| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Gwen Knapp Center for Lupus and Immunology Research, Committee on Immunology and Department of Pathology, and
Ben May Institute for Cancer Research, University of Chicago, Chicago, IL 60637; and
Cancer Biology Program, Hematology/Oncology Division, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA 02215
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
upon Ag stimulation. Analysis with various hCD1d-expressing
targets and use of Ag presentation inhibitors indicated the recognition
of hCD1d by CTLs did not involve species or tissue-specific ligands nor
require the processing pathways of endosomes or proteasomes.
Additionally, the reactivity of hCD1d-specific CTLs was not affected by
acid stripping followed by brefeldin A treatment, suggesting that CTLs
may recognize a ligand/hCD1d complex that is resistant to acid
denaturation, or empty hCD1d molecules. Our results show that hCD1d can
function as an alloantigen for CD8+ CTLs. The hCD1d Tg mice
provide a versatile model for the study of hCD1d-restricted cytolytic
responses to microbial Ags. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human group 1 CD1 can present lipid and glycolipid Ags derived from
mycobacterial cell wall to different subsets of T cells, including
CD4+ T cells (17),
CD8+ T cells (18), and
CD4-CD8- T cells
(19, 20). Group 2 CD1 can also bind lipid Ags, such as
glycosylceramide and phospholipids (21, 22). Additionally,
mCD1.1 has been shown to bind hydrophobic peptides as well
(23). Both hCD1d and mCD1.1 can be recognized by a unique
subset of T cells, the NK T cells, which express a restricted range of
TCRs bearing a single invariant V
-chain (V14J281 in mice and
V24JQ in humans) paired with limited sets of V
-chain
(24, 25, 26). NK T cells can recognize CD1d in the absence of
exogenous Ags, but their reactivity can be enhanced by the addition of
synthetic lipid Ags, such as -galactosylceramide (-GalCer)
(3) (21, 27). Upon activation, NK T cells
promptly produce large amounts of cytokines, in particular IL-4.
Several studies suggest that NK T cells may have important functions in
regulating immune responses (28, 29, 30, 31). In addition to NK T
cells, T cells expressing diverse TCR - and -chains have also
been found to recognize mCD1. These include some
CD4+ T cells from class II-deficient mice
(32, 33), some CD8+ CTLs from mice
immunized with plasmid DNA containing chicken OVA (34),
and from mice immunized with a mCD1 transfectant coated with
CD1-binding peptide (23, 35). However, no hCD1d-restricted
CD4+ or CD8+ CTLs have been
isolated to date.
Expression of two biochemically distinct forms of hCD1d,
2-microglobulin
(2m)-associated and
non-2m-associated hCD1d, has been demonstrated
(9, 12, 36). The 2m-associated
hCD1d is a mature 48-kDa glycoprotein, which is expressed on the
surface of thymocytes and B cells. The
non-2m-associated hCD1d is a 37-kDa
nonglycosylated isoform, which is predominately expressed on the IEC.
The functional role of hCD1d on IEC has been implicated by studies
showing that anti-hCD1d mAb inhibit IEC-induced proliferation of
CD8+ T cells (37). Recently, it was
also shown that cross-linking hCD1d on the surface of human IEC
with the anti-CD1d Ab specifically induced epithelial IL-10
expression, which may serve to dampen the epithelial proinflammatory
signals (38).
Little is known about the Ag presentation by hCD1d other than its
ability to present -GalCer to
CD4-CD8- or
CD4+ NK T cells. To further the study of hCD1d
functional properties, we have derived transgenic (Tg) mice expressing
hCD1d molecules. We report in this work that hCD1d behaves as a
transplantation Ag in mice, as observed in rapid rejection of skin
grafts from hCD1d Tg mice. Furthermore, hCD1d Tg spleen cells are
effective in inducing CD8+ hCD1d-specific CTLs in
normal mice. The hCD1d-specific T cells are capable of killing both
hCD1d-positive mouse and human cells, suggesting CTLs recognize hCD1d
as an intact molecule. Our data demonstrate that, similar to hCD1a, b,
and c, hCD1d could function as a restriction element for CTLs.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Genomic clone containing full-length hCD1D was
isolated from human genomic library (Stratagene, La Jolla, CA). The
chimeric gene in which hCD1D gene was driven by the
H2-Kb promoter was constructed as shown in Fig. 1
.
