|
|
||||||||




Departments of
*
Medical Microbiology and Immunology and
Pathology, University of Ulm, Ulm, Germany;
Roche Genetics, F. Hoffmann LaRoche, Basel, Switzerland;
Kirin Brewery, Pharmaceutical Research Laboratory, Gunma, Japan; and
¶ Ludwig Institute for Cancer Research, Epalinges, Switzerland
| Abstract |
|---|
|
|
|---|

T cells from either normal C57BL/6 (B6) or
MHC-II-deficient (A
-/- or A
-/-) B6
donor mice engrafted into congenic immunodeficient
RAG1-/- B6 hosts induced an aggressive inflammatory bowel
disease (IBD). Furthermore, CD4+ T cells from
CD1d-/- knockout (KO) B6 donor mice but not those from
MHC-I-/- (homozygous transgenic mice deficient for
2-microglobulin) KO B6 mice induced a colitis in
RAG-/- hosts. Abundant numbers of in vivo activated
(CD69highCD44highCD28high)
NK1+ and NK1- CD4+ T cells were
isolated from the inflamed colonic lamina propria (cLP) of transplanted
mice with IBD that produced large amounts of TNF-
and IFN-
but
low amounts of IL-4 and IL-10. IBD-associated cLP Th1 CD4+
T cell populations were polyclonal and MHC-II-restricted when derived
from normal B6 donor mice, but oligoclonal and apparently
MHC-I-restricted when derived from MHC-II-deficient
(A
-/- or A
-/-) B6 donor mice. cLP
CD4+ T cell populations from homozygous transgenic mice
deficient for
2-microglobulin KO B6 donor mice engrafted
into RAG-/- hosts were Th2 and MHC-II restricted. These
data indicate that MHC-II-dependent as well as MHC-II-independent
CD4+ T cells can induce a severe and lethal IBD in
congenic, immunodeficient hosts, but that the former need the latter to
express its IBD-inducing potential. | Introduction |
|---|
|
|
|---|

T cell
subset, a congenic, severely immunodeficient host, and an intact gut
flora of the host. A defect in the immunoregulation of mucosal T cell
responses is supposed to play a key role in the pathogenesis of colitis
in these models, with exaggerated IFN-
and TNF-
responses as
major mediators of this disease (reviewed in (9, 10).
These pathogenic T cell responses may be driven by Ags of the
intestinal flora to which these T cells are normally tolerant. In the
host developing colitis, neither the inducing
CD4+ T cell subset nor the stimulus that drives
the Th1-biased T cell activation in the mucosa has been identified.
Splenic CD4+ T cells with a
CD45RBhigh surface phenotype have an enhanced
IBD-inducing potential in H-2d
C.B-17scid/scid (SCID) mice (reviewed in Ref.
11). It is unresolved which IBD-inducing
CD4+ T cell subset(s) the
CD45RBhigh surface phenotype identifies, and it
has been shown that CD45RBhigh as well as
CD45RBlowCD4+ T cell can
induce a colitis in this adoptive transfer system (8, 12).
IBD-associated effector T cell populations accumulating in the inflamed
colonic lamina propria (cLP) in different mouse models can be
conventional CD4+ 
+ T
cells, 
+ and 
+
T cells (13), CD8+

+ T cells (14), or unusual
TCR
-
+ T cells
(15). Key questions of the pathogenesis of T cell-induced
IBD are unresolved. Preclinical mouse models offer an attractive
approach to clarify some aspects of these common and debilitating
diseases that may pave the way for the rational design of novel
therapeutic approaches.
Development of epitope recognition of most CD4+

T cells is MHC-II-dependent. A small subset of
MHC-II-independent CD4+ 
T cells is present
in normal and MHC-II-deficient (A
-/- or
A
-/- knockout (KO)) mice. MHC-II-independent
CD4+ 
T cells in
A
-/- mice are thymically derived, appear
early in ontogeny, localize preferentially to the B rather than to the
T cell areas in peripheral lymphoid organs, exhibit the phenotype of
resting or activated memory T cells, and have a diverse TCR
repertoire that is potentially functional (16, 17). These
T cell populations are heterogeneous with respect to phenotype, array
of peptides and/or glycolipids recognized, and restriction elements
used for specific recognition (18, 19, 20, 21, 22). Only one subset
within the MHC-II-independent CD4+ T cells, i.e.,
CD1d-restricted NK1+ T cells expressing the NK1
marker (NKT) cells is well defined (reviewed in Refs. 23, 24). NKT cells express the semi-invariant V
14J
281 V
8.2
TCR, bear the NK1.1 marker (in appropriate mouse strains), are found
mainly in thymus and liver, rapidly produce the cytokines IFN-
and
IL-4 after stimulation, and recognize the glycolipid
-galactosyl
ceramide (
-GalCer) in the context of CD1d (17, 25, 26, 27, 28, 29, 30, 31).
