| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Receptor-Ig Prevents Th2 Cell-Type Colitis1
,
,¶
*
Department of Gastroenterology, Research Institute, International Medical Center of Japan, Tokyo, Japan;
Immunobiology Vaccine Center, Departments of Microbiology and
Oral Biology, University of Alabama, Birmingham, AL 35294;
Biogen, Inc., Cambridge, MA 02142; and
¶ Department of Mucosal Immunology, Research Institute for Microbial Diseases, Osaka University, Osaka, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
receptor-Ig fusion protein (LTR-Ig) in utero,
a treatment that blocked the formation of both Peyer’s and colonic
patches. Mice deficient in colonic patches developed focal acute ulcers
with Th1-type responses, whereas lesions in normal mice were of a
diffuse mucosal type with both Th1- and Th2-type cytokine production.
We next determined whether LTR-Ig could be used to treat colitis in
normal or Th2-dominant, IFN-
gene knockout
(IFN--/-) mice. Four weekly treatments with LTR-Ig
resulted in deletion of Peyer’s and colonic patches with significant
decreases in numbers of dendritic cells. This pretreatment protected
IFN--/- mice from trinitrobenzene sulfonic
acid-induced colitis; however, in normal mice this weekly treatment was
less protective. In these mice hypertrophy of colonic patches was seen
after induction of colitis. We conclude that Th2-type colitis is
dependent upon the presence of colonic patches. The effect of LTR-Ig
was mediated through prevention of colonic patch hypertrophy in the
absence of IFN-. Thus, LTR-Ig may offer a possible treatment for
the Th2-dominant form of colitis. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(Th1-type), and in situ analysis of
lesions revealed increased IFN- production (2).
Treatment of mice with anti-IL-12 mAb markedly decreased the
severity of TNBS colitis (2, 5). We previously used this
model to show that mice that develop either predominant
CD4+ Th1- or Th2-type responses exhibit distinct
inflammatory lesions: Th1-dominant IL-4 gene knockout
(IL-4-/-) mice developed focal penetrating
ulcers, whereas colitis induced in Th2-dominant
IFN--/- mice was characterized by diffuse,
atrophic changes in crypts accompanied by marked hypertrophy of colonic
patches (6, 7). These characteristics of Th2-type
hapten-induced colitis share the feature of ulcerative colitis in
humans, which is considered to be Th2-mediated enteric inflammation
(8, 9). Th2-type TNBS colitis was also accompanied by
marked hypertrophy of colonic patches containing M cells and a
follicle-associated epithelium covering the dome region. In addition,
TNBS-specific Th2-type cytokine responses occurred in the colonic
patches themselves (6).
Because there exists an association between colonic patch development
and Th2-type responses, it is important to assess the role of colonic
patches in colitis. It has been shown that lymphotoxin (LT) signaling
is required for the genesis of Peyer’s patches and lymph nodes
(10). For example, LT
gene-deleted
(LT-/-) mice are deficient in both Peyer’s
patches and lymph nodes (11). Furthermore, treatment of
LT-/- mice with an agonist mAb to
LTR-induced lymph node development (12), whereas
knockout of the LTR gene again resulted in failure of mice to
develop Peyer’s patches and lymph nodes (13).
Interestingly, treatment with soluble LTR-Ig fusion protein
(LTR-Ig) in utero resulted in deletion of Peyer’s patches and
peripheral lymph nodes; however, mesenteric and sacral lymph nodes
(SLN), which drain the small and large intestines, remained intact
(14, 15). Although the precise cellular and molecular
mechanisms for the genesis of gut-associated lymphoreticular tissues
(GALT) is not yet known, it is clear that membrane-associated
LT12 on activated lymphocytes and LTR expressed by macrophages
and dendritic cells partially account for the development of this
mucosal inductive site (16).
Blockade of LTR in adult mice with LTR-Ig prevents germinal
center formation in spleen and results in impaired IgG anti-sheep
RBC Ab responses (17). Membrane LT12 expression by B
cells and that of LTR by stromal cells play crucial roles in the
formation of follicular dendritic cell networks (10, 18, 19, 20, 21). These studies have suggested that LT12/LTR
interactions are necessary not only for ontogenesis of lymphoid tissue,
but also for maintaining the normal architecture required for immune
responses. The LT system is critical for the generation of chemokine
gradients that are essential for cellular positioning (22)
and for recruitment of NK and dendritic cells (23, 24).
Likewise, disruption of the surface LT or LIGHT (another ligand for
LTR)-LTR axis affects the generation of
CD8+ T cells in
anti-lymphochoriomeningitis virus immune responses (25, 26). Blocking of the LT/LIGHT axis with LTR-Ig was
shown previously to be effective in preventing colitis in the CD3
and CD45RBhigh T cell transfer models of colitis
(27).
In this study we have generated Peyer’s patch-deficient mice by
treatment with LTR-Ig fusion in utero to determine the effects of
this treatment on the genesis of colonic patches as well as on the
development of TNBS colitis. We also tested the idea that colonic
patches are associated with Th2-type responses in three different
aspects of murine colitis: 1) in mice deficient in Peyer’s patches, 2)
in adult mice treated with LTR-Ig only, and 3) in adult mice treated
with LTR-Ig with predominant Th2-type responses.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Fusion proteins were from the extracellular domain of either
murine LTR or human TNFR p55 fused to the hinge,
CH2, and CH3 domains of
human IgG1 (LTR-Ig or TNFR p55-Ig, respectively) and were prepared
as described previously (14, 28, 29).
