalibur (BD Bioscience, San Diego, Calif). In this study, 1.7 × 106 cells were stained with anti-CD4-PerCP and anti-CD25-APC (BD Pharmingen), together with a TCR Vβ repertoire kit (Beckman Coulter, Fullerton, Calif) containing PE- and/or FITC-conjugated anti-Vβ-antibodies according to the manufacturer's instructions. A control staining also was performed with anti-CD3-FITC, anti-CD8-PE, anti-CD4-PerCP, and anti-CD25-APC (BD Pharmigen) to set the gate for CD25high cells.

Statistical Analysis

All statistical evaluations were performed with the StatView Software (SAS Institute, Cary, NC). Simple regression analysis, a Wilcoxon signed paired test, or a nonparametric Mann-Whitney test was used to evaluate significance of differences.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. References

Colonic CD4+CD25high Cells from Patients with UC Suppress Colonic T Cell Activity

We evaluated the suppressive activity of colonic CD4+CD25high T cells on both colonic and blood effector/naïve T cells. CD4+CD25high T cells were isolated from colonic sigmoid resections of 3 patients with UC undergoing surgery for severe colitis, and the CD4+CD25high T cells were cocultured with autologous colonic or blood CD4+CD25low/− T cells. To discriminate CD4+CD25high T cells from effector CD4+CD25low/− T cells, the CD25 expression on CD8+ T cells was used as an internal control because CD8+ T cells express only moderate levels of CD25 (Fig. 1A).20 The colonic CD25high T cells proliferated poorly, whereas pure colonic CD4+ T cells proliferated vigorously in response to anti-CD3 antibodies (Fig. 1B). Depletion of CD25high T cells increased the proliferative capacity of the CD4+ T cells even further, and the proliferation of the colonic CD25low/− T cells was markedly reduced by coculture of colonic CD25high T cells at a ratio of 1:1 (Fig. 1B). The CD25high T cells suppressed the proliferation of the cocultured CD25low/− T cells in a dose-dependent fashion (Fig. 1C). Thus, colonic CD4+CD25high T cells isolated from patients with UCa suppress the proliferation of autologous colonic effector CD4+ T cells. In addition, we found that the colonic and blood CD25high T cells seemed to be equally suppressive because coculture of either CD25high T cell population with blood CD25low/− T cells diminished proliferation to a similar extent (Fig. 1D).

thumbnail image

Figure FIGURE 1. Colonic CD4+CD25high T cells suppress the proliferation of colonic and blood effector T cells in UC. The capacity of colonic and blood CD4+CD25high T cells to suppress the proliferation of CD4+CD25low/− colonic or blood effector T cells was examined. Lymphocytes were isolated from the sigmoid colon and blood of 3 patients with UC undergoing colonic surgery, and CD25high and CD25low/− cells were sorted by flow cytometry (A). The various cell populations were cultured either alone or together in the presence of APCs; the cells were stimulated with α-CD3 antibodies, and proliferation was determined on day 5. B, The suppressor-to-responder ratio was 1:1, and the proliferative response of the total CD4+ T cell population was set to 100% in each experiment. Values shown are the means and SD of 3 independent experiments. The Mann-Whitney test was used to evaluate differences. C, Coculture experiment with various ratios of CD25high and CD25low/− cells isolated from the sigmoid colon of an active UC patient. D, Coculture experiment with colonic or blood CD25high T cells cultured together (1:1 ratio) with blood CD25low/− T cells isolated from a UCa patient.

Download figure to PowerPoint

We also analyzed the ability of CD4+CD25high T cells to suppress cytokine secretion from colonic effector/memory T cells. After polyclonal stimulation, CD4+CD25low/− cells produced both IFN-γ (mean, 335 pg/mL) and IL-2 (mean, 30 pg/mL). In coculture experiments, CD4+CD25high cells suppressed the release of both of these cytokines, although the suppression was less consistent (11% and 56% mean suppression, respectively) than that seen for the proliferation. In contrast to the CD25low/− effector T cells, CD25high cells did not produce any detectable levels of IFN-γ or IL-2. However, both CD25low/− and CD25high cells produced IL-10 when stimulated alone (mean, 78 and 8 pg/mL, respectively), and IL-10 also was found in supernatants taken from cocultures of the cells (mean, 63 pg/mL).