Chimeric DNA fragment was injected into the pronuclei of fertilized
eggs of (C57BL/6 (B6) x CBA)F1 mice to
produce Tg founder mice. Tg-positive mice were identified by PCR using
primers specific for hCD1D exon 2
(5'-CGAGGGCCCCACGCCGGGCGATA-3') and exon 3 (5'-CAGAGAGCGGACGGTGTCC-3').
Two lines of Tg mice, line 1 and line 3, were established by crossing
Tg founder mice with B6 mice. Line 1 was chosen for further
backcrossing with B6 mice because it showed higher surface expression
of hCD1d than line 3 (data not shown). The Tg mice used in this study
have been backcrossed four to seven generations onto B6 background.
|
RNA was extracted from various tissues of Tg mice with TRIzol reagent (Life Technologies, Grand Island, NY). cDNA was prepared using random hexamer primers, and amplified by PCR using primers specific for hCD1D exon 2 and exon 3. The amount of template cDNA used in each reaction was normalized to the amount of hypoxanthine phosphoribosyltransferase mRNA amplified with primers 5'-GTTGGATACAGGCCAGACTTTGTTG-3' and 5'-GAGGGTAGGCTGGCCTATAGGCT-3'.
Cell lines and CD1 transfectants
RMA-S and L929 cells were transfected with Kb/hCD1D chimeric gene by electroporation, followed by G418 selection and FACS analysis to generate lines stably expressing hCD1d. The derivation of RMA-S and L929 transfectants expressing mCD1.1 or mCD1.2, and C1R transfectants expressing CD1a, b, c, d, or CD1d/a chimeric protein have been described previously (26, 39). Human cell lines, U937, THP-1, K-562, Jurkat, MOLT-3, MOLT-4, Raji, and JY, were kindly provided by Gijs van Seventer (University of Chicago, Chicago, IL). All transfectants and human cell lines were maintained and cultured in RPMI 1640 medium (Life Technologies), supplemented with 10% heat-inactivated FCS (Sigma, St. Louis, MO), 2 mM L-glutamine, 100 U/ml penicillin and streptomycin, and 50 µM 2-ME (RPMI 10). Human intestine epithelial cell lines (HT29, T84, and CACO2) were kindly provided by Eugene Chang (University of Chicago). Con A- and LPS-induced blasts were prepared by incubating spleen cell suspension (5 x 106) with Con A (2.5 µg/ml; Sigma) or LPS (5 µg/ml; Sigma) for 72 h in RPMI 10. Bone marrow-derived macrophages were obtained by culturing bone marrow cells (2 x 105 cells/ml) for 6 days in RPMI 10 supplemented with 30% L929 cell supernatant.
Abs, cell preparations, and FACS analysis
The following Abs were purchased from PharMingen (San
Diego, CA): anti- CD11a (2D7), anti-CD102 (3C4), anti-CD54
(3E2), FITC anti-H2-Kb (AF6-88.5), FITC
anti-CD8 (53-6.7), PE anti-CD4 (RM4-5), PE anti-NK1.1
(PK136), FITC anti-H2-IAb (M5114),
biotinylated goat anti-mouse IgG2b, and biotinylated goat
anti-mouse IgG1. The hCD1d-specific Abs, 51.1 (mIgG2b) and 42.1
(mIgG1), have been described previously (9). PK136,
anti-mouse NK1.1; YTS.169.69, anti-mouse CD8; GK1.5,
anti-mouse CD4; 16-1-11N, anti-H2-Kk; Y3,
anti-H2-Kb; B22,
anti-H2-Db; BB7.2, anti-HLA-A2 (mIgG2b);
and 4D12, anti-HLA-B5 (mIgG1) were obtained from the American Type
Culture Collection (ATCC, Manassas, VA).