MHC-II-independent, CD1d-restricted NKT cells are absent from
CD1d-/- KO mice (32, 33, 34). The
presence and function of MHC-II-independent CD4+
T cells in the gut is not clear. Although NK-like T cells have been
found in the intraepithelial compartment of the small intestine of mice
(35), NK1+ NKT cells are absent from
the intestinal intraepithelial lymphocyte and lamina propria
lymphocyte (LPL) populations (36). Recently,
CD1d-glycolipid tetramer staining has revealed
NK1-CD4-CD8-
tetramer-positive T cells with the characteristic semi-invariant TCR in
lymph nodes and the small intestine (22, 37, 38).
We have transferred splenic CD4+ 
T cells
developing in a normal (MHC-I, MHC-II, CD1d-expressing),
MHC-II-deficient (A
-/- or
A
-/- KO) or MHC-I-deficient (homozygous
transgenic mice deficient for
2-microglobulin
(
2m-/-) or
CD1d-/-) environment into congenic, severely
immunodeficient RAG1-/- hosts to test their
IBD-inducing potential. Unexpectedly, the adoptively transferred
MHC-II-independent CD4+ T cells efficiently
induced colitis. This observation points to a new feature of
MHC-II-independent CD4+ T cells, their potential
to trigger inflammatory Th1-type reactions in the mucosa.
| Materials and Methods |
|---|
|
|
|---|
Normal B6 mice, B6 MHC-II-/-
(A
-/- or A
-/-) mice (16, 17),
C57BL/6J-Rag1tm1Mom
(RAG1-/-) mice (39) (The Jackson Laboratory,
Bar Harbor, ME), CD1d-deficient (CD1d-/-) B6 mice
(34) and B6 MHC-I-/-
(
2m-/-) mice (40) were used.
Mice were bred and kept under specific pathogen-free conditions in the
animal facility of Ulm University. RAG1-/- mice were
transplanted at 812 wk of age.
Flow cytometry (FCM) analyses of the surface phenotype and intracellular cytokine expression
Cells were suspended in PBS/0.3% w/v BSA supplemented with
0.1% w/v sodium azide. Nonspecific binding of Abs to Fc receptor was
blocked by preincubating cells with the mAb 2.4G2 directed against the
Fc
RIII/II CD16/CD32 (1 µg mAb/106 cells/100 µl).
Cells were incubated with 0.5 µg/106 cells of the
relevant mAb for 30 min at 4°C and washed. In most experiments, cells
were subsequently incubated with a second-step reagent for 10 min at
4°C. Three-color FCM analyses were performed on a FACSCalibur (BD
Becton Dickinson, Mountain View, CA). The forward narrow angle light
scatter was used as an additional parameter to facilitate exclusion of
dead cells and aggregated cell clumps. The following reagents and mAb
were obtained from BD PharMingen (Hamburg, Germany): PE-conjugated
anti-CD3
mAb 145-2C11, FITC- and PE-conjugated anti-CD4 mAb
GK1.5, biotinylated anti-CD4 mAb RM4-5, FITC-conjugated
anti-CD8
mAb 53-6.7, biotinylated anti-CD44 (Pgp-1) mAb IM7,
biotinylated anti-CD45RB mAb 23G2, biotinylated anti-CD28 mAb
37.51, biotinylated anti-CD69 mAb H1.2F3, PE-conjugated
anti-NK1.1 mAb PK136, biotinylated anti-CD40L mAb MR1, and
FITC-conjugated anti-CD62L (L-selectin) mAb MEL-14. PE-conjugated
streptavidin was obtained from PharMingen. SA-Red670 was obtained from
Life Technologies (Berlin, Germany).
Cells (106 cells/ml) were stimulated with 50
ng/ml PMA and 500 ng/ml ionomycin in the presence of 10 µg/ml
brefeldin A in RPMI 1640/10% FCS for 12 h at 37°C with 5%
CO2. Cells were harvested, washed twice in
staining buffer (PBS without
Mg2+/Ca2+, 0.3% w/v BSA,
0.1% w/v sodium azide), incubated (15 min, 4°C) with purified 2.4G2
Ab to block nonspecific binding of Ab to Fc receptors, washed with
staining buffer, resuspended in staining buffer, and surface stained
with the relevant Abs. Cells were washed with staining buffer, labeled
with the second-step reagent, and washed twice. Cells were then
resuspended in 100 µl Cytofix/Cytoperm solution for 20 min at 4°C
and washed twice in 1 ml 1x Perm/Wash solution. Fixed and
permeabilized cells were resuspended in 100 µl of 1x Perm/Wash
solution. Cells were stained for 30 min at 4°C with 1 µg
mAb/106 cells of FITC-conjugated anti-IL-4
mAb BVD4-1D11, FITC-conjugated anti-IL-10 mAb JES5-16E3,
FITC-conjugated anti-TNF-
mAb MP6-XT22, FITC-conjugated
anti-IFN-
mAb XMG1.2, or appropriate negative control Abs
(FITC-conjugated rat IgG1 mAb R3-34; PE-conjugated rat IgG1 mAb R3-34).