In conclusion, colonic CD4+CD25high T cells from patients with UC in exacerbation are able to suppress both proliferation and cytokine secretion of autologous colonic nonregulatory T cells.

Increased Frequency of Colonic CD4+CD25high Tregs in Patients with Active UC

We next examined the frequency of the CD25high Tregs in the colonic mucosa in health and disease. We compared the frequencies of colonic Tregs in the ascending colon and sigmoid colon from UCa or UCr patients with those of healthy control subjects. In the ascending colon and sigmoid colon, UCa patients had significantly increased frequencies of CD25high Tregs among CD4+ T cells relative to both patients with UC in remission and control subjects (Fig. 2A). In contrast, the CD25high Treg frequency tended to be slightly increased in the blood of UCr patients compared with UCa patients or control subjects. The ratio of CD25high Tregs to CD4+CD25low activated T cells also was significantly increased in both the ascending colon and the sigmoid colon of UCa patients compared with UCr patients or control subjects (Fig. 2B). In contrast, the ratio of blood CD25high Tregs to activated blood T cells was similar in all investigated subject groups. Consequently, the colonic CD25high Treg frequency was ≈2.5-fold higher in both colonic regions relative to blood in UCa patients (P = 0.003) (Fig. 2A and 2B).

thumbnail image

Figure FIGURE 2. Regional distribution of CD4+CD25high T cells in the ascending (Asc) colon and sigmoid (Sigm) colon in UCa patients, UCr patients, and control subjects (Ctrl). The frequency of CD25high cells among CD4+ LPLs was analyzed in the ascending and sigmoid colons and blood from UCa and UCr patients and control subjects by flow cytometry. A, Each symbol represents the fraction of CD25high cells in 1 individual; horizontal lines depict median values. B, The ratio of CD25high to CD25low cell frequencies in each individual. Horizontal lines indicate median values. Mann-Whitney test was used (A and B) to evaluate differences. C, The frequency of CD25high T cells among CD4+ T LPLs isolated from the ascending colon relative to the sigmoid colon. Connecting lines show values from paired samples taken from the same individual. Wilcoxon signed-paired test was used to evaluate differences.

Download figure to PowerPoint

We also analyzed whether there were any regional differences in the distribution of CD25high Tregs in the ascending colon and sigmoid colon in UCa patients, UCr patients, and control subjects. The frequency of CD25high Tregs among CD4+ T cells was significantly higher in the ascending colon compared with the sigmoid colon of healthy control subjects (Fig 2C). This regional difference in the frequency of colonic CD25high Tregs was not found in either UCa or UCr patients (Fig. 2C). Taken together, our data show that UCa patients with extensive colonic inflammation have increased frequencies of CD25high Tregs in the sigmoid colon and ascending colonic mucosa compared with healthy control subjects and that the colonic regional variation in CD25high Treg frequency found in normal individuals is absent in patients with UC.

Colonic CD4+CD25high Tregs and Colonic Biopsies from Patients with Active UC Are Characterized by High FOXP3 Expression

FOXP3 is a transcription factor previously shown to be preferentially expressed in blood CD25high Tregs of humans and mice.19,21 Makita et al5 have recently shown that purified colonic CD25high T cells of healthy subjects express high levels of FOXP3. We investigated the association between expression of FOXP3 and CD25 in inflamed human mucosa. The FOXP3 mRNA level was ≈5-fold higher in colonic CD4+CD25high T cells than in colonic CD4+CD25low/− T cells (Fig. 3A). Thus, our results show that most FOXP3-expressing cells in the colonic mucosa are contained within the CD25high fraction. We also found that colonic CD25high cells expressed ≈2-fold-higher levels of FOXP3 relative to blood CD25high cells (Fig 3A, B).