Single cell suspensions from thymus, spleen, lymph nodes, and Peyer’s patches were prepared using standard procedure and stained in immunofluorescence buffer (HBSS containing 2% FBS and 0.1% NaN3) using combinations of fluorescent-conjugated Abs for 30 min at 4°C. The IEC were prepared and purified through the discontinuous 25/40/70% Percoll gradient centrifugation, as described by Yamamoto et al. (40). Cells that layered between the 40 and 25% interface were collected as IEC. For isolating hepatic epithelial cells, mouse livers were perfused for 10 min with perfusion I medium (in mmol/L: NaCl, 120; KCI, 5; KH2PO4, 0.4; Na2HPO4, 0.2; NaHCO3, 25; EGTA, 0.5; D-glucose, 5.5; pH 7.4), then for 10 min with perfusion II medium (in mmol/L: NaCl, 120; KCI, 5; KH2PO4, 0.2; NaHCO3, 25; MgSO4, 0.4; MgCl2, 0.5; CaCl2, 3; D-glucose, 5.5; pH 7.4) containing 0.05% collagenase, 0.5 ml insulin (5 mg/ml), and 0.8 U trypsin inhibitor per unit tryptic activity in the collagenase (41). Then hepatocytes were minced, and separated on a Percoll density gradient described as above. Thymic stromal cell suspensions were prepared by digesting fetal thymi in 0.1% trypsin, 0.5 mM EDTA for 40 min at 37°C. Digestion was stopped by addition of immunofluorescence (IF) buffer. After mechanical disruption of the lobe, cells were harvested and washed two times with IF buffer before cell surface staining experiments. Cells (106) were stained with anti-hCD1d, followed by biotin-conjugated goat anti-mouse IgG2b and a third incubation with streptavidin-conjugated PE and FITC anti I-Ab. The stained cells were analyzed by flow cytometry using a FACSCalibur (Becton Dickinson, Mountain View, CA) with the CellQuest software.
Skin grafting
Female hCD1d Tg+ or
Tg- mice (6–8 wk old) were used as donors.
Full-thickness sections of skin (
1 x 1 cm in size) were
harvested from the tail of donors and grafted onto the dorsal side of
female C57BL/6. Bandages were removed on day 7 post transplant, and
grafts were monitored for 55 days for evidence of rejection. Rejection
was defined as complete necrosis of the skin grafts.
Generation of anti-hCD1d CTLs
B6 x CBA/F1 mice were primed with 107 irradiated hCD1d+ splenocytes (in B6 background) through i.p. injection and footpad injection. After 10 days, lymphocyte suspensions were prepared from draining lymph nodes and spleen, and then cultured with irradiated hCD1d+ splenocytes (2–5 x 106 cells/ml) in RPMI 10 medium. One week later, cultures were restimulated with hCD1d+ splenocytes and maintained in supplemented Mischell Dutton medium (SMDM) with IL-2 supplement (20 U/ml). IL-2 for restimulations was partially purified from the supernatant of EL4.IL2 cells (ATCC). After that, the CTLs were restimulated weekly with the irradiated L929/hCD1d transfectants. CTL clones were established by limiting dilution method in the presence of rIL-2 (10 U/ml; PharMingen) and 2.5 x 104 irradiated L929/hCD1d transfectants. The established clones were maintained by weekly stimulation with irradiated L929/hCD1d transfectants. The CTL activity was tested by 51Cr release assay, as described below.
CTL assay
One million target cells were labeled with 50 µCi [51Cr]sodium chromate (Amersham, Arlington Heights, IL) for 45 min at 37°C. Target cells (1 x 104 cells) were incubated with effector T cells in round-bottom microtiter wells. After 4 h at 37°C, 100 µl of supernatant from each well was collected and assayed for 51Cr release. The percentage of specific 51Cr release was calculated by the following equation: (experimental release - spontaneous release)/(maximal release - spontaneous release) x 100. For Ab-blocking studies, the CTL activity of clones against various targets was tested in the presence of 25% of supernatant or 30 µg/ml of purified mAbs.
Inhibitor treatment of target cells
Target cells were incubated with lactacystin (40 µM; Calbiochem, San Diego, CA), chloroquine (20 µM; Sigma), or brefeldin A (1 µg/ml, Sigma) for 18 h before CTL assay (42, 43). For the acid-stripping experiment, target cells (RMA-S/hCD1d, RMA, and P388) were washed with HBSS, and incubated for 90 s with acid-stripping medium (0.3 M glycine-HCl and 1% BSA in water, pH 2.4) at cell densities of 2 x 107 cells/ml (43). Culture medium (100x vol) was added to neutralize pH. Cells were washed three times and suspended in RPMI 10 at 106/ml. Completeness of acid stripping was assessed by flow cytometric analysis on RMA and P388 cells using anti-Kb and anti-Kd Ab, respectively.
