Published online before print October 13, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
T cells in different types of IBD animal models
* Medizinische Klinik I, and
Institute of Pathology, Charité, University Medicine Berlin, Campus Benjamin Franklin, Berlin, Germany
1Correspondence: Medizinische Klinik I Charité - Universitätsmedizin Berlin, Campus Benjamin Franklin, Hindenburgdamm 30, Berlin D-12200, Germany. E-mail: joerg.hoffmann{at}charite.de
ABSTRACT
The role of 
T cells in inflammatory bowel disease (IBD) is still controversial. Although T cells induce IBD in immunodeficient animals, others suggest a protective role of T cells. Therefore, this study was conducted in order to elucidate the effect of T cell depletion/deficiency on different IBD animal models. Mice depleted of or deficient in T cells were exposed to dextran sodium sulfate (DSS) in order to induce colitis. In addition, T cells were depleted in mice with terminal ileitis (TNF
ARE) or colitis due to interleukin 2 deficiency (IL-2 ko). Finally, DSS-induced colitis was studied in mice deficient in interferon gamma (IFN- ko) upon T cell depletion. Depletion of T cells aggravated DSS-induced colitis and terminal ileitis of TNFARE mice. Exacerbated DSS-induced colitis was also found in T cell-deficient mice. IL-2 ko mice showed increased mortality upon early (starting at 4 wk of age) but not late depletion (starting at 8 wk of age). Early T cell depletion or deficiency resulted in increased IFN- production by both lamina propria lymphocytes and splenocytes in every model investigated herein. In IFN- ko mice, T cell depletion did not affect the development and course of DSS-induced colitis. The protective effect of T cells in IBD was confirmed in various IBD animal models. Particularly, during the early phase of intestinal inflammation, T cells appear to be important. The mechanism seems to involve the control of IFN- production and epithelial regeneration.
Key Words: colitis • epithelial regeneration • interferon • terminal ileitis
INTRODUCTION
T cells bearing a receptor composed of a gamma and a delta chain ( T cells) make up 40–50% of intraepithelial lymphocytes (IEL) in mice and men but less in peripheral lymphoid organs (<5%) [1
, 2
]. These IEL play a dual role in protecting the intestinal mucosa: they provide first-line defense against infectious agents, and, subsequently, they are involved in intestinal immunoregulation [3
4
5
]. This is accomplished by supporting IgA production and by maintaining homeostasis of the intestinal epithelium through an intranet with 
ß T cells and epithelial cells [3
4
5
]. Upon injury, IEL clear necrotic epithelium and promote epithelial cell growth probably via secretion of keratinocyte growth factor (KGF) [4
, 6
7
8
9
10
]. In addition, they are involved in tolerance induction by secretion of interleukin 10 (IL-10) and in tumor defense by interferon-gamma (IFN-) production [11
, 12
].
Conflicting evidence exists regarding the role of T cells in inflammatory bowel disease (IBD). On the one hand, increased numbers of T cells were found in the mucosa and blood of IBD patients [13
14
15
] and data from TCR- ko mice and an adoptive transfer model suggest that T cells induce colitis [16
17
18
]. On the other hand, T cells seem to protect the epithelium against injury since rat TNBS-induced colitis was aggravated by T cell depletion [19
, 20
]. Additionally, impaired epithelial repair was observed in T cell ko mice exposed to DSS, probably due to insufficient KGF production [7
]. Therefore, it is currently unclear whether T cells have a more proinflammatory or antiinflammatory role in IBD.
In the present study, the role of T cells in various IBD animal models was examined. The colitis course upon T cell depletion/deficiency was studied in an inducible model (DSS-induced colitis) and in an immunological model (IL-2 ko mice), both representative for human ulcerative colitis. In addition, the course of terminal ileitis upon T cell depletion was studied in mice heterozygously deficient of TNFARE, a Crohn’s disease animal model [21
, 22
]. Our data clearly show that depletion of or deficiency in T cells aggravates intestinal inflammation in every investigated model, particularly in the early phase of colitis. This might be due to the regulatory role of T cells with regard to IFN- secretion by ß T cells as development and course of DSS-induced colitis in IFN- ko mice was unaffected by T cell depletion.