Cells were washed twice in 1x Perm/Wash solution (250 x
g) and resuspended in staining buffer;
104 cells were analyzed by FCM using a FACScan
equipped with a 15-mW argon laser (BD Becton Dickinson) using the
CellQuest software (BD Becton Dickinson).
CD4+ T cells used for adoptive transfer
CD4+ T cells were aseptically purified from spleen cells depleted of CD8+ T cells by treatment with anti-CD8 Ab and low toxicity rabbit complement (Cedarlane Laboratories, Hornby, Ontario, Canada) following the manufacturers instructions. CD4+ T cells were enriched to >98% purity by positive selection on MACS separation columns (Milteny Biotec, Bergisch-Gladbach, Germany). Briefly, cell suspensions were washed in MACS buffer (PBS without Mg2+ and Ca2+ supplemented with 2 mM EDTA and 0.5% BSA) and incubated 20 min at 4°C with MACS CD4 MicroBeads. The magnetically labeled positive fraction was retained in a magnetic field on VS+ MACS columns. The purity of the positively separated CD4+ population was routinely >98%. Into RAG-/- B6 mice, 3 x 105 CD4+ cells were injected i.p. At biweekly intervals, the transplanted mice were weighed and their clinical condition was monitored.
Isolation of lymphoid cell populations from transplanted mice
Transplanted RAG- mice were sacrificed by cervical dislocation. Single cell suspensions were aseptically prepared from the spleen, the mesenteric lymph nodes, and the lamina propria of the intestine. Colonic LPLs were isolated as described (12, 41, 42).
Generation of myeloid dendritic cells (DC) from bone marrow
The in vitro generation of myeloid DC from murine bone marrow has been described (43). Briefly, bone marrow cells prepared from femurs were depleted of CD4+CD8+B220+ lymphocytes and MHC-class-II+ cells (Miltenyi Biotec) by MACS sorting. These bone marrow cells depleted of T cells, B cells, and maturing myeloid cells were cultured at a density of 106 cells/ml in 6-well plates (Nunc, Wiesbaden, Germany) in serum-free UltraCulture medium (BioWhittaker, Verviers, Belgium) supplemented with 5 ng/ml GM-CSF and 10 ng/ml FL (PeproTech, Rocky Hill, NJ), 2 mM glutamine, and antibiotics. Cultures were incubated at 37°C in humidified air supplemented with 5% CO2. On days 3 and 5, cells were fed by medium exchange. From day 7 of cultures, nonadherent CD11c+ cells were purified by magnetic bead separation (Miltenyi Biotec), pulsed with bacterial lysates, and used as stimulator cells for CD4+ T cells.
Cytokine determination by ELISA
The release of cytokines by CD4+ T cells
was detected by a conventional double-sandwich ELISA. For detection and
capture, the mAb R4-6A2 and biotinylated mAb AN18 were used for IFN-
(PharMingen). Extinction was analyzed at 405/490 nm on Spectra-Max
equipment (Molecular Devices, Sunnyvale, CA) using the Softmax Pro
software (Molecular Devices).
Histopathological examinations
Tissue samples were taken from various locations of the small and large intestine (duodenum, jejunum, terminal ileum, cecum, ascending colon, transverse colon, recto-sigmoid colon). Tissue was fixed in neutral buffered formalin, embedded in paraffin, sectioned on a microtome, mounted on slides, and stained with hematoxylin/eosin.
| Results |
|---|
|
|
|---|
-/- or A
-/-), or
MHC-I-deficient (
2m-/-,
CD1d-/-) B6 mice
We analyzed the surface phenotype, the inducible expression of
proinflammatory cytokines, the response to
-GalCer, and the
diversity of the TCR V
repertoire of CD4+ T
cells from the B6 lines that were used as donors in the adoptive
transfer experiments described below. Approximately one-third of the
mononuclear cells from the spleen or cLP from normal, MHC-II-deficient
(A
-/- or A
-/-) or
MHC-I-deficient (
2m-/-
or CD1d-/-) B6 mice were
CD3+ T cells. As expected,
A
-/- and A
-/-
mice had few CD4+ T cells, and
2m-/- mice had few
CD8+ T cells.