Generation of Peyer’s patch-deficient mice
Timed pregnant BALB/c mice were purchased from The Jackson
Laboratory (Bar Harbor, ME) and were maintained in sterile cages under
pathogen-free conditions at Biogen (Cambridge, MA) and under specific
pathogen-free conditions in the immunocompromised mouse facility at the
Immunobiology Vaccine Center, University of Alabama (Birmingham, AL).
Mice were provided sterile food and water ad libitum and were free of
microbial pathogens as determined by Ab screening and routine
histologic analysis of organs and tissues. Because it has been shown
that the genesis of lymph nodes and Peyer’s patches can be blocked in
a sequential manner by varying the gestation day of LTR-Ig treatment
(14, 15), we generated mice that had disrupted
Peyer’s patches but that retained mesenteric and SLN for subsequent
studies of colonic patch formation and TNBS colitis development. To do
this, pregnant mice were injected i.v. with 200 µg LTR-Ig on
gestational days 14 and 17 as described previously (14, 15). Mice born in the same facility and of the same age as those
in the experimental group were used as controls.
Treatment of adult mice with fusion proteins and mAbs
IFN- gene-deficient (IFN--/-) mice
and background/age-matched, normal BALB/c mice were purchased from The
Jackson Laboratory and kept in our animal facility at the Immunobiology
Vaccine Center or at the Research Institute, International Medical
Center of Japan (Tokyo, Japan), under specific pathogen-free conditions
as described above. Adult mice (8 wk of age) were given 200 µg
LTR-Ig, TNFR p55-Ig, or control normal human IgG (control IgG) i.p.
Young adult mice (8 wk of age) were given these fusion proteins on days
-14, -7, 0 (the day of induction of colitis), and 7. In some
experiments groups of mice were given 1 mg rat anti-IFN- mAb
(clone XMG 1.2) by the i.p. route on days 0 and 7. Control groups of
mice received i.p. doses of normal rat IgG (Jackson ImmunoResearch
Laboratories, West Grove, PA).
Induction of TNBS colitis
Mice were given rectally a solution of TNBS (Research Organics, Cleveland, OH) dissolved in a mixture of PBS (pH 7.2) and then mixed with an equal volume of ethanol at a final concentration of 2% TNBS in 50% ethanol. After being anesthetized with ketamine/xylazine, mice were given intrarectally 36 µg TNBS/g body weight on days 0 and 7 using a glass microsyringe equipped with a gastric intubation needle. Tissues and cells were assessed 3 days later (day 10).
Myeloperoxidase (MPO) assay
MPO activity in murine colonic tissues was determined by a method reported previously (30). Briefly, colonic tissues were homogenized in 5 mM phosphate buffer (pH 6.0) and centrifuged at 30,000 x g for 30 min at 4°C. The pellet was suspended in 50 mM phosphate buffer containing 0.5% hexadecyltrimethyl ammonium bromide. After centrifugation at 20,000 x g, the supernatant was subjected to MPO assay in a spectrophotometer with a rate assay system (U3200; Hitachi, Tokyo, Japan).
Histologic analysis and immunohistochemistry
The colon was removed from its mesentery to the pelvic brim by blunt dissection, opened longitudinally, and fixed in 5% glacial acetic acid in ethanol (v/v). After embedding in paraffin, 4-µm serial sections were prepared and stained with hematoxylin and eosin for histologic grading. Some specimens were snap-frozen, and 4-µm frozen sections were prepared. Histologic grading was performed on coded slides without knowledge of mouse group and according to the criteria described previously (6). For immunohistochemistry, frozen sections were fixed with cold acetone and incubated with 10% normal rabbit serum for blocking. Sections were then incubated with anti-mouse MAdCAM-1 mAb (x100; BD PharMingen, San Diego, CA), biotin-anti-mouse CD11c (x100, BD PharMingen), biotin-peanut agglutinin (biotin-PNA; x100, Vector Laboratories, Burlingame, CA), or FDC-M1 mAb (20 µg/ml; provided by Dr. K. Maeda, Yamagata University, Yamagata, Japan) for 1 h. Bound anti-MAdCAM-1 mAb or mAb FDC-M1 was detected with FITC-labeled anti-rat IgG (Southern Biotechnology Associates, Birmingham, AL). CD11c and PNA stainings were detected with FITC-streptavidin. For T and B cell analysis, each section was stained with FITC-anti-B220 and biotin-anti-CD3 mAb (BD PharMingen), followed by incubation with tetramethyl rhodamine-streptavidin (Southern Biotechnology Associates). Sections were examined with a fluorescence microscope (BX50/BXFLA, Olympus, Tokyo, Japan) equipped with a CCD camera and an image analyzer (ATTO densitograph; ATTO, Tokyo, Japan). Combination images of T and B cell staining were analyzed using identical settings in Photoshop 4.0 (Adobe Systems, San Jose, CA).
Isolation of lymphoid cells
Peyer’s and colonic patches were excised from the intestinal wall and washed once with RPMI 1640 (Cellgro; Mediatech, Washington, D.C.), and single cells were dissociated with collagenase (type V, Sigma, St. Louis., MO) at a concentration of 0.5 mg/ml in RPMI 1640 with 100 U/ml penicillin, 100 µg/ml streptomycin, and 40 µg/ml gentamicin for 20 min at 37°C (31). The cell dissociation step was performed twice more using fresh collagenase solution each time. Colonic lamina propria lymphocytes were prepared as described previously (32), with modification. Briefly, after excision of all visible lymphoid follicles, the colonic tissue was treated with 1 mM EDTA in PBS for 20 min to remove the epithelium. The tissue was then digested with collagenase (type V; Sigma) for 20 min, and this step was repeated one more time. The single-cell suspensions were then pooled and washed with RPMI 1640 two additional times. Mononuclear cells were further purified using a discontinuous Percoll gradient to avoid contamination with epithelial cells (32). The SLN were teased apart with forceps, and the resulting cell suspensions were washed twice more in RPMI 1640.