thumbnail image

Figure FIGURE 3. Expression of FOXP3 in colonic and blood T cell populations and colonic mucosa in patients with UC and control subjects (Ctrl). CD4+ T cells isolated from the sigmoid colon (A) and blood (B) of patients with UC undergoing colonic surgery were sorted into CD4+CD25high and CD4+CD25low/− cells with flow cytometry. The FOXP3:GAPDH mRNA levels were analyzed by quantitative RT-PCR. Values shown are the means and SD of normalized values from 2 independent experiments. C, Levels of FOXP3:PBGD mRNA in whole tissue material isolated from the ascending and sigmoid colon from UCa patients, UCr patients, and control subjects as determined by RT-PCR. Each symbol represents values from 1 individual; horizontal lines depict median values. Mann-Whitney test was used to evaluate differences.

Download figure to PowerPoint

Next, we analyzed the total FOXP3 expression in homogenized colonic tissue from the different subject groups. We found 3-fold-higher FOXP3 mRNA levels in biopsies from the ascending colon and 6-fold-higher FOXP3 mRNA levels in the sigmoid colon of UCa patients relative to control subjects. Furthermore, the FOXP3 mRNA levels tended to be higher in the ascending colon and 5-fold higher in the sigmoid colon in UCa patients relative to UCr patients (Fig 3C).

Phenotypic Properties of Colonic CD4+CD25high Tregs in UC

To further characterize and compare the CD25high Treg populations in the ascending colon and sigmoid colon of patients with UC, we investigated the expression of surface markers by flow cytometry. Blood CD25high Tregs have a low expression of the activation marker CD69, whereas most colonic CD25high Tregs, regardless of regional localization, expressed CD69. No difference in CD69 expression was noted between subject groups in either blood or intestinal mucosa (data not shown). Furthermore, the expression of CD69 on colonic and blood CD25high Tregs was comparable to that found on CD4+CD25low T cells in colon and blood, respectively.

To investigate whether the colonic CD25high Treg population expanded in situ in response to specific antigen(s) or was nonspecifically recruited to the mucosa from the peripheral CD25high Treg pool as a result of the inflammatory conditions, the T cell receptor repertoire of colonic and blood CD25high Treg populations was compared. The Vβ repertoire of colonic CD25high Tregs from UCa patients was polyclonal and overall did mirror the Vβ repertoire of blood CD25high Tregs (Fig. 4A). In addition, the Vβ repertoire of the colonic CD4+CD25low/− T cell population from patients with UC was polyclonal and reflected the Vβ repertoire recorded on the same cell population in blood (Fig. 4A).

thumbnail image

Figure FIGURE 4. Phenotypic properties of colonic CD4+CD25high T cells in UC. A, The TCR Vβ repertoire of blood and colonic CD4+ CD25high Tregs and CD4+CD25low/−T effector cells from UCa patients was analyzed by flow cytometry. The results shown are from 1 of 2 experiments. B, Frequencies of CD25high and CD25low/− CD4+ T cells expressing CTLA-4 were evaluated in PBMCs and LPLs from the ascending colon and sigmoid colon in UCa patients, UCr patients, and control subjects (Ctrl) by flow cytometry. Values shown are mean and SD (n = 6). Black bars represent CD4+ CD25high T cells; white bars, CD4+ CD25low/− T cells.

Download figure to PowerPoint

Because CTLA-4 has been suggested to be important for the suppressive mechanisms of CD25high Tregs,2,17 we analyzed the CTLA-4 expression on colonic and blood CD25high Treg populations. CD25high Tregs expressed more CTLA-4 than CD25low/− cells in all tissues investigated, but the frequency of CTLA-4-expressing CD25high Treg was augmented in the colon compared with blood in all subject groups investigated (P = 0.003) (Fig. 4B). CTLA-4 expression was comparable on CD25high Tregs of the ascending colon and sigmoid colon in both health and disease. However, an increased frequency of colonic CD25high Tregs expressing CTLA-4 was recorded in both the ascending colon and sigmoid colon of UCa patients compared with UCr patients (Fig. 4B).