MATERIALS AND METHODS
Animals
Female C57BL/6 mice were obtained from the Bundesinstitut für Risikobewertung (Berlin, Germany), and female C57BL/6 mice deficient in T cells ( TCR ko), as well as deficient in IFN- (IFN- ko) from The Jackson ImmunoResearch Laboratory (Bar Harbor, ME), all at 8 wk of age. They were housed under conventional conditions receiving standard chow and water ad libitum.
Interleukin 2 deficient (IL-2 ko) mice were provided by I. Horak (Berlin, Germany). They were housed under specific pathogen-free conditions. Homozygous IL-2 ko mice were bred using heterozygous parents. The genotype was established by polymerase chain reaction (PCR) analysis of DNA from tails using DNeasy Tissue Kit (Qiagen, Berlin, Germany) using the following primers: sense 5'-TCG AAT TCG CCA ATG ACA AGA CGC T-3', antisense 5'-CTA GGC CAC AGA ATT GAA AGA TCT-3', and antisense 5'-GTA GGT GGA AAT TCT AGC ATC ATC C-3'.
Breeder TNFARE mice were provided by G. Kollias, Athens, Greece. They were housed under specific pathogen-free conditions. Heterozygous TNFARE/+ mice were bred from heterozygous (TNFARE/+) and wild-type (++) parents. The following primers were used for genotyping: antisense 5'-CCT TCC TCA CAG AGC CAG CCC CCT C-3'and sense 5'-AAT TAC GGT TAG GCT CCT GTT TCC-3'.
Monoclonal antibodies
For T cell depletion sterile filtered 13D5 (murine TCR; hamster IgG3; ATCC), was injected intraperitoneally (i.p.). Hamster IgG (Dianova, Hamburg, Germany) served as control. The following mAbs were used in vitro: anti-Hamster Biotin (Jackson ImmunoResearch Laboratories), anti-rat Cy5 (Jackson ImmunoResearch Laboratories), CY5-conjugated AffiniPure F(ab')2 fragment donkey anti-rat IgG (H+L) (Jackson ImmunoResearch Laboratories), Cy3-conjugated Streptavidin (Jackson ImmunoResearch Laboratories), 145-2C11 (murine CD3
; hamster IgG1; ATCC); 37.51 (murine CD28; hamster IgG2; BD Biosciences, Berlin, Heidelberg, Germany); GL3-FITC (murine TCR; hamster IgG2; BD Biosciences). 13D5 and 145-2C11 were purified from supernatants using spinner flasks followed by affinity chromatography employing protein G sepharose.
T cell depletion in vivo
Depletion of T cells was performed by i.p. injection of the anti- TCR mAb 13D5 at a dose of 400 µg in 200 µL PBS followed by 200 µg in 200 µL on days 3 and 6. In all other models, the initial administration of 400 µg 13D5 was followed by a weekly dose of 200 µg in 200 µL until the 12th week. Control mice received hamster IgG. Mice were monitored for weight loss, rectal prolapse, and diarrhea. Animals were killed by cervical dislocation whenever a weight loss >20% and/or obvious signs of pain and/or lethargy occurred. Otherwise, mice with DSS-induced colitis were killed on day 11 after colitis induction, and all other mice were killed at 14 wk of age unless otherwise indicated.
Induction of colitis
In depletion studies, colitis was induced in C57BL/6 (wt and IFN- ko) mice by administering 3% dextran sulfate sodium (DSS; ICN, Berlin, Germany) via the drinking water, while TCR ko and control mice, respectively, were fed 2.5% DSS.
Histological staining and scoring
Tissue samples of cecum and proximal colon were fixed in 4% formaldehyde/PBS and embedded in paraffin. The sections (5 µm) were stained with hematoxylin and eosin. The degree of inflammation was scored by a blinded pathologist using a modified scoring system as described by Neurath et al. [23
]: 0, no signs of inflammation; 1, low level of leukocyte infiltration (10–30 leukocytes per high power field (hpf)); 2, moderate level of leukocyte infiltration (31–70 leukocytes per hpf); 3, high level of leukocyte infiltration (>71 leukocytes per hpf), high vascular density, thickening of the bowel wall; 4, transmural infiltrations, loss of goblet cells, high vascular density, strong bowel wall thickening, ulcerations, and cryptic abcesses.
For detection of fibrin thrombi a modified Weigert’s differential stain for fibrin was used as described previously [24 ].
Immunohistology
For verifying successful T cell depletion in the intestine, colon sections were filled with diluted (1/3 with PBS) optimal cutting temperature embedding medium (O.C.T., Tissue Tek, Miles Diagnostik Division), snap frozen, and stored at –80°C before use. Five- to ten-micrometer sections were stained for GL3+ T cells ( TCR+), according to Wands et al. [25
].