Culture conditions and proliferation assays
Cells from SLN were conjugated with TNBS by treatment of 1
x 107 cells with 1 ml of 0.3 mg/ml TNBS in RPMI
1640 at room temperature for 15 min. Excess TNBS was removed by washing
with RPMI 1640 and centrifugation of conjugated cells. For Ag-specific
stimulation, conjugated cells were cultured in RPMI 1640 supplemented
with 10% FCS, sodium pyruvate, L-glutamine, HEPES buffer,
50 µM 2-ME, 100 U/ml penicillin, 100 µg/ml streptomycin, 40 µg/ml
gentamicin, and 1 µg/ml amphotericin B (complete medium) at 37°C in
an atmosphere of 5% CO2 in air. For the
proliferation assay (6), 2 x
105 TNBS-conjugated cells were added to wells of
96-well plates and cultured for 3 days, and 0.5 µCi/well
[3H]TdR was added 18 h before cell
harvesting. The levels of [3H]TdR incorporation
was determined by scintillation counting. Cells cultured without
treatment with TNBS were used as controls. In some experiments
TNBS-treated cells were cultured in complete medium for 24 h at a
concentration of 3 x 106 cells/ml. Culture
supernatants were obtained and then assessed for cytokine levels as
described below. Cells were stained with FITC-anti-CD4 mAb (BD
PharMingen) and subjected to sorting with a
FACStarPlus (BD Biosciences, Mountain View, CA)
to obtain purified CD4+ T cells (99%
CD4+ T cells). Cytokine-specific RT-PCR was
performed as described previously (33). In some
experiments quantitative RT-PCR for mRNA expression of IFN- and IL-4
was performed using ABI PRISM 7700 with Taqman probes (PE Applied
Biosystems, Foster City, CA). The PCR product for each cytokine was
used as a standard for template DNA.
Cytokine assays
Culture supernatants were subjected to cytokine-specific ELISA
as described previously (32, 34, 35). In brief, microtiter
plates were coated with mAbs to individual cytokines and incubated
overnight at 4°C. After blocking with 3% BSA in PBS at 37°C for
2 h, diluted samples were added to wells and incubated overnight
at 4°C. The wells were then washed and incubated with detection mAbs,
and bound Ab was detected using a peroxidase-labeled anti-biotin
mAb (Vector Laboratories). The substrate used for peroxidase activity
was 3,3',5,5'-tetramethylbenzidine. The following anti-cytokine
mAbs were used for coating or in the biotinylated form for detection,
respectively, in this ELISA: anti-IFN-, R4-6A2 and XMG 1.2 mAbs;
anti-IL-2, JES6-1A12 and JES6-5H4 mAbs; anti-IL-4, BVD4-1D11
and BVD6-24G2 mAbs; anti-IL-5, TRFK-5 and TRFK-4 mAbs;
anti-IL-6, MP5-20F3 and MP5-32C11 mAbs; and anti-IL-10,
JES5-2A5 and JES5-16E3 mAbs. The ELISAs were capable of detecting 0.4
ng/ml IFN-, 0.05 U/ml IL-2, 3.0 pg/ml IL-4, 0.2 U/ml IL-5, 0.2 ng/ml
IL-6, and 0.04 ng/ml IL-10. The levels of cytokine production by
Ag-stimulated T cells were calculated by subtracting the results of
control cultures that were not stimulated with TNBS.
Statistics
The data are expressed as the average ± 1 SD and were compared using the Mann-Whitney U test. The results were analyzed using the StatView-J 4.4 statistical program (Abacus Concepts, Berkeley, CA) for Macintosh computers and were considered statistically significant if p < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
R-Ig
(generation of colonic patch-deficient mice)
In the initial experiments mice were treated in utero with
LTR-Ig to determine whether loss of Peyer’s patches could be
associated with a failure to form colonic patches. No visible Peyer’s
patches were present in these mice; however, mesenteric and SLN
developed normally as reported previously (14, 36). The
mice treated in utero with LTR-Ig did not have visible colonic
patches. To verify this, histologic sections of small and large
intestines were assessed for the presence of follicles. No
organized structures were seen in either small or large intestine, and
only scattered aggregates of B cells were noted in the mucosal
layer. These B cell aggregates were also observed in untreated normal
BALB/c mice (data not shown).
TNBS colitis in mice treated in utero with LTR-Ig
We next induced TNBS colitis in 10-wk-old control and
LTR-Ig-treated colonic patch-deficient mice by administering two
doses TNBS-ethanol 1 wk apart. The histologic features of TNBS colitis
observed on day 10 in untreated mice and colonic patch-deficient mice
were remarkably different (Fig. 1
). As
noted previously (6), the inflammation induced by TNBS in
normal BALB/c mice demonstrated crypt distortion, loss of goblet cells,
and mononuclear cell infiltration in the mucosal layer with enlarged
colonic patches (Fig. 1, A and C) and occasional
ulcers. In contrast, colonic patch-deficient mice also developed
colonic inflammation, which was clearly different from the control
group. Open ulcers were found in all mice, whereas crypt lesions were
less frequent (Fig. 1, B and D). Crypts were more
hypertrophic and elongated, resulting in a thickened colonic wall.
Transmural cell infiltration was seen with a predominant infiltration
of polymorphonuclear leukocytes, even in the submucosa under the intact
lamina muscularis mucosae (Fig. 1, B and D). An
infiltration of polymorphonuclear leukocytes in colonic patch-deficient
mice was verified by the finding of significant levels of MPO activity
(Fig. 2A). This MPO activity
was increased in the colon of mice after TNBS enema and was higher in
colonic patch-deficient than in normal mice. No colonic patches were
seen after induction of TNBS colitis. To control for vehicle-only
effects, colonic patch-deficient mice were also given a 50% ethanol
enema, This treatment resulted in only minimal superficial erosion and
was identical with that induced in normal mice. We conclude that both
mouse groups developed inflammation, but with distinct characteristics.