Cytokine assay
CTLs (105 cells/well) were cultured with
the same number of irradiated stimulators in round-bottom 96-well plate
in a total volume of 200 µl/well. After 48 h, the supernatants
were harvested for cytokine assay. A sandwich ELISA was used to
determine the concentration of IFN-, IL-4, and IL-2. Abs specific
for cytokines and recombinant mouse cytokines were obtained from
PharMingen and used according to the manufacturer’s directions.
Cloning and sequencing of TCR genes
Total RNA was isolated from CTL clones using TRIzol reagent
(Life Technologies). First strand cDNA synthesis and PCR of dC-tailed
cDNA were conducted with 5' rapid amplification of cDNA end (RACE)
system (Life Technologies, Rockville, MD), according to the
manufacturer’s protocol. In brief, first strand cDNA synthesis was
done using Superscript II reverse transcription and C-specific
primer (5'-CAGGAGGATTCGGAGTCCCA-3') or C-specific primer
(5'-CCAGAAGGTAGCAGAGACCC-3'). The synthesized cDNA was then isolated
with GlassMax DNA Isolated Spin Cartridge, tailed with TdT, and
amplified by PCR. Oligonucleotide primers used were as follows:
abridged anchor primer (5'-ACTAGTACGGGIIGGGIIGGGIIG-3') and
C-inner primer (5'-CTGTCCTGAGACCGAGGATC-3') for TCR gene
amplification; and abridged anchor primer and C-inner primer
(5'-CCTGGGT GGAGTCACATTTCTC-3') for TCR gene amplification.
The PCR products were cloned into pGEM-T vector (Promega, Madison, WI).
Positive clones were screened by nested PCR with abridged anchor primer
pairing with the primer specific for either C
(5'-ACTGGTACACAGCAGGGTCTG-3') or C
(5'-CCTGGGTGGAGTCACATTTCTC-3'). Nucleotide sequences were determined
by PCR sequencing method using Taq DNA polymerase, BigDye
Terminator Cycle Sequencing Kit (PE Applied Biosystems, Foster
City, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The expression of hCD1D in the different tissues of Tg mice was
examined by RT-PCR. hCD1D message was detected from all tissues tested
from Tg animals, including thymus, spleen, lymph nodes, liver, kidney,
intestine, and skin (Figs. 1
and 2A). Cell surface expression
of hCD1d in the Tg mice was determined by immunofluorescence staining
with mAbs specific to hCD1d (mAb51.1) (Fig. 2B). High levels
of surface expression can be detected on the majority of lymphocytes
isolated from spleen and lymph nodes of the Tg animals, suggesting the
hCD1d epitope recognized by mAb51.1 was not affected by the species
origin of the associated 2m. A bimodal
staining pattern was observed from Tg thymocytes with high levels of
hCD1d expression on mature thymocytes (CD3high)
and low levels of hCD1d expression on immature thymocytes
(CD3low) (data not shown). Surface expression of
hCD1d can also be detected on epithelial cells from the thymus,
intestine, and liver (Fig. 2B). The expression pattern of
hCD1d in the Tg mice is similar to that of H2-Kb
molecules.
|
To determine whether hCD1d can serve as a transplantation Ag, B6
mice were grafted with tail skin from hCD1d Tg+
and Tg- littermate control mice. The donor mice
used in skin graft experiments were backcrossed seven generations onto
B6 background. All skin grafts (n = 10) from
hCD1d+ mice were rapidly rejected with a mean
survival time of 14 days. Five of six skin grafts from
hCD1d- mice showed no sign of rejection and
survived for at least 55 days. One of the hCD1d-
skin grafts was rejected at day 28, which might be due to the remaining
minor histocompatibility Ag disparities. These results reveal that
hCD1d molecules are recognized as functional transplantation Ags in
mice (Fig. 3).
|
To investigate whether hCD1d is capable of stimulating
hCD1d-restricted CTL response, we immunized B6 x
CBA/F1 (H-2b/k) mice with
hCD1d Tg+ splenocytes. The lymphocytes isolated
from the primed mice were stimulated with irradiated
Tg+ splenocytes in vitro. After 2 wk of culture
with Tg+ splenocytes
(H-2b background), the CTLs were
restimulated with H-2-mismatched hCD1d-transfected L929 cells
(H-2k background), to eliminate
H-2b-restricted CTLs. RMA-S/hCD1d
(H-2b) and L929/hCD1d
(H-2k) transfectants were used to screen
hCD1d-specific CTLs. Two CTL lines, BN1 and BN4, which lysed both
transfectants but not untransfected parental cells, were established
from two individual mice. Lysis of H-2-mismatched hCD1d-positive cells
indicated that these CTL lines recognized hCD1d as an intact molecule
and not as an hCD1d-derived peptide presented by a mouse MHC molecule
(Fig. 4). Inhibition of target cell lysis
by H2-Kb, Db, or hCD1d-specific mAb
showed that only the Ab to hCD1d exerted a complete inhibitory effect,
which further supports the notion that BN1 and BN4 CTLs recognized
hCD1d as a restriction molecule (Fig. 4).
|
FACS analysis of BN1 and BN4 showed that they are
CD8+CD4- and negative for
NK cell surface markers (data not shown), suggesting that they are
distinct from the NK T cell subset. Consistent with this finding, both
lines did not express invariant V14J281 transcript (data not
shown), the canonical TCR rearrangement found in most of the
mCD1-restricted NK T cells. Thus, we used 5'-rapid amplification of
cDNA end (RACE) protocol followed by DNA sequencing to determine the
TCR usage of CTL clones derived from line BN1 and BN4. All four clones
from line BN1 expressed V5J41 and V8.3D2J2.4, and all
eight clones derived from line BN4 expressed V17J25 and
V2D1J1.6. These CTL clones secreted significant amounts of
IFN- upon Ag stimulation, but not IL-2 or IL-4 (Table I). CTL clones derived from these two CTL
lines were used for further study.
|
Unlike some mouse and human NK T cells that can recognize CD1
molecules from both species (44), BN1 and BN4 CTLs did not
recognize mCD1.1- and mCD1.2-transfected RMA-S cells (Fig. 5). These CTLs lysed various cell types
derived from hCD1d+ Tg mice, including Con A
blasts, LPS blasts, kidney fibroblasts, and bone marrow-derived
macrophages, suggesting that the recognition of hCD1d by these CTLs did
not involve tissue-specific ligands. To test whether these CTLs can
recognize hCD1d expressed on human cells, we examined the reactivity of
these CTLs with a variety of human cell lines. Fig. 6A shows that BN1 and BN4
readily lysed several human cell lines, including MOLT-3, MOLT-4,
Jurkat, THP-1, and JY. An effective lysis of
hCD1d+ human cells by mouse
anti-hCD1d-specific CTLs suggested the recognition did not involve
a species-specific ligand. The degree of lysis by CTLs to different
cell lines largely correlated with the levels of hCD1d surface
expression on the cell lines, as detected by anti-hCD1d Ab, 51.1
(Fig. 6B). It is worth noting that both CTLs showed similar
reactive patterns toward various target cells, with the exception that
BN4 consistently had low reactivity to MOLT-4 and bone marrow-derived
macrophages from Tg mice.
|
|
To determine whether anti-hCD1d CTLs recognize other hCD1
molecules, human CD1A-, B-, C-, or D-transfected
C1R cells were used as targets. Both clones showed preferential
reactivity toward hCD1d transfectant (Fig. 7). Clone BN1 showed some
cross-reactivity to CD1c and CD1a transfectants, while BN4 did not
react with CD1a, b, c transfectants, even at the high E:T ratio (data
not shown).
|
shows that the hCD1d/a
transfectants could be killed as efficiently as the hCD1d
transfectants, suggesting that the recognition of these CTLs did not
involve endosomally derived ligands.
To gain insights into the cellular processes required to generate
epitopes recognized by these anti-hCD1d CTLs, inhibitors known to
interfere with discrete stages of Ag processing were used, specifically
for proteasomal degradation (lactacystin) and endosomal acidification
(chloroquine) requirements. As shown in Fig. 8A, neither lactacystin nor
chloroquine affected the recognition of hCD1d-specific CTLs, suggesting
that epitopes recognized by anti-hCD1d CTLs do not involve
proteasomally derived peptides and endosomally processed Ags. Acid
treatment of MHC class I+ target cells has been
shown to denature class I/peptide complexes, resulting in the loss of
recognition by CTLs (43, 48). However, treatment of
RMA-S/hCD1d transfectants with glycine/HCl (pH 2.4) followed by
brefeldin A (to minimize new hCD1d expression) did not significantly
diminish the killing by hCD1d-specific CTLs, suggesting that these CTLs
recognized either a ligand/hCD1d complex that is resistant to acid
denaturation, or empty hCD1d molecules. However, we cannot eliminate
the possibility that the number of remaining epitopes for CTLs in
acid-stripped target cells may still be above the threshold levels of
CTL recognition due to very high levels of epitope expression by
transfectants. To verify the efficacy of various treatments, parallel
experiments have been performed using CTLs specific to H2-M3, a mouse
class Ib molecule. Fig. 8B showed that lactacystin,
brefeldin A, and acid treatment plus brefeldin A effectively blocked
the allorecognition of M3-specific T cells.