Frequency of Colonic CD4+CD25high Tregs Correlates with Colonic and Systemic Disease Activity

Finally, we investigated whether there was any correlation between the frequency of colonic CD25high Tregs and disease activity in patients with extensive UCa. The frequency of CD25high Tregs in the ascending colon was positively correlated to disease activity in the same colonic region on the basis of morphological changes in mucosal biopsies. The same pattern was seen in the sigmoid colon (Fig. 5A). We then investigated whether there was any correlation between colonic CD25high Treg frequency and systemic disease activity, as judged by the levels of CRP in serum. We found that the CRP level in serum was positively correlated to the frequency of CD25high Tregs in both the ascending colon and sigmoid colon (Fig. 5B). Neither colonic disease activity nor CRP values, however, correlated to the frequency of CD25high Tregs in the periphery (data not shown). Thus, the frequency of mucosal CD25high Tregs in UCa patients is correlated to the colonic and systemic disease activity score.

thumbnail image

Figure FIGURE 5. Colonic CD4+CD25high T cell frequencies correlate with colonic and systemic UC activity scores. The frequencies of CD25high cells among CD4+ T LPLs isolated from the ascending colon and sigmoid colon of UCa were correlated to (A) colonic disease activity (score 1-10) in the respective colonic segments and to (B) CRP in serum. Simple regression analysis was used to evaluate the correlation.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. References

The presence of CD4+CD25+ Tregs has been reported to prevent the outbreak of colitis and to cure established intestinal inflammation in experimental animal models of IBD.2–4 Despite the promising data of the putative therapeutic effects of Tregs obtained from such experimental colitis models, the role of Tregs in human IBD is largely unknown. Recently, Makita et al5 demonstrated that colonic Tregs from patients with IBD can suppress the activity of blood effector T cells. However, the important question of whether lamina propria T cells from the inflamed colons of patients with IBD are responsive to Treg-mediated suppression has not previously been addressed. Therefore, in the present study, we analyzed whether colonic effector T cells can be suppressed by colonic Tregs from patients with UC. We found that colonic Tregs are indeed able to suppress both the proliferation and the cytokine production of colonic effector T cells. Thus, our results indicate that Tregs isolated from the site of active inflammation retain their suppressive activity and that effector T cells from inflamed tissues are able to respond to the suppression.

Our study also shows that the patients with UCa have higher frequencies of colonic Tregs relative to patients with UCr and healthy control subjects. Our data are compatible with the work recently presented by Makita et al,5 who reported increased ratios of cells with a Treg phenotype in inflamed lesions of active patients with IBD compared with noninflamed mucosa of the healthy individuals. In addition, we found a strong association between amplified local and systemic activity of the disease and the presence of colonic Treg frequencies. The reason for the insufficient suppression of the intestinal inflammation despite the accumulation of an apparently suppressive colonic Treg pool in patients with UC remains unclear. However, the regulatory effect of the colonic Treg population on colonic effector T cells from patients with UC shown in our study was obtained with high suppressor-to-responder ratios (1:1 and 1:2), and the suppressive effect was lost at lower ratios. In vivo, even higher proportions of Tregs may be required to fully suppress inflammation because cell-to-cell contact is likely to be important for suppression. The beneficial effects of Tregs seen in the murine colitis settings also were obtained by administrating excessive numbers of Tregs.22–24 Thus, although Tregs accumulate in the intestinal mucosa of patients with IBD, the frequency may not be enough to restrain the intestinal inflammation. In support of this hypothesis, Maul et al6 found that patients with diverticulitis harbored significantly increased frequencies of cells with a Treg phenotype in inflamed lesions compared with patients with active IBD.