Isolation of lamina propria lymphocytes and splenocytes
Lamina propria lymphocytes (LPL) were isolated, as described previously [20
]. Briefly, the colon was opened and cut into pieces. The fragments were shaken in HBSS (Ca2+/Mg2+-free) containing 1 mM DTT and 2.5 µM EDTA for 45 min with two changes of media. The pellet was washed, and DNA and collagene were digested using collagenase and DNase. Subsequently, the tissue was passed through a 70-µm mesh cell strainer (BD Biosciences) followed by Percoll density gradient centrifugation (100%/40%). LPL were collected from the interphase, washed, and resuspended in complete medium (RPMI 1640 medium containing 10% FCS, 100 U/mL penicillin/streptomycin, 3 mM glutamine, and 50 µM ß-mercaptoethanol). The viability was always greater than 90% determined by trypan blue dye exclusion.
The spleen was grinded and passed through a cell strainer followed by lysis of erythrocytes and resuspended in complete medium.
Flow cytometry
Splenocytes and LPL were double-stained u 145-2C11-PE (CD3+) and GL3-FITC (TCR+) (BD Biosciences) in order to control the T cell depletion. A total of 106 cells were stained in staining buffer according to the manufacturer
instructions. Only viable cells were analyzed using a FACS Vantage cell sorter (Becton/Dickinson).
Cytokine assay
Isolated LPL and splenocytes were incubated at 106 cells/mL in 24-well plates (NUNC, Wiesbaden, Germany) coated with 145-2C11 (10 µg/mL) and 37.51 (1 µg/mL) added at 37°C in a humidified atmosphere with 5% CO2. Supernatants were taken after 48 h and examined for cytokine secretion (IL-10, TGF-ß, and IFN-) by sandwich ELISA using antibodies, as well as recombinant protein standards for IL-10 (R&D Systems), IFN- (BD Biosciences), and TGF-ß (Promega, Mannheim, Germany) according to the manufacturer instructions.
Proliferation of LPL and splenocytes
LPL and splenocytes were incubated in triplicates for 96 h at 5 x 104 cells/well in 96-well round-bottom plates (NUNC) coated with 145-2C11 (10 µg/mL) in the presence of 1 µg/mL 37.51 at 37°C in 5% CO2 humidified air. Each well received 0.5 µCi of [3H]thymidine (Amersham Pharmacia, Little Chalfont, Buckinghamshire, UK) during the last 18 of 96 h of culture. Incorporated [6-3H]thymidine was harvested on glass fiber membrane and detected by liquid scintillation counting (LKB Wallac, Turku, Finland).
Blood count
Blood samples were taken at the beginning (tail vein) and end (heart blood) of the experiments for blood count using an automatic cytometer (Beckmann Coulter, Krefeld, Germany).
Statistical analysis
Statistical analysis was carried out using SPSS for Windows. Survival was analyzed using Kaplan-Meier analysis. For other comparisons, Mann-Whitney U test was used. Values were expressed as mean (95% confidence intervals) and standard error of the mean (SEM). Differences were considered statistically significant for P < 0.05.
RESULTS
Depletion of T cells by administration of the monoclonal antibody 13D5
To verify T cell depletion by administration of the mAb 13D5, splenocytes and LPL were analyzed by flow cytometry, and colon sections were analyzed by immunohistology. Because of the abundance of splenocytes, T cell depletion was examined during the experiments by flow cytometry of splenocytes using a noncross-blocking anti-TCR mAb. Splenic T cells were always found to be reduced >85% (data not shown), whereas TCR+ LPL were on average reduced 85% (from 4.8% GL3+ T cells in controls to 0.41% GL3+ T cells in 13D5-treated animals). This was confirmed by immunofluorescence staining of colon sections showing 1.26 ± 0.6 GL3+ T cells per hpf in controls compared with 0.28 ± 0.1 GL3+ T cells per hpf in 13D5 treated animals (P<0.05).