|
Accessory molecules, such as ICAM/LFA-1, are known to play
important roles in hCD1d-restricted NK T cell-mediated cytotoxicity
(49), and therefore we examined the role of these
molecules in cytotoxicity mediated by hCD1d-restricted CTLs. The
requirement for accessory molecules by BN1 and BN4 was studied by Ab
blocking of the cytotoxic response against RMA-S/hCD1d and L929/hCD1d
transfectants (Fig. 9). The result
demonstrated that anti-CD11a (LFA-1) completely blocks the CTL
killing of RMA-S/hCD1d transfectants, but has no effect on L929/hCD1d
transfectants. Anti-CD54 (ICAM-1) and anti-CD102 (ICAM-2) also
partially blocked the lysis of RMA-S/hCD1d transfectants by both CTLs.
The differential effect of anti-LFA-1 and anti-ICAMs Abs on
hCD1d-transfected RMA-S (LFA-1+,
ICAMs+) and L929 cells
(LFA-1+, ICAMs-)
correlated with the expression levels of LFA-1 and ICAMs on these two
target cell lines (data not shown). This result indicates that an
LFA-1/ICAM interaction is critical for the cytotoxicity of BN1 and BN4
for hCD1d-expressing T cell line, but not fibroblast cell lines, and
the requirement for accessory molecules by BN1 and BN4 to exert CTL
activity is similar. Although these CTL clones express CD8 on their
cell surface, anti-CD8 Ab did not block the CTL response,
suggesting the affinity of the TCRs and hCD1d is sufficient enough to
trigger the CTL response.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Most of the allo-specific CTLs to MHC class I molecules recognize
epitopes that are dependent on both MHC molecules and specifically
bound peptides (51, 52, 53, 54). In contrast, our data showed that
the recognition of hCD1d by anti-hCD1d CTLs may not involve a
specific ligand, as hCD1d-specific CTLs can react with various
hCD1d-expressing cells of either mouse or human origin. Thus, the
hCD1d-restricted CTLs may recognize empty hCD1d molecules or,
alternatively, the target could be a complex of hCD1d and a broadly
distributed, conserved cellular Ag. If both BN1 and BN4 CTLs recognized
empty hCD1d, one would expect that the reactivity of CTLs to different
targets would entirely depend on the surface expression levels of hCD1d
on the target cells; furthermore, the reactivity patterns of both CTLs
against varied hCD1d-expressing cells should be the same. Although this
was the case to a certain extent, differential reactivity between BN1
and BN4 to some hCD1d+ target cells was detected.
For example, BN4 had lower reactivity to hCD1d+
bone marrow-derived macrophages and MOLT-4 cell line than BN1 at all
ranges of E/T ratio tested (Figs. 5 and 6). Our Ab-blocking experiment
indicated the requirement for accessory molecules is similar between
BN1 and BN4 (Fig. 9). Therefore, it is most likely that differential
activity between BN1 and BN4 may depend on the relative abundance of
the conserved cellular ligand(s) for hCD1d on different cells. Because
the recognition of hCD1d-restricted CTLs is resistant to acid
denaturation and independent of TAP and proteasomal activity, it is
possible that anti-hCD1d-specific CTLs may recognize nonpeptide
Ags, probably cellular lipid Ags, in the context of hCD1d. This notion
is supported by the recent finding that some mCD1-restricted T cells
recognized cellular phospholipids in a CD1-dependent manner
(55).
The response of hCD1d-specific CTLs to hCD1d-expressing target cells
was not enhanced by addition of -GalCer or glycosylation variants of
ceramides (data not shown). These results indicated that hCD1d-specific
CTLs have different ligand specificity from CD1d-restricted NK T cells.