NK1+CD4+ T cells were found
in all mice tested. A 13% fraction of the splenic T cells from
normal and MHC-I-deficient mice were NK1+, and
the majority of splenic CD4+ T cells from
A
-/- and A
-/-
(MHC-II-/-) B6 mice were
NK1+. Furthermore, 515% of cLP
CD4+ and CD8+ T cells from
normal mice and the majority of cLP CD4+ T cell
populations from A
-/- and
A
-/- KO mice were
NK1+. Within the splenic
NK1+ CD3+ T cell
populations in both normal and MHC-II-deficient B6 mice 3045% were
CD4+CD8-, while the
remaining NK1+ T cells were
CD4-CD8- or
CD4-CD8+. Splenic
NK1+CD4+ T cells showed an
effector/memory T cell phenotype with high expression of costimulator
molecules but also high CD45RB expression (data not shown).
Twice as many splenic CD4+ T cells from
MHC-II-deficient than from normal B6 mice produced high levels of
TNF-
and/or IFN-
after a 12-h incubation with phorbol ester and
ionomycin (Fig. 1
A). Splenic
CD4+ T cells from normal and MHC-II-deficient
(A
-/- or A
-/-) B6
mice proliferated and produced IFN-
in response to stimulation by
Con A, anti-CD3
Ab or
-GalCer; splenic
CD4+ T cells from MHC-I-deficient
(
2m-/-,
CD1d-/-) B6 mice responded to polyclonal
stimulation by Con A or anti-CD3
Ab but not to stimulation by
-GalCer (Fig. 1
B, and data not shown). cLP
CD4+ T cells from all lines released IFN-
in
response to polyclonal T cell stimulation by Con A or anti-CD3
Ab but did not respond to
-GalCer stimulation (Fig. 1
B,
and data not shown). The TCR V
repertoire of
CD4+ T cells from the spleen and the cLP of
normal B6 or MHC-II-deficient A
-/- KO B6
mice was diverse (data not shown).
|
When splenic CD4+ T cells from either normal
or MHC-II-deficient (A
-/- or
A
-/- KO) B6 mice were injected into congenic
RAG1-/- B6 hosts (3 x
105 cells/mouse), these mice showed diarrhea and
rectal prolapse within 36 wk posttransfer (Fig. 2
A). Many mice lost >20% of
their body weight within 36 wk posttransfer. Loss in body weight did
not strictly correlate with disease severity, because some transplanted
mice died early posttransfer without significant loss of body weight,
although they showed a histopathology of severe IBD. This IBD was
progressive, severe, and always lethal. The course of the IBD in B6
RAG1-/- hosts induced by the transfer of
congenic CD4+ T cells was more aggressive and
showed less interindividual variability than did the IBD in SCID mice
induced by the transfer of BALB/c-derived CD4+ T
cells that we have described in detail previously (12). Of
particular interest was the observation that CD4+
T cells derived from MHC-II-deficient (A
-/-
or A
-/-) B6 donor mice induced an aggressive
form of the disease. Transfer of titrated numbers of purified
CD4+ T cells from B6
A
-/- donor mice into
RAG1-/- hosts confirmed the efficient
IBD-inducing potential of MHC-II-independent CD4+
T cells (Fig. 2
B). CD4+ T cells from
A
-/- and A
-/- KO
B6 donor mice were equally efficient in inducing IBD (data not shown).
These data reveal a new feature of MHC-II-independent
CD4+ T cells: their unexpected efficiency in
inducing mucosal inflammation.
|
|
We isolated cLP T cells from RAG1-/- mice
with severe IBD that were transplanted with splenic
CD4+ T cells from MHC-II-dependent or
-independent B6 donor mice. The phenotype of IBD-associated cLP
CD4+ T cells indicated that they are activated in
situ evident by low expression of CD62L and high expression of CD69 and
CD44 in all cLP CD4+ T cell populations tested. A
fraction (2030%) of cLP CD4+ T cells from
RAG1-/- hosts reconstituted with
CD4+ T cells from normal or
A
-/- B6 donor mice was
NK1+. Surface expression of NK1.1 by cLP
CD4+ T cells was lower than by cLP NK cells. In
FCM analyses, we characterized the surface phenotype of the
NK1+ and NK1- subsets of
the cLP CD4+ T cell populations because
differences were apparent in the surface phenotype of cLP
NK1+ vs NK1-
CD4+ T cells (Fig. 4
). cLP NK1+
CD4+ T cells from diseased
RAG-/- hosts expressed higher surface levels of
CD45RB than did NK1- CD4+
T cell in the same LPL population. Surface expression of CD28 was
higher in NK1+ CD4+ T cells
than in NK1- CD4+ T cells.