Depletion of T cells aggravates DSS-induced colitis in mice
To elucidate the role of T cells on tissue injury and immunological processes in murine DSS-induced colitis model, C57BL/6 mice received either a T cell-depleting antibody (13D5, n=9) or control IgG (n=5) on days 0, 3, and 6 followed by administration of 3% DSS via the drinking water. Six mice (three of each treatment group) had to be sacrificed after 8 days due to excessive weight loss, while the remaining animals were killed on day 11. Data from the mice killed on day 8 were pooled with the data of those that survived until the end of the experiment. Histopathological examinations revealed more severe colitis in T cell-depleted mice compared with controls (cecum: 3.9±0.1 vs. 3.0±0.0, P<0.05; colon: 4.0±0.0 vs. 3.0±0.0, P<0.05). However, no difference was found between both groups regarding weight course, survival, and colon length. INF- secretion of anti-CD3/CD28-stimulated LPL was significantly enhanced in T cell-depleted mice compared with controls (8.4±2.4 ng/mL vs. 1.4±0.6 ng/mL, P<0.05). A trend toward increased IL-10 secretion by LPL was seen upon anti-CD3/CD28 stimulation (1.7±0.6 ng/mL vs. 0.4±0.1 ng/mL, P=0.1).
Increased severity of DSS-induced colitis in TCR ko mice
Depletion of T cells led to more severe colitis, as well as increased secretion of the proinflammatory cytokine IFN-. To exclude side effects induced by the mAb itself TCR ko mice (n=7) and wild-type (wt) control mice (n=7) were exposed to 2.5% DSS. All animals had to be sacrificed 7 days after colitis induction due to severe disease. Both, TCR ko and wild-type mice displayed massive weight loss and rectal bleeding but no rectal prolapse due to colitis induction. As in T cell-depleted animals, DSS-induced colitis was microscopically more severe in TCR ko than in wt mice (Fig. 1
). Although wt mice revealed only mild inflammatory infiltrates, TCR ko showed severe acute tissue damage, extensive erosions, vascular fibrin thrombi, and destructed epithelial layers (Fig. 2
). In addition, increased IFN- and decreased IL-10 production were found in anti-CD3/CD28-stimulated splenocytes (Table 1
). LPL of T cell ko mice also showed increased IFN- and IL-10 production compared with controls as in the previous depletion experiment (Table 1)
.
 View larger version (10K): [in a new window] |
Figure 1. Aggravation of DSS-induced colitis due to T cell deficiency. TCR ko mice or wt control mice were exposed to 2.5% DSS in the drinking water for 7 days (n=7, each). Quantitative histopathological assessment of DSS-induced colitis shows significantly increased inflammation both in the cecum and colon. Shown are the histological scores (–) as box plots and the level of significance determined by Mann Whitney U test (P<0.05).
|
 View larger version (56K): [in a new window] |
Figure 2. Histopathology of increased DSS-induced colitis due to T cell deficiency. (A) Colon section (100x) of DSS-colitic wt mice: the colon appears normal with a low level of inflammatory cell infiltration and an intact epithelial layer; histological score = 1. (B) colon section (200x) of DSS treated TCR ko mice: highly inflamed colon with vascular fibrin thrombi (arrow), and loss of epithelial layer; histological score = 4. (C) SFOG stain of cecum section (200x): vascular fibrin thrombi are ruddled, erythrocytes appear yellow, and connective tissue is blue.
|
|
View this table: [in a new window] |
Table 1. T Cell Deficiency Enhances Interferon- Production in DSS-induced Colitis
|
T cells in IL-2 ko mice T cells, we asked whether these effects might be just due to epithelial injury mediated by DSS or whether these effects occur in immunological IBD animal models. Therefore, IL-2 ko mice were depleted of T cells (n=18) or received control IgG (n=13). The most striking effect of early T cell depletion (starting at 4 wk of age) was increased mortality (Fig. 3
). However, no statistically significant difference was found in mean histological colon scores (2.2±0.5 vs. 2.0±0.5, P=0.597). Still, early T cell depletion resulted in significantly increased IFN- secretion by LPL compared with controls (10.0±4.5 ng/mL vs. 0.6±0.05 ng/mL, P<0.05). Further, production of IL-10 was decreased in anti-CD3/CD28-stimulated splenocytes of T cell-depleted animals compared with controls (0.5±0.2 ng/mL vs. 1.4±0.2 ng/mL, P<0.05).