Mice treated with LTR-Ig exhibited higher scores for acute
ulcer-related lesions and lower scores for chronic-type inflammation
and crypt changes (Fig. 2B).
|
|
The SLN of mice with TNBS colitis were enlarged and contained
lymphoid cells undergoing proliferation. In a previous study
(6) we extensively assessed TNBS-induced cytokine
responses in lamina propria, colonic patches, and SLN and showed that
the cytokine responses in the SLN accurately reflect the
immunopathologic responses in colonic patches and lamina propria. Based
on this analysis, we assessed cells from SLN of colonic patch-deficient
mice for TNBS-specific proliferative and cytokine responses. Cultures
of SLN cells exhibited brisk proliferative responses to
TNBS-conjugated-self Ag (Fig. 2
A). Cytokine production,
which accompanied the TNBS-induced proliferative responses, was
measured 24 h after stimulation (Fig. 3A). Following restimulation
with TNBS, colonic patch-deficient mice did not produce IL-4, whereas
control mice produced IFN-, IL-4, and IL-5. To confirm this,
purified CD4+ T cells from cultures were
subjected to cytokine-specific RT-PCR. The expression of mRNA in
TNBS-specific CD4+ T cells was similar to the
results obtained for secreted proteins (data not shown). Thus, in
colonic patch-deficient mice, mucosal Th2-type responses were
suppressed. Furthermore, to confirm this, lamina propria mononuclear
cells were isolated from the colon. The total RNA was extracted, and
cytokine-specific mRNA was quantified by real-time RT-PCR. The number
of copies of mRNA for IL-4 was significantly lower in colonic
patch-deficient mice. We conclude from these results that diffuse
crypt-type inflammation is accompanied by Th2-type responses and is
dependent upon the presence of colonic patches. In contrast, focal
ulcer-type lesions occur independently of colonic patches.
|
mAb
The results to this point suggest that inflammation in colonic
patch-deficient mice is dependent upon IFN-, but not IL-4. We tested
this assumption by treatment of TNBS colitis with anti-IFN- mAb.
Treatment of murine TNBS colitis with anti-IFN- mAb was
ineffective, and IFN--/- mice developed
colitis, which was as severe as that in wild-type (WT) mice (6, 7). However, in colonic patch-deficient mice, blockade of
IFN- markedly reduced the severity of inflammation (Fig. 2). We
conclude that the inflammatory changes in colonic patch-deficient mice
are largely mediated by IFN-, and that Th2-type inflammatory
responses appear to be dependent upon the presence of colonic
patches.
LTR-Ig treatment of adult mice decreases the cellularity of both
Peyer’s and colonic patches
Our past studies in Th2-type TNBS-induced colitis revealed
atrophic changes in the mucosal epithelium, with distorted crypts and
fibrosis (6, 7). We proposed that this could be a model
for the chronic stage of IBD, especially in ulcerative colitis. In this
study we used colonic patch-deficient mice to provide further evidence
that Th2-type inflammatory responses require the presence of colonic
patches. Our model allowed us to treat Th2-type TNBS colitis by
abrogating functional colonic patches. For this purpose we took
advantage of the fact that signals through the LTR are also
important in the maintenance of secondary lymphoid tissue in adult mice
(10, 18, 19, 20, 21). Thus, we assessed the effects of blocking
LTR signals by LTR-Ig on TNBS colitis induced in adult WT and
IFN--/- mice (a model for Th2-dominant
colitis). Because the beneficial effect of anti-TNF- mAb in
experimental colitis (37, 38) and in treatment of Crohn’s
disease has been documented (39, 40), TNFR p55-Ig was also
tested for comparison. In our initial study we tested the effect of
weekly i.p. administration of LTR-Ig on Peyer’s and colonic patches
in adult mice without colitis. We found that weekly doses of LTR-Ig
induced a flattened appearance in Peyer’s patches (Fig. 4A). Approximately three or
four doses were required for maximum effects on Peyer’s patches, and
this treatment also reduced by one-half the total number of
macroscopically visible Peyer’s patches in LTR-Ig-treated adult
mice. When treatment was discontinued, the Peyer’s patches recovered
cellularity and increased in size (Fig. 4A, right
panel). Interestingly, the cecal patches did not undergo
visible changes in size or shape following administration of LTR-Ig
(data not shown). Colonic patches, which are naturally flat, were
assessed by histologic analysis. Two consecutive treatments with
LTR-Ig resulted in diminished colonic patches with smaller germinal
centers, which stained less intensely. The T and B cell zones were
essentially intact (Fig. 4B, bottom row). After
four treatments, colonic patches were further reduced in size, with
only vestiges remaining, and were detected only in histologic sections
(Fig. 4B). In untreated mice, CD11c-positive dendritic cells
were present in the subepithelial dome of colonic patches and also
scattered in the follicles (Fig. 4B, middle row). Two weekly
doses of LTR-Ig markedly eliminated CD11c-positive dendritic cells
from both areas of colonic patches (Fig. 4B). Similar
effects were also observed in IFN--/- mice
treated with LTR-Ig (data not shown). These effects of LTR-Ig
suggested the possibility that this treatment abrogates the function of
colonic patches by eliminating cells, including dendritic cells, and
reducing cell-to-cell interactions in the colonic patches.