Expression of CD40L was low in NK1+ and
NK1- CD4+ T cells.
|
and IFN-
(Fig. 5
-/- or A
-/- B6
mice expressed these two potent proinflammatory cytokines than did cLP
CD4+ T cells from normal B6 donor mice. This may
be related to the more aggressive course of colitis seen after transfer
of MHC-II-independent CD4+ T cells, despite the
observation that lower numbers of cLP CD4+ T
cells were recovered from the cLP of diseased
RAG-/- mice transplanted with
A
-/- or A
-/- B6
donor-derived T cells than in those transplanted with
CD4+ T cells from normal B6 donor mice. cLP and
splenic CD4+ T cells from MHC-II-competent or
MHC-II-deficient (A
-/- or
A
-/-) donor mice engrafted in diseased
RAG1-/- mice released IFN-
in response to
polyclonal T cell activation (by Con A or anti-CD3
mAb) but not
to
-GalCer (data not shown). The analysis of the TCR V
repertoire
of CD4+ T cells isolated from the cLP of
individual transplanted RAG-/- mice with
colitis indicated that cLP CD4+ T cell
populations derived from normal B6 donor mice always had a polyclonal
TCR V
repertoire. In contrast, three of five analyzed
RAG-/- mice transplanted with
CD4+ T cells from A
-/-
B6 donor mice showed a preferential usage of TCR V
5, indicating an
oligoclonal TCR V
repertoire of the disease-associated cLP
CD4+ T cells (data not shown).
|
-) B6 donor mice is MHC-I- or
MHC-II-dependent. Myeloid DC grown from bone marrow in
GM-CSF/FL-supplemented, serum-free cultures were used as stimulator
cells. These presenting cells were pulsed with bacterial lysates
derived from diseased RAG-/- mice as described
(44, 45). The data in Fig. 6
when restimulated with syngeneic (MHC-I- and
MHC-II-expressing), lysate-pulsed but not nonpulsed DC. This response
was MHC-II restricted because it was completely blocked by an
anti-Ab mAb. CD4+ T
cells from A
-/- B6 mice obtained from the
cLP of diseased, transplanted RAG-/- hosts
released IFN-
when restimulated with syngeneic normal or
MHC-II-deficient, lysate-pulsed DC (Fig. 6
when stimulated with MHC-I-deficient,
lysate-pulsed DC from
2m-/- KO B6 mice (Fig. 6
-/- B6 donor mice have a preferential
MHC-I-dependent reactivity. These data suggest that disease-associated
CD4+ T cells, raised in an MHC-II-dependent or
MHC-II-independent way, differ in their restricted recognition
pattern.
|
MHC-II-independent CD4+ T cells efficiently
induced a colitis after transfer into RAG1-/-
hosts. CD1d-restricted and
-GalCer-reactive NKT cells are the only
well-characterized, MHC-II-independent CD4+ T
cell subset (27, 30, 46, 47). In cLP
CD4+ T cell populations, we found
NK1+ T cells. cLP T cells from transplanted and
diseased hosts did not respond to
-GalCer-pulsed
CD1d+ DC (data not shown). This indicated that
well-characterized CD1d-restricted NK1+
CD4+ T cells reactive to
-GalCer are not
involved in IBD induction. This was confirmed in transfer experiments
using CD4+ T cells from
CD1d-/- KO B6 donor mice.
RAG1-/- B6 hosts injected with purified,
splenic CD4+ T cells from
CD1d-/- KO B6 mice developed a colitis (Fig. 2
). The time course of this disease, its histopathology, and the
phenotype of the cLP CD4+ T cells isolated from
diseased RAG1-/- resembled the IBD induced by
transfer of normal B6 CD4+ T cells (data not
shown). Furthermore, transfer of CD4+ T cells
from J
281-/- KO B6 donor mice into
RAG1-/- B6 mice induced a colitis with a
clinical course and histopathology similar to that induced by normal B6
CD4+ T cells (data not shown). cLP
CD4+ T cells stimulated with lysate-pulsed DC
from CD1d-/- B6 KO mice released IFN-
indicating that this MHC-I-like molecule is not involved (data not
shown). In the last set of experiments, we transferred
CD4+ T cells from MHC-I-deficient
2m-/- KO B6 mice into
RAG1-/- hosts to test whether MHC-I-dependent
CD4+ T cells are required to drive the disease
process.
CD4+ T cells from MHC-I-deficient
2m-/- KO B6 donors induce no (or only very
mild) colitis in RAG1-/- KO hosts
CD4+ T cells were found in the spleen and
cLP of MHC-I-deficient
2m-/- KO B6 mice.