 View larger version (6K): [in a new window] |
Figure 3. Increased mortality of IL-2 ko mice upon early T cell depletion. IL-2 ko mice were treated i.p. with the T cell depleting mAb 13D5 (——, n=18) or hamster IgG (– – –, n=13). Antibody treatment started at 4 wk of age (initial concentration of 400 µg followed by weekly doses of 200 µg). Shown is the survival rate (%) and the level of significance using Kaplan Meier analysis.
|
T cell depletion than in controls (spleen weight: 806±80 mg vs. 525±84 mg, P<0.05; Fig. 4
). In contrast, splenocyte proliferation was decreased upon T cell depletion compared with control IgG-treated IL-2 ko mice (proliferation index: 1.1±0.3 vs. 12.7±6.7, P<0.05). In addition, T cell-depleted IL-2 ko mice showed a trend toward lower white blood cell counts compared with controls (4,800±2,400 per µL vs. 10,000±1,500 per µL, P=0.053).
 View larger version (41K): [in a new window] |
Figure 4. Increased splenomegaly in IL-2 ko mice upon early T cell depletion. IL-2 ko mice were depleted of T cells as described in Figure 3
. (A) Spleen of a T cell-depleted IL-2 ko mouse compared with the spleen of a hamster IgG-treated IL-2 ko mouse (B), and the spleen of a wild-type mouse (C).
|
T cells leads to increased inflammation (DSS-induced colitis) or increased mortality (IL-2 ko) in animal models representative of ulcerative colitis. To investigate the role of T cells in small bowel inflammation heterozygous TNF ARE mice were either depleted of T cells starting at 4 wk of age using a depleting mAb 13D5 or treated with control IgG. Mice were killed after 14 wk, and histological examination revealed more pronounced terminal ileitis in the T cell-depleted group (n=24) compared with control IgG treated mice (n=21), as the mean histological ileum score was increased (3.4±0.2 vs. 2.3±0.3, P<0.05). As in every other investigated IBD models IFN- secretion of LPL was increased, as previously reported [20
].
Because different mechanisms might be involved in the initiation and the perpetuation of intestinal inflammation, the influence of T cell depletion at a later time point, i.e., on already established colitis in IL-2 ko mice, was studied next. Therefore, depletion of T cells in IL-2 ko mice started at 8 wk of age 
2 wk after the first signs of colitis. IL-2 ko mice with established colitis were depleted of T cells (n=17) or received control IgG (n=16). Depletion of T cells in established colitis had no effect on survival (Fig. 5)
or the histological degree of colitis (mean histological colon scores: 2.1±0.3 vs. 1.6±0.4, P=0.167). However, late T cell depletion again led to increased spleen weight (760±60 mg vs. 540±31 mg, P<0.01). The proliferative response of splenocytes and white blood cell counts though was not statistically significant different (data not shown). Further, late T cell depletion resulted in decreased IL-10 production by both splenocytes and LPL upon anti-CD3/CD28 stimulation (Table 2
). No significant changes were found regarding the splenic IFN- production of IL-2 ko mice after late depletion though IFN- secretion of LPL was again increased (Table 2)
. Regarding white blood cell counts, there was no statistically significant difference between T cell-depleted IL-2 ko mice and controls (data not shown).
 View larger version (4K): [in a new window] |
Figure 5. No effect of T cell depletion on established colitis in IL-2 ko mice. Depletion of T cells (——, n=17) and hamster IgG treatment (– – –, n=18) was performed as described in Figure 3
. In IL-2 ko mice with established colitis, mAb treatment started at 8 wk of age. Shown is the survival rate (%) and the level of significance using Kaplan Meier analysis.
|
|
View this table: [in a new window] |
Table 2. Suppression of IL-10 Secretion upon T Cell Depletion in IL-2 ko Mice with Established Colitis
|
ko mice T cell depletion had no effect on DSS-induced colitis was increased upon T cell depletion in every investigated IBD model, the development and the course of DSS-induced colitis in IFN- ko mice were studied upon T cell depletion (n=13) and control IgG treatment (n=11). The depletion of T cells did not affect the development and course of DSS-induced colitis in IFN- ko mice, as there were no statistically significant differences with regard to weight course (weight loss: 24.9±1.5% vs. 24.1±1.6%, P=0.930) or mortality (mean survival time of T cell depleted mice: 9.15 days vs. controls 8.64 days, P=0.4192). The intestinal tissue damage was pronounced in both treatment groups, but there were no statistically significant differences regarding macroscopical or histological scores (Table 3
). Further, the colons were shortened upon DSS-induced colitis to a similar degree in both groups (4.5±0.4 vs. 4.8±0.3, P=0.360). Similarly, no statistically significant differences were detected in IL-10 and TGF-ß secretion by LPL or splenocytes upon CD3/CD28 stimulation (data not shown). As IFN- ko mice are deficient in IFN-, there was no IFN- secretion measurable. Finally, blood counts revealed no statistically significant differences regarding WBC, PLT, PBL, and PMN upon T cell depletion (data not shown). |
View this table: [in a new window] |
Table 3. No Effect on Macroscopical and Histological Scores upon T Cell Depletion in IFN- ko Mice with DSS-induced Colitis
|
The present study provides clear evidence that depletion of or deficiency in T cells enhances intestinal inflammation in different IBD animal models, particularly in the early phase. As a possible mechanism, T cells seem to control IFN- production, probably by ß T cells, during inflammation.