|
R-Ig- and TNFR p55-Ig-treated
mice
In the next study we induced TNBS colitis in mice pretreated with
LTR-Ig. Adult mice were treated with either LTR-Ig or TNFR p55-Ig
and were then given TNBS enemas following the third dose of fusion
proteins. A second dose of TNBS was given at the same time as fusion
proteins on day 7. These treated mice were then subsequently assessed
for colitis and cellularity of patches on day 10 (Fig. 5). Pretreatment of both WT and
IFN--/- mice with two doses of LTR-Ig
markedly decreased the cellularity of Peyer’s and colonic patches (day
0, Fig. 5). In contrast, TNFR p55-Ig treatment did not affect the
number of cells in either Peyer’s or colonic patches. After TNBS
enema, both WT and IFN--/- mice treated with
control IgG had >2-fold more colonic patch cells (day 10, Fig. 5B). In contrast, LTR-Ig-treated,
IFN--/- mice exhibited markedly reduced
numbers of colonic patch cells (Fig. 5B). In
LTR-Ig-treated WT mice, most mice showed colonic patch hypertrophy
after induction of colitis, and cell yields increased to levels
comparable to those seen in mice treated with control IgG (day 10, Fig. 5B). Low numbers of cells were seen in Peyer’s patches of
LTR-Ig-treated WT and IFN--/- mice even
after induction of TNBS colitis. In mice treated with TNFR p55-Ig, the
number of colonic patch cells increased after TNBS treatment, even
though it remained lower than that in WT or
IFN--/- mice treated with control-Ig.
|
In WT mice, LTR-Ig treatment decreased the disease severity
only slightly; however, this was not significantly different from
control mice (Fig. 6, left
panel). LTR-Ig treatment eliminated colonic patches more
efficiently in IFN--/- mice, and this group
had the lowest histologic scores (Fig. 6, right panel).
Treatment of normal mice with TNFR p55-Ig did not affect disease. To
examine the effect of LTR-Ig in a Th2-type environment, we next
treated WT adult mice with both LTR-Ig and anti-IFN- mAb. As
noted previously (6), treatment with anti-IFN- mAb
alone did not protect mice from TNBS colitis; however, coadministration
of LTR-Ig with anti-IFN- mAb resulted in significant
protection (Fig. 6, middle panel).
|
-/- mice. The most dramatic histologic
changes induced by LTR-Ig treatment occurred in
IFN--/- mice, where the severity of colonic
inflammation was significantly reduced (Figs. 6 and 7, A and C).
However, in WT mice, the same treatment was less effective (Figs. 6 and 7, B and D). To identify vestiges of colonic
patches, serial sections of colonic tissues were prepared. In
LTR-Ig-treated IFN--/- mice, the vestiges
of colonic patches were of reduced thickness, and the lymphoid cells
were replaced by epithelial cells (Fig. 7C). In contrast,
colonic patches from LTR-Ig-treated WT mice were of normal size
(Fig. 7D). We also examined the architecture of colonic
patch vestiges on day 10 by immunohistologic analysis. Assessment of T
and B cell subsets in colonic patches of LTR-Ig-treated WT and
IFN--/- mice did not show significant
changes from the group treated with control IgG (data not shown).
Mucosal homing receptor MAdCAM-1+ venules were
present in vestiges of colonic patches of
IFN--/- mice and in the enlarged colonic
patches of LTR-Ig or control IgG-treated WT mice (Fig. 8). Germinal centers were absent in
LTR-Ig-treated WT and IFN--/- mice (Fig. 8). Dendritic cells were absent in LTR-Ig-treated
IFN--/- mice after induction of colitis,
whereas LTR-Ig-treated WT mice showed scattered
CD11c+ cells, although they did not form clusters
under an epithelial cell layer as were seen in control IgG-treated mice
with TNBS colitis (Fig. 8). Follicular dendritic cells were abundant in
B cell zones in the enlarged colonic patches of both WT and
IFN--/- mice treated with control IgG. In
IFN--/- mice treated with LTR-Ig,
follicular dendritic cells had almost completely disappeared from
vestiges of colonic patches.
|
|
TNBS-specific responses were assessed in SLN cultures. No T cell
proliferative responses were seen in IFN--/-
mice treated with LTR-Ig or TNFR p55-Ig, whereas all groups of WT
mice exhibited TNBS-specific T cell proliferative responses (Fig. 9A). Cytokine levels followed
the degree of proliferative responses induced in culture. In
IFN--/- mice treated with LTR-Ig or TNFR
p55-Ig, a reduction in Th2-type cytokine synthesis was noted (Fig. 9B). The SLN cells from WT mice treated with LTR-Ig,
TNFRp55-Ig, or control IgG all produced IFN-, IL-2, IL-4, and IL-5
(Fig. 9B). Reduction of these cytokines by
CD4+ T cells in LTR-Ig-treated
IFN--/- mice was confirmed by
cytokine-specific RT-PCR (Fig. 9C, lane 4). Thus,
although both LTR-Ig and TNFRp55-Ig tended to protect
IFN--/- mice from TNBS-induced colitis, only
LTR-Ig treatment resulted in elimination of colonic patches and
prevention of colitis.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
R-Ig in
utero, a treatment known to eliminate Peyer’s patches
(15), and we now report that this eliminates colonic
patches as well. Cells taken from the SLN of these treated mice also
lacked TNBS-induced Th2-type responses, which confirmed and extended
our previous results, suggesting that colonic patches supported
CD4+ Th2-type immune responses in TNBS colitis
(6). The diffuse mucosal type of inflammation is
characteristic of colitis observed in the Th2 cytokine-dominant state,
and this form occurs in both IFN--/- and
IL-12-/- mice. In contrast, a focal penetrating
lesion is associated with up-regulation of the Th1-type cytokines,
IFN- and TNF- (7). These findings point to a
significant role for colonic patches in the maintenance of diffuse
mucosal lesions through Th2-type immune responses. Alterations in
immune responses to luminal Ags in colonic patch-deficient mice were
rather modest. Significant IgA Ab responses to orally administered Ag
were seen in mice treated with LTR-Ig in utero, and Ag-specific Th
cell responses shifted to mesenteric lymph nodes (36). In
the present study trinitrophenyl-specific serum IgG and fecal IgG and
IgA Ab responses after induction of colitis in colonic patch-deficient
mice were comparable to those in normal mice (data not shown). These Ab
responses to mucosal Ags appear to differ from local inflammatory
responses in terms of dependence upon the presence of GALT. Our results
suggest that TNBS colitis, which is mediated by Th2-type responses, is
directly dependent upon immunological events in colonic patches
themselves. Furthermore, a lack of colonic patches was associated with
loss of Th2-type responses and was not replaced by the presence of
mesenteric or SLN.