After transfer into RAG1-/- hosts, these cells
induced no or only mild histopathological signs of colitis during a 10-
to 16-wk observation period that were not accompanied by clinical signs
(Fig. 2
A). Signs of a mild colitis were found only in 5/22
RAG1-/- hosts transplanted with purified
CD4+ T cells from
2m-/- KO B6 donor mice
(Fig. 3
F); the other hosts showed an essentially normal
histology without signs of inflammation (Fig. 3
A). The
surface phenotype of cLP CD4+ from
RAG1-/- mice transplanted with T cells from
2m-/- KO B6 donors was
similar to that described above for cLP CD4+ T
cells from RAG1-/- hosts transplanted with T
cells from normal, MHC-II-deficient or CD1d-deficient B6 donor mice
(Fig. 7
A). Striking
differences were observed in the cytokine expression profile of cLP
CD4+ T cells from nondiseased
RAG1-/- hosts transplanted with
CD4+ T cells from MHC-I-deficient
2m-/- KO B6 mice. Many
of these T cells expressed IL-4 but few produced IFN-
or TNF-
(Fig. 7
B). This Th2-biased cytokine expression pattern is in
contrast to the Th1-biased pattern described above for cLP
CD4+ T cells from RAG1-/-
B6 mice with colitis transplanted with CD4+ T
cells from normal B6, MHC-II-deficient (A
-/-
or A
-/-) B6, or
CD1d-/- KO B6 mice. These data point to
MHC-I/
2m-dependent (but not CD1d-restricted)
CD4+ T cells that promote the Th1-biased
polarization of mucosal T cell responses in the colon.
|
| Discussion |
|---|
|
|
|---|

T cells. Their development seems to
depend on conventional MHC-Ia molecules, MHC-Ib molecules (e.g., Qa-1),
or CD1d molecules. Identification of MHC-II-independent
CD4+ T cell subsets is difficult with the
currently available tools because no reliable markers are available to
identify this CD4+ T cell subsets. The NK1 marker
is not expressed by all MHC-II-independent T cells, its expression is
not stable (48), and it is expressed on many activated,
conventional CD4+ and CD8+
T cells in vivo (49, 50). MHC-II-independent
CD4+ T cells are present in the noninflamed
intestinal mucosa. MHC-II-independent CD4+ T
cells were found in the intestinal lamina propria in
A
-/- or A
-/-
B6 mice.
The main finding of our study was the unexpected potency of
MHC-II-independent CD4+ T cells to induce
colitis. CD4+ T cells from either
A
-/- or A
-/- KO
B6 mice efficiently induced a colitis in congenic immunodeficient
RAG1-/- B6 hosts. This excludes the possibility
that minor populations of MHC-II (A
E
)-dependent
CD4+ T cells from A
-/-
KO mice can induce colitis or that minor histocompatibility differences
between donor and host play a role (A
-/- KO
mice were generated in B6 mice; A
-/- KO mice
were from the 14th backcross to B6). The MHC-II-independent
CD4+ T cell subset that induced colitis is
unknown. These T cells are present in normal,
MHC-II-/- (A
-/- or
A
-/-), CD1d-/-,
J
281-/- but not in
MHC-I-/-
(
2m-/-) B6 mice. They are
inducible to rapidly express proinflammatory cytokines (TNF-
,
IFN-
), and they repopulate and expand in the immunodeficient host.
IBD-associated cLP CD4+ T cells from
A
-/- or A
-/- KO
B6 donor mice showed variable NK1 expression, did not respond to
-GalCer, and showed (in most but not all mice) an oligoclonal TCR
V
5 bias. MHC-II-independent CD4+ T cells
involved in inducing IBD are not the well-characterized
CD4+ NKT cells. NKT cells are a fairly constant T
cell subset of
106 cells/organ in many
lymphoid and nonlymphoid organs of the mouse. The majority of
CD4+ NKT cells are in vivo activated, rapidly
release cytokines after stimulation, express intermediate levels of CD3
and TCR
, and use an invariant V
14J
281 TCR
-chain (with
an invariant CDR3 region containing no N-region additions/deletions)
paired with V
8.2, V
7, or V
2 (but not V
5) TCR
-chains
(with a variable CDR3 region). CD4+ NKT cells
recognize
-GalCer (25, 26, 27, 28, 30) in the context of CD1d
(17, 24, 36). In contrast to the restricted TCR V
and
V
repertoire of CD1d-restricted,
-GalCer-specific
CD4+ NKT cells, other MHC-II-independent
CD4+ T cells show a diverse TCR
repertoire
and do not recognize
-GalCer. CD4+ T cells
from A
-/- and
A
-/- B6 donor mice that repopulated spleen,
mesenteric lymph nodes, and gut mucosa of
RAG1-/- hosts showed a diverse TCR V
repertoire at an early stage of the disease, but a restricted TCR V
usage pattern as the disease progressed. In 3 of 5 late-stage,
transplanted IBD+ mice that we analyzed in
detail, >80% of the T cells expressed TCR V
5. This was not
observed in splenic or cLP CD4+ T cells isolated
from 8 RAG1-/- hosts transplanted with
CD4+ T cells from normal,
CD1d-/- or J
281-/-
KO mice (data not shown). MHC-II-independent CD4+
T cell populations that expand in the immunodeficient host during the
emergence of IBD, thus, tend to become oligoclonal. This suggests that
a selection process drives mucosal repopulation.