We also analyzed Treg frequencies in the blood of the different study groups. Consistent with the results presented by Maul et al,6 we found decreased frequencies of blood Treg in patients with UCa relative to patients with UCr. This supports the notion that during exacerbation, Tregs are recruited from the blood to the inflamed tissue. We also found that the T cell receptors of colonic and blood Tregs have similar polyclonal Vβ T cell receptor repertoires. Thus, there is an accumulation of Tregs with a polyclonal Vβ T cell receptor repertoire analogous to that of blood Tregs in the colonic mucosa of patients with UCa. This indicates that the increase in colonic Tregs in inflamed tissue is caused by recruitment from the peripheral Treg pool, not by local expansion of antigen-specific clones in the gut.

We showed that most individuals with healthy colonic mucosa have lower Treg frequencies in the sigmoid compared with the ascending colon. The finding of lower Treg frequencies, and thus a reduced ability to suppress effector T cells responses, at the left side of the colon of healthy individuals may explain at least in part why distal inflammation is typical at the onset of UC. The normal segmental variation of colonic Treg frequencies was lost in patients with established UC, suggesting that more Tregs are recruited to the site at the onset of colitis but that it is not enough to stop disease progression.

Our finding of an accumulation of Tregs able to suppress colonic effector T cells at the site of inflammation is consistent with studies of Tregs in patients with rheumatoid arthritis. Several groups have demonstrated Treg suppression of effector T cells isolated from inflamed joints, and patients with rheumatoid arthritis have increased frequencies of Tregs in inflamed joints relative to blood.25–28 Both UC and rheumatoid arthritis are remitting and relapsing chronic inflammatory disorders that afflict patients from the time of onset throughout the lifespan. Both diseases are characterized by tissue destruction and accumulation of leukocytes in the inflamed organ. Furthermore, arthritis is recognized as an extraintestinal manifestation of patients with UC. It is therefore intriguing to note that these immunologically closely related diseases share the feature of having functional Treg populations but still have insufficient regulation of T cell activity in the inflamed areas.

It cannot be entirely excluded that the suppressive activity of Tregs and their ability to regulate other cells may be influenced by factors present in the local in situ environment that are lacking in the in vitro culture of separated cells. For example, Toll-like receptor 2 (TLR2) triggering, together with T cell receptor signaling, results in a temporal loss of the suppressive Treg capacity that is regained after removal of the TLR2 trigger,29 and TLR8 ligation reverses the suppressive function of Tregs.30 Furthermore, studies have demonstrated that IL-6 produced by TLR-activated dendritic cells renders antigen-specific T cells nonreceptive to Treg suppression.31 However, lipopolysaccharide and flagellin signaling via TLR4 and TLR5, respectively, potently increased the suppressive capacity and enhanced FOXP3 expression of CD4+CD25+ Tregs.32,33 Thus, it is clear that further studies are needed to fully understand the role of factors influencing the capacity of Tregs to control inflammation in vivo.

In conclusion, this article demonstrates that colonic Tregs from patients with UC have a suppressive effect on colonic effector T cells in vitro. However, the frequency of colonic Tregs increases with colonic and systemic disease activity, suggesting that Tregs may be outnumbered by other inflammatory cells or that their suppressive activity may be influenced by the in vivo environment. Nevertheless, our results demonstrate that colonic Tregs from patients with colitis have the potential to downregulate T cell-mediated inflammation, given the right stimuli. Thus, Tregs are important targets for future immunotherapy against colitis, and future trials may be designed to investigate whether transfer of in vitro-propagated Tregs to patients with colitis or enhanced in vivo recruitment, expansion, and/or activation of Tregs may be able to control mucosal inflammation.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. References

This study was supported by the Swedish Medical Research Council; AstraZeneca R & D, Mölndal, Sweden; Nanna Swartz Foundation; Sahlgren's Academy Hospital Foundation; Tore Nilsson Foundation; Wilhelm & Martina Lundgren Foundation; Nio Meter Liv Foundation; Goljes Foundation; and the Swedish Society of Medicine.

References

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. References