Based upon these results, we assumed that abrogation of immune function
of colonic patches would ameliorate TNBS colitis in a Th2-dominant
state. We again used LTR-Ig for this purpose. In the initial study
we found that pretreatment of adult mice with LTR-Ig resulted in
loss of cellularity in both Peyer’s and colonic patches with disrupted
follicular dendritic cell networks, and this response was most dramatic
in IFN--/- mice. The precise mechanism for
this observed effect is not yet known. Down-regulation of MAdCAM-1 was
not seen after treatment with LTR-Ig, even in flattened colonic and
Peyer’s patches. Blocking of the LTR signaling pathway in normal
mice inhibited the formation of germinal centers in colonic patches,
but these mice still exhibited colonic patch hypertrophy after
induction of TNBS colitis. In contrast, colonic patches in
IFN--/- mice were diminished even after
induction of colitis and were accompanied by a lack of dendritic cells
in both subepithelial dome and follicles in the vestiges of lymphoid
follicles. Previous studies have shown that membrane-bound LT is
essential for maintaining the architecture of follicular dendritic cell
networks (11, 18, 19, 20, 21), and the use of mAb to block
LT in adult mice causes decreased expression of chemokines such
as B lymphocyte chemoattractant and secondary lymphoid tissue
chemokine, a potent T cell attractant expressed by follicular stromal
cells (22). Decreased numbers of dendritic cells in
Peyer’s and colonic patches would result in decreased production of
these chemokines, with subsequent lack of recruitment of mucosal
lymphocytes. In our study the protective effect of LTR-Ig treatment
was more obvious in the absence of IFN- and was closely associated
with the loss of colonic patches. In contrast, in the presence of
IFN- (normal mice) the colonic patches were initially eliminated
efficiently by pretreatment with LTR-Ig; however, hypertrophy of
patches was reinduced by TNBS enema treatment. It was evident that
local inflammation induced the redevelopment of colonic patches in
LTR-Ig-treated mice, because Peyer’s patches in the small intestine
were absent in mice with TNBS colitis, whereas colonic patches became
hypertrophic. Because dendritic cells reappeared in colonic patches
after induction of colitis in LTR-Ig-treated WT mice, this clearly
suggests the strong relation of colonic patch hypertrophy and
recruitment of dendritic cells into colonic patches.
Our studies have clearly shown that the LT12/LTR signaling
pathway maintains normal dendritic cell numbers and germinal centers in
both Peyer’s and colonic patches. Our results further suggest the
presence of inflammation-specific pathways for recruiting lymphocytes
into colonic patches that are independent of the LT12/LTR
pathway. In this regard, once tissue damage occurs, the production of
IFN- and other inflammatory cytokines synergistically up-regulates
other cytokines, inflammatory adhesion molecules, and chemokines, which
may then accelerate the homing of lymphocytes into colonic patches to
overcome the effects of LTR-Ig treatment. In
IFN--/- mice, the effect of TNFRp55-Ig
treatment was intermediate between those of control IgG and LTR-Ig
treatments in preventing colitis and colonic patch hypertrophy. Because
TNFR p55-Ig had no effect on the cellularity of colonic patches without
colitis, this suggests that TNFR p55-Ig blocked only an
inflammation-specific pathway. Thus, signaling via LTR may play a
significant role in the constitutive maintenance of cellularity of
colonic patches in WT mice, and additional inflammatory responses may
cause colonic patch hypertrophy. In contrast, in
IFN--/- mice colonic patch hypertrophy and
inflammation itself were dependent on the LT12/LTR signaling
pathway as well as maintenance of both Peyer’s and colonic patches in
the uninflamed gut.
In contrast to the limited protective effect of LTR-Ig in WT mice in
our study, previous reports of experimental colitis following T cell
reconstitution of SCID mice demonstrated the efficacy of both LTR-Ig
and anti-TNF- mAb treatment in preventing colonic inflammation
(27). This latter study suggested that down-regulation of
adhesion molecules by LTR-Ig was one of the mechanisms for the
anti-inflammatory effects. In this T cell-adoptive transfer system,
the induction of colitis primarily depended upon the homing of
lymphocytes into the intestine. Because the tissue destruction occurred
after infiltration and expansion of adoptively transferred T cells into
the colon, the terminal inflammatory responses may be controlled by
LTR-Ig in this experimental system by preventing the initial
infiltration of lymphocytes into uninflamed mucosa. In contrast, in the
case of TNBS colitis, rectal administration of TNBS with ethanol
initially breaks the intestinal barrier, and haptenated Ags then enter
the lamina propria. Therefore, this influx would trigger an initial
inflammatory response, including the production of TNF- and IFN-.