This is consistent with the finding that the hCD1d-specific CTLs did
not express characteristic invariant TCR found in CD1-restricted NK T
cells. Our analysis of TCR sequences of hCD1d-specific CTLs showed that
all CTL clones derived from each individual mouse have identical DNA
sequences. The homogeneity of CTL population from each individual mouse
may be due to the limited TCR repertoire against hCD1d molecules and/or
preferential clonal expansion during in vitro restimulation.
Although the expression of hCD1d on human IEC lines, HT29, T84, and
CACO2, has been demonstrated (56), we found that
hCD1d-specific CTLs fail to recognize the target Ag expressed on these
IEC lines (data not shown). It has been shown that the majority of
hCD1d on IEC is nonglycosylated and non-2m
associated (36). Our data suggested that hCD1d-specific
CTLs could distinguish the different conformational state of hCD1d
molecules.
Tg mice expressing HLA class I and class II molecules have been used to provide a suitable animal model for the study of the functions of HLA molecules. The ability of hCD1d to serve as a restriction element in allorecognition for CD8+ CTLs suggested that similar to group 1 CD1, hCD1d might present unique microbial Ags to CTLs. We are attempting to challenge the hCD1d Tg mice with various bacterial pathogens and examine whether the hCD1d might play a role as a restriction element for the microbial Ags in vivo. Crossing hCD1d Tg mice onto CD1-deficient background will further provide an animal model to study the functional role of hCD1d in T cell development and immune response against infectious disease.
|
Acknowledgments |
|---|
-GalCer; Dr. Kistern Fischer Lindahl (University of Texas
Southwestern Medical Center, Dallas, TX) for providing anti-M3
CTLs; Hanh Nguyen for critical reading of the manuscript; and Miriam
Fay for technical assistance. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Chyung-Ru Wang, University of Chicago, Gwen Knapp Center for Lupus and Immunology Research, Room 412, 924 East 57th Street, Chicago, IL 60637-5420.
3 Abbreviations used in this paper: hCD1, human CD1; -GalCer, -galactosylceramide; 2m, 2-microglobulin; IEC, intestinal epithelial cell(s); m, mouse; Tg, transgenic.
Received for publication October 11, 2000. Accepted for publication January 4, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-microglobulin-dependent surface expression of functional mouse CD1.1. J. Exp. Med. 182:1913.2-microglobulin-associated form. Int. Immunol. 11:383. T cell pool. J. Immunol. 162:366.+ T cells. Nature 372:691.[Medline]
14 NKT cells by glycosylceramides. Science 278:1626. chain is used by a unique subset of major histocompatibility complex class I-specific CD4+ and CD4-CD8- T cells in mice. J. Exp. Med. 180:1097.24+ CD4-CD8- T cells. J. Exp. Med. 186:109.-galactosylceramide directs conventional T cells to the acquisition of a Th2 phenotype. J. Immunol. 163:2373.-galactosylceramide polarizes CD1-reactive NK T cells towards Th2 cytokine synthesis. Eur. J. Immunol. 29:2014.[Medline]
2-microglobulin-independent MHC class Ib molecule expressed by human intestinal epithelium. Science 265:259.-galactosylceramide by natural killer T cells is highly conserved through mammalian evolution. J. Exp. Med. 188:1521. chain in the T cell antigen receptor complex by /
T cells localized in epithelia. J. Exp. Med. 179:365.-glycosylceramide-pulsed monocyte-derived dendritic cells. J. Immunol. 164:4458. modulates CD1d surface expression on intestinal epithelia. Am. J. Physiol. 271:C276.This article has been cited by other articles:
|
|
 |
|
  G. Sarikonda, H. Wang, K.-J. Puan, X.-h. Liu, H. K. Lee, Y. Song, M. D. Distefano, E. Oldfield, G. D. Prestwich, and C. T. Morita Photoaffinity Antigens for Human {gamma}{delta} T Cells J. Immunol., December 1, 2008; 181(11): 7738 - 7750. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Toyofuku, Y. Yasunami, K. Nabeyama, M. Nakano, M. Satoh, N. Matsuoka, J. Ono, T. Nakayama, M. Taniguchi, M. Tanaka, et al. Natural Killer T-Cells Participate in Rejection of Islet Allografts in the Liver of Mice Diabetes, January 1, 2006; 55(1): 34 - 39. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. Lipscomb and B. J. Masten Dendritic Cells: Immune Regulators in Health and Disease Physiol Rev, January 1, 2002; 82(1): 97 - 130. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