Histopathological evidence of colitis was observed in only 5 of 22
RAG1-/- mice transplanted with
CD4+ T cells from
2m-/- KO B6 mice (in
four independent experiments). Histologically, these five animals
showed only mild colitis late in a 10- to 16-wk observation period. The
surface phenotype of cLP CD4+ T cells from these
mice did not differ from that of cLP CD4+ T cells
from severely diseased RAG1-/- mice
transplanted with CD4+ T cells from normal B6,
A
-/- KO or A
-/-
KO B6 mice (Fig. 7
A). The only striking difference observed
was the cytokine expression profile of cLP CD4+ T
cells from transplanted RAG1-/- mice. Although
cLP CD4+ T cells from diseased
RAG1-/- mice transplanted with T cells from
normal, A
-/- or
A
-/- B6 mice produced TNF-
and IFN-
but no IL-4, only few cLP CD4+ T cells derived
from
2m-/- KO B6 mice
produced TNF-
or IFN-
but many produced IL-4 and IL-10 (Fig. 7
B). The lack or very mild course of colitis, thus,
coincided with a Th2-biased polarization of the T cell response in the
cLP. A control of aggressive CD4+ T cells
(derived from CD45RBhigh precursors) by
suppressive CD4+ T cells (derived from
CD45RBlow precursors) in the pathogenesis of
colitis has been proposed (3, 11, 51, 52, 53, 54). A predominance
of suppressive CD4+ T cells in
2m-/- B6 mice may
explain the low incidence of colitis induction. When MHC-II-independent
CD4+ T cells (from
A
-/- B6 mice) were either mixed to
MHC-I-independent CD4+ T cells (from
2m-/- B6 mice) or
transferred into disease-free RAG-/- hosts
repopulated for 814 wk with CD4+ T cells from
2m-/- donor B6 mice, a
colitis developed (unpublished data). This argues against a dominant
suppressive activity of CD4+ T cells from
2m-/- donor mice. It
has been shown that IL-2-/- x
2m-/- double KO mice
spontaneously develop a CD4+ T cell-dependent
colitis (55). In contrast to our data, MHC-II-dependent
but not MHC-I-dependent CD4+ T cells are critical
for the development of colitis in this model. It has been reported that
the transfer of 45 x 105
CD45RBhigh CD4+ T cells
into MHC-II-/- x
RAG-/- double KO hosts does not induce colitis
(56). This graft is expected to contain
103 MHC-II-independent
CD4+ T cells. In the absence of MHC-II-dependent
CD4+ T cells, this small number of transferred
MHC-I-dependent CD4+ T cells, thus, does not
induce colitis. Alternatively, MHC-I-dependent
CD4+ T cells in A
-/-
or A
-/- KO mice and in normal mice may
differ.
We reported: 1) the efficient induction of IBD by
CD4+ 
T cells from either normal,
CD1d-/- and J
281-/-
mice, or MHC-II-deficient (A
-/- or
A
-/-) mice; and 2) no (or inefficient)
induction of IBD by CD4+ 
T cells from
MHC-I-deficient
(
2m-/-) mice.
Therefore, dysregulation of T cell reactivity seems to be an important
initiating event in the pathogenesis of colitis. We described three
experimental conditions in which MHC-I- and/or MHC-II-dependent
CD4+ T cells were either present or absent
(1). Transfer of CD4+ T cells from
A
-/- or A
-/- KO
mice: in the absence of conventional MHC-II-restricted
CD4+ T cells, large numbers of MHC-II-independent
CD4+ T cells expand and induce IBD
(2). Transfer of CD4+ T cells from
normal, CD1d-/- or
J
281-/- mice: in the presence of low numbers
of MHC-II-independent CD4+ T cells and large
numbers of conventional MHC-II-restricted CD4+ T
cells, MHC-II-dependent CD4+ T cells expand
polyclonally, develop a Th1-biased reactivity, and induce IBD
(3). Transfer of CD4+ T cells from
2m-/- mice: only
conventional MHC-II-restricted CD4+ T cells
expand polyclonally that do not induce IBD because they default to a
Th2-biased reactivity (in the absence of regulatory MHC-II-independent
CD4+ T cells). MHC-I-dependent
CD4+ T cells, thus, seem to facilitate
Th1-biased, mucosal CD4+ T cell responses. A
regulatory role of NKT cells has been proposed in the pathogenesis of
graft-versus-host disease and autoimmune disease (57, 58, 59).