Increasing evidence suggests that intraepithelial T cells play an important role in healing processes of the epithelial barrier, as well as in mucosal defense toward enteric pathogens [2
, 4
, 7
, 8
, 26
27
28
29
30
]. To this end, T cell-deficient mice show impaired epithelial regeneration and reduced mucosal IgA production [27
, 31
]. In addition, exaggerated immunopathology was reported in T cell-deficient mice infected with Listeria monocytogenes, Klebsiella, Myobacteria tuberculosis, and Nocardia asteroides [4
, 26
, 30
, 32
, 33
]. Among these, nocardiosis (lung) led to severe mucosal pathology [4
, 30
]. Hence, T cells were shown to be required for an adequate inflammatory response by neutrophils in order to clear damaged epithelial cells. Similar findings were described for lung injury due to ozone [30
]. Our own group reported previously of increased mortality of rats with TNBS-induced colitis after T cell depletion, a type of colitis that also leads to marked epithelial injury [19
]. During the course of the present study, two groups reported similar findings in T cell-deficient mice exposed to DSS also in accordance with the data presented herein [7
, 8
]. Chen et al. showed increased tissue injury during and delayed tissue repair after DSS exposure. Further, they demonstrated that T cells not only produce KGF but that KGF ko mice are more susceptible to DSS suggesting that, at least in mice, T cells control intestinal homeostasis by producing KGF. The second group presented evidence that T cells control neutrophil infiltration in DSS-induced colitis leading to fatal disease in most T cell-deficient mice, possibly by the secretion of the chemokine KC, the murine homolog to interleukin 8. Regarding the clinical course, we found no differences between T cell-depleted and control mice with DSS-induced colitis. In both groups, rapid weight loss and rectal bleeding occurred to a similar extent. However, marked changes were seen on histology. This probably reflects the role of T cells in wound healing processes. Therefore, tissue injury and the resulting histological damage are not always reflected by clinical signs. Also in TNFARE mice, the clinical course was different from the pathology since no clinical signs were observed in spite of massive histological damage [20
]. It is tempting to speculate that in the long run difference in the microscopic degree of inflammation could have consequences on clinical signs of colitis, although this remains to be seen in future studies.
Additionally, T cells seem to control IFN- production, probably secreted by ß T cell. This hypothesis is based on the observation that significantly increased IFN- production was found in every IBD model examined. Against this hypothesis is evidence that late T cell depletion occurs in IL-2 ko mice though affecting IFN- production by LPL was not associated with either increased mortality or enhanced colitis. However, depletion of T cell in IFN- ko mice neither affected the development nor the course of DSS-induced colitis, suggesting that T cells control the secretion of IFN- by other lymphocytes, e.g., ß T cells. This control function is an important factor during the regulation of bowel inflammation. Szczepanik et al. previously reported that T cells are able to down-regulate IFN- production by targeted contact sensitivity-effector cells, i.e., T cells [34
]. The hypothesis that T cells control IFN- production by ß T cells seems to be contrary to reports on IFN- production by T cells in tumor immunity and infections [12
, 35
]. However, the deficiency of T cells always led to increased IFN- production in every IBD animal model investigated herein, representing both ulcerative colitis (DSS-induced colitis, IL-2 ko) and Crohn’s disease (TNFARE). Therefore, T cells seem to have the potential to do both: 1) mount a strong IFN- response toward enteric pathogens or tumor cells and, 2) down-regulate inflammatory responses by controlling IFN- production of probably ß T cells. At present, it is unknown how T cells differentiate so sharply between these quite different types of injury. On the other hand, in infectious disease models, T cells are able to differentiate between pathogens and mount either a Th1 response or a Th2 response [12
]. The regulatory role of T cells was previously observed in several other inflammatory disease models, such as uveitis, dermatitis, lupus, and orchitis [36
, 37
]. Further, CD8+ T cells mediated tolerance in a murine insulin-dependent diabetes model with increased IL-4 and particularly IL-10 secretion [11
]. Only a milder course of experimental autoimmune encephalomyelitis was reported in T cell-deficient mice [38
], which stands in contrast to depletion studies [39
]. The depletion studies presented herein therefore confirm our previous studies in rat TNBS-induced colitis and reports by others on colitis in T cell-deficient mice. To this end, the protective role of T cells is not restricted to chemically induced IBD models but can also be seen in immunological models, at least in the Crohn’s disease model (TNFARE) investigated here.
T cell depletion in IL-2 ko mice led to increased mortality after "early" but not "late" depletion; however it did not affect the degree of microscopic or macroscopic colitis. In addition, T cell depletion resulted in increased intestinal IFN- production, but there was no difference between "early depleted" or "late depleted" mice. It is therefore unclear whether the increased mortality in the "early" group had anything to do with the colitis. Importantly, IL-2 ko mice do not only develop colitis but also hemolytic anemia and splenomegaly [40
, 41
]. By 8 weeks of age 50% of these mice die due to this hematological illness. So it might be possible that the increased mortality of IL-2 ko mice upon "early" T cell depletion is rather caused by the severe hematologic disorder, exemplified by severe splenomegaly rather than colitis. Splenomegaly after T cell depletion is mediated by enhanced ß T cell proliferation as described previously by Kaufmann et al. using the anti- TCR mAb GL3 [42
]. Increased ß T cell proliferation was also described in naive IL-2 ko mice due to impaired Fas-mediated apoptosis [41
]. As a result of splenomegaly T cell-depleted IL-2 ko mice had lower leukocyte counts than the control group. It is therefore tempting to speculate that increased mortality is rather caused by more severe splenomegaly than by the colitis itself. Against this hypothesis stands that IL-2 ko mice, which were "late depleted" of T cells showed similar splenic weights, as "early depleted" mice. Taken together, the increased mortality after "early depletion" of T cells in IL-2 ko mice cannot be explained conclusively at present and has to be studied in future experiments.
Former reports suggest that T cells are involved in the induction of oral tolerance [11
, 43
, 44
]. In the present study, we found only minor effects on IL-10 production with decreased systemic and sometimes even increased LPL secretion after T cell depletion. However, increased secretion of IL-10 by LPL is a typical feature of DSS-induced colitis [45
, 46
] and was also reported in UC patients [47
]. Therefore, the production of IL-10 by T cells is unlikely to be the cause for the protective role of this T cell subset in IBD.
Recent reports and the data presented herein strongly support the notion that T cells have a protective role in IBD. However, others have reported differently. All of these groups used either TCR or TCRß ko mice [16
17
18
], which were recently shown to have markedly altered T cell functions [48
]. One can speculate that the proinflammatory role of T cells in these models has to do with these changes as increasing evidence now suggests a protective role in immunocompetent models. Future studies have to deal with the potential of T cells to actively suppress IBD.
In summary, the present data demonstrate that T cells have a protective role not only in "injury" colitis models (DSS-induced and TNBS-induced colitis) but also in immunological models (at least TNFARE, possibly IL-2 ko). Finally, some evidence suggests that T cells are able to control IFN- production by ß T cells in every investigated IBD models herein.
ACKNOWLEDGEMENTS
This work was supported by Deutsche Morbus Crohn/Colitis ulcerosa Vereinigung (DCCV e.V.), Leverkusen, Germany, by the German Ministry of Education and Research (BMBF/DLR) in the competence network on IBD, Core Facility Berlin, Germany, and by the Deutsche Forschungsgemeinschaft (DFG), Bonn, Germany (SFB633 B3 and Z1). We thank I. Horak, Berlin/Germany, and G. Kollias, Athens/Greece, for kindly providing IL-2-deficient and TNFARE mice, respectively, as well as Simone Spiekermann for her assistance in immunohistology.
Received November 29, 2005; revised July 26, 2006; accepted August 19, 2007.
REFERENCES
cells: a right time and a right place for a conserved third way of protection Annu. Rev. Immunol. 18,975-1026[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