Indeed, these Th1-type cytokines are up-regulated in early phases of
TNBS colitis, and this is followed by Th2-type cytokine production and
hypertrophy of colonic patches occurring at a later stage. In
IFN--/- and IL-12-/-
mice, the early burst of IFN- is absent, but these mice do develop a
diffuse, mucosal type of colitis (7). This initial
up-regulation of proinflammatory cytokines probably overcomes the
effect of LTR-Ig in WT mice with TNBS colitis. LTR-Ig was
effective for treatment of TNBS colitis in
IFN--/- mice, most likely due to the lack of
an early Th1 inflammatory response. Thus, LTR-Ig efficiently
prevented the Th2-dominant type of inflammation in
IFN--/- mice by disturbing the function of
colonic patches. Indeed, when WT mice were treated with both
anti-IFN- mAb and LTR-Ig, a significant reduction in disease
was seen. Further investigation is required to clarify the precise role
of IFN- in recruiting dendritic cells into colonic patches; however,
our current findings in IFN--/- mice are of
significance in terms of a model for the chronic type of colitis.
Although LTR-Ig provided limited benefits to WT mice, it does offer
a new modality for treatment of human IBD, a new weapon in the current
arsenal of proposed treatment that includes the neutralization of
TNF- in humans (39, 40, 41) or IL-12 in the mouse model
(2, 5). It is possible that disturbance of immunological
events in GALT ameliorates inflammation mediated by Th2-type cytokine
responses without induction of high levels of IFN- or TNF-.
Ulcerative colitis is often suggested to consist of this type of
inflammation (8, 9). Colonic patches are also very likely
to be involved in the pathogenesis of IBD. Lymphoid follicles with an M
cell-containing follicle-associated epithelium have been reported to be
abundant in the human colon, although they are smaller in size than
Peyer’s patches of the small intestine (42, 43). These
findings suggest that colonic patches are a functional type of GALT for
Ag uptake and presentation and for T and B cell interactions.
Furthermore, evidence from animal models (44) and human
studies (45, 46, 47) has shown that the indigenous flora is
also important in the pathogenesis of IBD. Because the colon is the
major colonization site for commensal bacteria, colonic GALT are likely
to have specific functions in the maintenance of the homeostasis of
mucosal immunity in the presence of bacterial Ags. In contrast to
Crohn’s disease, which develops numerous secondary lymphoid follicles
in all layers of the small and large intestinal walls, ulcerative
colitis does not cause increased numbers of lymphoid follicles, but
instead leads to an enlargement of colonic patches that line the
mucosal layer (our unpublished observations). This suggests a primary
role for colonic patches in the initiation of inflammation in
ulcerative colitis. It is plausible that suppression of abnormal
immunological events in these follicles by blocking the
LT12/LTR signaling pathway, which may occur constitutionally
in IBD patients, has beneficial effects in treating this disease.
In conclusion, our results have shown that anti-inflammatory
effects of LTR-Ig are the result of a mechanism distinct from that
involved in the neutralization of TNF- or the blocking of Th1-type
cytokine pathways. By abrogating both Peyer’s and colonic patch
development, this treatment effectively abrogates the Th2 type of TNBS
colitis. Based upon our results, we envision that LTR-Ig would be
beneficial in the treatment of ulcerative colitis and perhaps also of
IBD cases refractory to other therapies.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Taeko Dohi, Department of Gastroenterology, Research Institute, International Medical Center of Japan, 1-21-1 Toyama, Shinjuku-ku, Tokyo 162-8655, Japan. E-mail address: dohi{at}ri.imcj.go.jp
3 Abbreviations used in this paper: IBD, inflammatory bowel disease; GALT, gut-associated lymphoreticular tissues; LT, lymphotoxin; LTR-Ig, LT receptor-Ig fusion protein; MPO, myeloperoxidase; PNA, peanut agglutinin; SLN, sacral lymph nodes; TNBS, 2,4,6-trinitrobenzene sulfonic acid; WT, wild type.
Received for publication November 17, 2000. Accepted for publication June 21, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-mediated oral tolerance. J. Exp. Med. 183:2605.-dependent fashion. J. Exp. Med. 187:1009. receptor. Immunity 9:71.[Medline]
receptor controls organogenesis and affinity maturation in peripheral lymphoid tissues. Immunity 9:59.[Medline]
complex is required for the development of peripheral lymphoid organs. J. Exp. Med. 184:1999. complexes on the surface of mouse lymphocytes. J. Immunol. 159:3288.[Abstract]
receptor by radioresistant stromal cells and of lymphotoxin and tumor necrosis factor by B cells. J. Exp. Med. 189:159. and tumor necrosis factor are required for stromal cell expression of homing chemokines in B and T cell areas of the spleen. J. Exp. Med. 189:403. gene-targeted mice. J. Immunol. 163:1350.-deficient mice show defective antiviral immunity. Virology 260:136.[Medline]
receptor: molecular genetics, ligand binding, and expression. J. Immunol. 155:5280.[Abstract]
T cell-deficient mice have impaired mucosal immunoglobulin A responses. J. Exp. Med. 183:1929. and IL-5. J. Immunol. 145:68.[Abstract]
and T cells. Eur. J. Immunol. 25:2743.[Medline]
TCR+ intestinal intraepithelial lymphocytes: Th1- and Th2-type cytokine production by CD4+CD8- and CD4+CD8+ T cells for helper activity. Int. Immunol. 5:1473. antibody (infliximab) therapy profoundly down-regulates the inflammation in Crohn’s ileocolitis. Gastroenterology 116:22.[Medline]
in Crohn’s disease. Lancet 349:521.[Medline]
This article has been cited by other articles:
|
|
 |
|
  K. A. Kudsk, F. E. Gomez, W. Kang, and C. Ueno Enteral Feeding of a Chemically Defined Diet Preserves Pulmonary Immunity but Not Intestinal Immunity: The Role of Lymphotoxin {beta} Receptor JPEN J Parenter Enteral Nutr, November 1, 2007; 31(6): 477 - 481. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Little, L. V. Bell, L. J. Cliffe, and K. J. Else The Characterization of Intraepithelial Lymphocytes, Lamina Propria Leukocytes, and Isolated Lymphoid Follicles in the Large Intestine of Mice Infected with the Intestinal Nematode Parasite Trichuris muris J. Immunol., November 15, 2005; 175(10): 6713 - 6722. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. W. Spahn, H.-P. Eugster, A. Fontana, W. Domschke, and T. Kucharzik Role of Lymphotoxin in Experimental Models of Infectious Diseases: Potential Benefits and Risks of a Therapeutic Inhibition of the Lymphotoxin-{beta} Receptor Pathway Infect. Immun., November 1, 2005; 73(11): 7077 - 7088. [Full Text] [PDF]
|
|
|
|
|
|
M.-N. Kweon, M. Yamamoto, P. D. Rennert, E. J. Park, A.-Y. Lee, S.-Y. Chang, T. Hiroi, M. Nanno, and H. Kiyono Prenatal Blockage of Lymphotoxin {beta} Receptor and TNF Receptor p55 Signaling Cascade Resulted in the Acceleration of Tissue Genesis for Isolated Lymphoid Follicles in the Large Intestine J. Immunol., April 1, 2005; 174(7): 4365 - 4372. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. G. Levisetti, A. Suri, K. Frederick, and E. R. Unanue Absence of Lymph Nodes in NOD Mice Treated With Lymphotoxin-{beta} Receptor Immunoglobulin Protects From Diabetes Diabetes, December 1, 2004; 53(12): 3115 - 3119. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Yamamoto, M.-N. Kweon, P. D. Rennert, T. Hiroi, K. Fujihashi, J. R. McGhee, and H. Kiyono Role of Gut-Associated Lymphoreticular Tissues in Antigen-Specific Intestinal IgA Immunity J. Immunol., July 15, 2004; 173(2): 762 - 769. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. V. Tumanov, D. V. Kuprash, J. A. Mach, S. A. Nedospasov, and A. V. Chervonsky Lymphotoxin and TNF Produced by B Cells Are Dispensable for Maintenance of the Follicle-Associated Epithelium but Are Required for Development of Lymphoid Follicles in the Peyer's Patches J. Immunol., July 1, 2004; 173(1): 86 - 91. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Stopfer, D. N. Mannel, and T. Hehlgans Lymphotoxin-{beta} Receptor Activation by Activated T Cells Induces Cytokine Release from Mouse Bone Marrow-Derived Mast Cells J. Immunol., June 15, 2004; 172(12): 7459 - 7465. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kawamura, T. Kanai, T. Dohi, K. Uraushihara, T. Totsuka, R. Iiyama, C. Taneda, M. Yamazaki, T. Nakamura, T. Higuchi, et al. Ectopic CD40 Ligand Expression on B Cells Triggers Intestinal Inflammation J. Immunol., May 15, 2004; 172(10): 6388 - 6397. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. C. Voon, L. S. Subrata, M. Karimi, D. Ulgiati, and L. J. Abraham TNF and Phorbol Esters Induce Lymphotoxin-{beta} Expression through Distinct Pathways Involving Ets and NF-{kappa}B Family Members J. Immunol., April 1, 2004; 172(7): 4332 - 4341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T W Spahn and T Kucharzik Modulating the intestinal immune system: the role of lymphotoxin and GALT organs Gut, March 1, 2004; 53(3): 456 - 465. [Full Text] [PDF]
|
|
|
|
|
|
R. A. Fava, E. Notidis, J. Hunt, V. Szanya, N. Ratcliffe, A. Ngam-ek, A. R. de Fougerolles, A. Sprague, and J. L. Browning A Role for the Lymphotoxin/LIGHT Axis in the Pathogenesis of Murine Collagen-Induced Arthritis J. Immunol., July 1, 2003; 171(1): 115 - 126. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-C. Chen, M.-J. Hwang, Y.-C. Chou, W.-H. Chen, G. Cheng, H. Nakano, T.-Y. Luh, S.-C. Mai, and S.-L. Hsieh The Role of Apoptosis Signal-regulating Kinase 1 in Lymphotoxin-beta Receptor-mediated Cell Death J. Biol. Chem., April 25, 2003; 278(18): 16073 - 16081. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kuai, E. Nickbarg, J. Wooters, Y. Qiu, J. Wang, and L.-L. Lin Endogenous Association of TRAF2, TRAF3, cIAP1, and Smac with Lymphotoxin beta Receptor Reveals a Novel Mechanism of Apoptosis J. Biol. Chem., April 11, 2003; 278(16): 14363 - 14369. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Kato, K. Fujihashi, R. Kato, T. Dohi, K. Fujihashi, Y. Hagiwara, K. Kataoka, R. Kobayashi, and J. R. McGhee Lack of oral tolerance in aging is due to sequential loss of Peyer's patch cell interactions Int. Immunol., February 1, 2003; 15(2): 145 - 158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. W. Spahn, H. Herbst, P. D. Rennert, N. Lugering, C. Maaser, M. Kraft, A. Fontana, H. L. Weiner, W. Domschke, and T. Kucharzik Induction of Colitis in Mice Deficient of Peyer's Patches and Mesenteric Lymph Nodes Is Associated with Increased Disease Severity and Formation of Colonic Lymphoid Patches Am. J. Pathol., December 1, 2002; 161(6): 2273 - 2282. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