To our knowledge, we provide the first evidence that MHC-II- and
CD1d-independent CD4+ T cells can modulate the
polarization of mucosal T cell responses in the gut.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Jörg Reimann, Department of Medical Microbiology and Immunology, University of Ulm, Helmholtzstrasse, 8/1, D-89081, Ulm, Germany. ![]()
3 Abbreviations used in this paper: IBD, inflammatory bowel disease; cLP, colonic lamina propria; KO, knockout; NKT cells, T cells expressing the NK1 marker;
-GalCer,
-galactosyl ceramide; LPL, lamina propria lymphocyte;
2m-/-, homozygous transgenic mice deficient for
2-microglobulin; FCM, flow cytometry; DC, dendritic cell(s); B6, C57BL/6. ![]()
Received for publication August 22, 2000. Accepted for publication January 3, 2001.
| References |
|---|
|
|
|---|

+ T cells in the pathogenesis of murine inflammatory bowel disease. Eur. J. Immunol. 24:2803.[Medline]
expression by T cells. J. Exp. Med. 187:1225.
+ and TCR
+ T cells in an experimental model of colitis. Eur. J. Immunol. 27:17.[Medline]
gene in C57.BL/6 mice. Int. Immunol. 5:957.
+ cells: new clues to their origin, specificity, and function. J. Exp. Med. 182:633.
14 NKT cells by glycosylceramides. Science 278:1626.
-galactosylceramide by natural killer T cells is highly conserved through mammalian evolution. J. Exp. Med. 188:1521.
-galactosylceramide specifically stimulates V
14+ NK T lymphocytes. J. Immunol. 161:3271.
-galactosylceramide polarizes CD1-reactive NK T cells towards Th2 cytokine synthesis. Eur. J. Immunol. 29:2014.[Medline]
-galactosylceramide demonstrates its immunopotentiating effect by inducing interleukin (IL)-12 production by dendritic cells and IL-12 receptor expression on NKT cells. J. Exp. Med. 189:1121.
-galactosylceramide directs conventional T cells to the acquisition of a Th2 phenotype. J. Immunol. 163:2373.
+ cells in the liver of mice. J. Exp. Med. 180:699.
2m, MHC class I proteins, and CD8+ T cells. Science 248:1227.
T cells in SCID mice with colitis. Eur. J. Immunol. 26:3074.[Medline]

T cells of athymic origin in histocompatible severe combined immunodeficient (SCID) mice. Clin. Exp. Immunol. 112:443.[Medline]
but not interleukin4 in the suppression of T helper type 1-mediated colitis by CD45RBlow CD4+ T cells. J. Exp. Med. 183:2669.
14-J
281 transgenic nonobese diabetic mice against diabetes. J. Exp. Med. 188:1831.This article has been cited by other articles:
![]() |
P. Bochtler, C. Wahl, R. Schirmbeck, and J. Reimann Functional Adaptive CD4 Foxp3 T Cells Develop in MHC Class II-Deficient Mice J. Immunol., December 15, 2006; 177(12): 8307 - 8314. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Bienvenu, B. Martin, C. Auffray, C. Cordier, C. Becourt, and B. Lucas Peripheral CD8+CD25+ T Lymphocytes from MHC Class II-Deficient Mice Exhibit Regulatory Activity J. Immunol., July 1, 2005; 175(1): 246 - 253. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Takeshita, T. Yamasaki, S. Akira, F. Gantner, and K. B. Bacon Essential role of MHC II-independent CD4+ T cells, IL-4 and STAT6 in contact hypersensitivity induced by fluorescein isothiocyanate in the mouse Int. Immunol., May 1, 2004; 16(5): 685 - 695. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. H. Boyle and J. S. Hill Gaston Breaking the rules: the unconventional recognition of HLA-B27 by CD4+ T lymphocytes as an insight into the pathogenesis of the spondyloarthropathies Rheumatology, March 1, 2003; 42(3): 404 - 412. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. Pinkoski, N. M. Droin, and D. R. Green Tumor Necrosis Factor alpha Up-regulates Non-lymphoid Fas-ligand following Superantigen-induced Peripheral Lymphocyte Activation J. Biol. Chem., October 25, 2002; 277(44): 42380 - 42385. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. H. Boyle, J. C. Goodall, S. S. Opat, and J. S. H. Gaston The Recognition of HLA-B27 by Human CD4+ T Lymphocytes J. Immunol., September 1, 2001; 167(5): 2619 - 2624. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |