Phenotypic and Functional Characterization of Long-Term
Cultured Rhesus Macaque Spleen-Derived NKT Cells1
Balgansuren Gansuvd,* William J. Hubbard,* Anne Hutchings,* Francis T. Thomas,*
Jeanine Goodwin,* S. Brian Wilson,†Mark A. Exley,‡and Judith M. Thomas2
Natural killer T cells are immunoregulatory cells, which have important roles in tolerance and autoimmunity, as demonstrated
primarily in mice and humans. In this study, we define the phenotype and function of V?24?T cells derived from the spleens of
rhesus macaques, a species increasingly used in models of immune tolerance. V?24?cells were isolated and expanded with
monocyte-derived immature dendritic cells in the presence of ?-galactosylceramide, IL-2, and IL-15. Rhesus NKT cells were
stained with mAbs against both V?24 and the invariant complementarity-determining region 3 epitope of the human V?24/J?Q
TCR. The cells were CD4, CD8 double negative and expressed CD56. Rhesus NKT cells also exhibited moderate to high expression
of CD95, CD45RO, CD11a, and ?7integrin, but did not express CD45 RA, CD62L, CCR7, CD28, and other activation, costimu-
latory molecules (CD69 and CD40L). By intracellular staining, >90% of unstimulated rhesus NKT cells expressed IL-10, but not
IFN-?. However, the latter was strongly expressed after stimulation. Rhesus NKT secreted large amounts of TGF-?, IL-13, and
IL-6, and modest levels of IFN-?, whereas IL-10 secretion was negligible and no detectable IL-4 was observed either intracellularly
or in culture supernatants. Functionally, the NKT cells and their supernatants suppressed T cell proliferation in allogeneic MLR.
We conclude that long-term cultured rhesus macaque spleen-derived V?24?T cells are semi-invariant double-negative cells with
effector memory phenotype. These cells are semianergic, polarized to a uniquely Th3 > T regulatory-1 regulatory cell phenotype,
and have regulatory/suppressive function in vitro. The Journal of Immunology, 2003, 171: 2904–2911.
3), NK1.1?(CD161?) T cells (4, 5), CD1d-restricted T cells (6–
8), and CD3?CD56?NK T cells (9). The relationships between
these overlapping populations are only now becoming clear. NKT
are a subpopulation of ?? T cells, many of which coexpress an
invariant V?14 or V?24 TCR in mouse and human, respectively,
along with other NK cell markers (1, 10). Although invariant NKT
cells have highly restricted V? and J? ?CR usage, their V? TCR
repertoire is less stringently defined (11). The invariant NKT cells
recognize the nonclassical MHC class I molecule, CD1d (6),
which presents ?-galactosylceramide (?-GalCer) to NKT cells
(12). Unlike invariant NKT cells, not all CD1d-restricted T cells
have a limited TCR repertoire, the latter using a diverse variety of
TCR V? and V? chains other than V?24/V?14 and V?11/V?8
(13–15). Moreover, not all invariant NKT cells express CD161,
while it is expressed by a subset of classical T cells (16). In con-
trast to NK cells, invariant NKT cells do not express CD16 and
CD57, while CD94 (7, 17) is variably expressed. The majority of
everal types of NKT cells have been reported. These in-
clude human V?24?, V?11?, and J?Q?, and mouse
V?14?, V?8, 2, 7?, and J?281?invariant NKT cells (1–
invariant NKT cells are CD4?CD8?double negative (DN),3with
the remaining populations being CD4?or CD8???, and CD3?
CD56?NKT cells that express the CD8?? homodimer (18).
NKT cell frequency and phenotype vary with tissue distribution.
The percentage of murine CD1d-reactive NKT cells is highest in
the liver (30–50%), with substantial numbers also present in bone
marrow (BM, noninvariant 20–30%). There are fewer CD1d-re-
active NKT cells in blood (4%) and spleen (3%), and still fewer in
lymph nodes (0.3%) and thymus (0.3–0.5%) (reviewed in Ref.
19). NKT cell phenotype and function also vary in relation to
tissue distribution. Murine thymic and liver NKT cells are CD1d
dependent and exhibit a CD4?or DN phenotype with high ex-
pression of CD69. Spleen and BM NKT cells are partially CD1d
independent and rarely express the invariant TCR, and most are
DN or CD8?with high expression of CD62 (20, 21). These find-
ings suggest murine CD1d-dependent NKT cells are activated and
segregate mainly to thymus and liver, whereas CD1d-independent
NKT cells resemble recirculating naive T cells. In addition, NKT
cells, either human umbilical cord blood or adult peripheral blood,
display an activated memory phenotype (22).
NKT cells are involved in immune modulation. They produce
both Th1- and Th2-type cytokines and exhibit suppressive activity
in allogeneic MLR responses (13, 21). Murine NKT cells inhibit
growth of certain tumors (23, 24), participate in the immune re-
sponse to nonpeptide mycobacterial components (25) and virus
infections (26, 27), and suppress acute lethal graft-vs-host disease
(28). CD1d-dependent NKT cells have also been implicated in an
anterior chamber-associated immune deviation model (29) and in
allotransplant tolerance (30).
*Department of Surgery, Division of Transplantation Immunobiology, University of
Alabama, Birmingham, AL 35294;†Diabetes Unit, Massachusetts General Hospital,
Harvard Medical School, Boston, MA 02215; and‡Cancer Biology Program, Hema-
tology/Oncology Division, Beth Israel-Deaconess Medical Center, Harvard Medical
School, Boston, MA 02215
Received for publication January 15, 2003. Accepted for publication July 14, 2003.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1This work is supported by National Institute of Diabetes and Digestive and Kidney
Diseases Grant 1U19DK57958 to J.M.T.
2Address correspondence and reprint requests to Dr. Judith M. Thomas, 564 Boshell
Diabetes Building, 1808 7th Avenue South, Birmingham, AL 35294. E-mail address:
3Abbreviations used in this paper: DN, double negative; ?-GalCer, ?-galactosylce-
ramide; BM, bone marrow; CDR3, complementarity-determining region 3; DC, den-
dritic cell; iDC, immature DC; MNC, mononuclear cell; rh, recombinant human; Tr1,
T regulatory-1; TSP, thrombospondin.
The Journal of Immunology
Copyright © 2003 by The American Association of Immunologists, Inc.0022-1767/03/$02.00
The mechanism of immune regulation by NKT cells remains
poorly defined, possibly related to differences in cytokines pro-
duced by NKT subsets. Many invariant NKT react with CD1d
tetramers, and CD4?CD1d tetramer?cells produce both Th1 and
Th2 cytokines, but are relatively biased to Th2, whereas CD4?
cells produce IFN-? and TNF-? (16, 17). Human BM-derived
CD1d-reactive NKT cells that suppress MLR also have a Th2 bias
(13). Costimulatory pathways involving CD28-CD80/CD86 and
CD40-CD154 may contribute to the regulation of Th1 and Th2
cytokine functions of NKT cells (31). For example, type 2 den-
dritic cells (DC2) can enhance the development of neonatal NKT
cells into NKT2 cells, which repolarize toward NKT1 in the pres-
ence of DC1 cells (32). Thus, NKT cytokine polarization probably
depends on coreceptor expression as well as the type of APC
To date, detailed studies of NKT cells have not been reported in
nonhuman primates, the penultimate preclinical model for inter-
ventional research in AIDS and organ transplantation tolerance. In
this study, we provide the first description of rhesus macaque
spleen-derived NKT cells and characterize their suppressive effect
Materials and Methods
Tissues from three normal, male rhesus macaques (Macacca mulatta), who
were selected for islet transplantation, were studied. All were free of
known pathogens. Surgical procedures were performed in accordance with
the National Institutes of Health Guide for the Care and Use of Primates
under supervision of the University of Alabama at Birmingham Institu-
tional Animal Care and Use Committee.
Preparation of mononuclear cells (MNCs)
Spleen cells were prepared from tissue specimens of normal donor rhesus
macaques. Spleens were cut into small pieces and minced gently on metal
mesh filter (75 ?m diameter). The filtered cells were washed twice and
suspended into RPMI 1640 (Life Technologies, Gaithersburg, MD) with
100 U/ml penicillin and 100 ?g/ml streptomycin (Life Technologies) con-
taining 10% heat-inactivated FCS. The MNCs were isolated by Ficoll-
Hypaque density (G ? 1.077 g/dl) gradient centrifugation at 1500 rpm for
CD14?cells were isolated from MNCs using MACS column (Miltenyi
Biotec, Auburn, CA) after staining with Microbeads conjugated with
mouse anti-human CD14 mAb (Miltenyi Biotec). Monocytes were cultured
in RPMI 1640 supplemented with 10% FCS in the presence of 800 U/ml
human rGM-CSF (rhGM-CSF) (R&D Systems, Minneapolis, MN) and 10
ng/ml rhIL-4 (R&D Systems) for 4 days at 37°C in a humidified atmo-
sphere with 5% CO2. Cells were washed with PBS containing 0.05% BSA
and 2% EDTA before use.
Enrichment and expansion of V?24?NKT cells
MNCs (1 ? 108) isolated from rhesus macaques, as described above, were
incubated with FITC-conjugated mouse anti-human V?24 mAb (Beckman
Coulter, Kansas, MO), followed by anti-FITC MACS beads (Miltenyi Bio-
tec), and applied onto a MACS column. The positively selected cells (3–
5 ? 105) were stimulated by irradiated (2300 rad) autologous CD14?cells
pulsed with ?-GalCer (100 ng/ml; Kirin Brewery, Maebashi, Cunma, Ja-
pan) in the presence of rhIL-2 (100 U/ml; R&D Systems) and rhIL-15 (10
ng/ml; Biosource International, Camarillo, CA). From the third week, the
NKT cells were stimulated by allogeneic monocyte-derived immature DCs
(iDC), pulsed with ?-GalCer in the same cytokines, as mentioned above.
The cytokines and ?-GalCer were added every 2–3 days.
Anti-human mAbs known to cross-react with rhesus macaques were se-
lected for this study. These included mAbs V?24 FITC (C15; Immunotech,
Marceille Cedex, France); V?11 PE (C21; Immunotech); CD8?? PC5
(B9.11; Immunotech); CD8? PE2ST8 (5H7; Immunotech); mAb clone
6B11-biotin, recognizing complementarity-determining region 3 (CDR3)
of V?24?NKT cells (33); CD3? FITC (SP34; BD PharMingen, San Di-
ego, CA); CD4 FITC (M-T477; BD PharMingen); CD161 FITC (DX12;
BD PharMingen); CD56 PE (MY31; BD PharMingen); CD45RA FITC
(5H9; BD PharMingen); CCR7 unconjugated (2H4; BD PharMingen);
CD28 PE (CD28.2; BD PharMingen); CD69 FITC (FN50; BD PharMin-
gen); CD154 FITC (TRAP1; BD PharMingen); CD25 FITC (M-A251; BD
PharMingen); CD122 PE (Mik?2; PharMingen); CD11a FITC (G25.2; BD
Biosciences, San Jose, CA); ?7integrin PE (FIB504; BD Biosciences);
CD62L PE (SK11; BD Biosciences); CD45RO biotin R-PE (UCHL1; An-
cell, Bayport, MN); CD1d unconjugated (CD1d55; kindly provided by S.
Porcelli, Albert Einstein College of Medicine of Yeshiva University,
Bronx, NY). For intracellular staining, anti-IFN-? FITC (MD-1; Bio-
source); anti-IL-4 FITC, PE (860A 4B3; Biosource); and anti-IL-10 PE
(JES3-9D7; Caltag Laboratories, Burlingame, CA) mAbs were used. FITC-
conjugated goat anti-mouse IgG (BD Biosciences) was used as a secondary
Ab for the anti-human CD1d mAb.
Cell surface phenotyping by flow cytometry
Before staining, the cells were washed and resuspended in PBS supple-
mented with 1% BSA and 0.01% NaN3, and subsequently incubated with
the mAbs for 30 min on ice. After washing twice, 2 ? 104labeled cells
were subjected to two- or three-color FACScan flow cytometry using an
Epics Elite Analyzer (Beckman Coulter).
For intracellular IFN-?, IL-10, and IL-4 staining, 2–5 ? 105NKT cells
were stimulated for 4 h at 37°C with 20 ng/ml PMA (Sigma-Aldrich, St.
Louis, MO) and 1 ?g/ml ionomycin (Sigma-Aldrich) in 1 ml of complete
medium containing the intracellular transport inhibitor brefeldin A (10 ?g/
ml; Sigma-Aldrich). The cells were then washed with staining buffer
(0.05% BSA/EDTA in PBS) and permeabilized with FACS permeabilizing
solution (BD Biosciences) for 10 min at room temperature. These cells
were resuspended in staining buffer and incubated for 30 min on ice in the
dark with FITC-conjugated anti-IFN-?, PE-conjugated anti-human IL-10,
or FITC/PE-conjugated anti-human IL-4 mAbs. Unbound Abs were re-
moved by two washes with staining buffer. Finally, the cells were resus-
pended in PBS and analyzed by flow cytometry.
Measurement of cytokines in culture supernatants
NKT cells (106/well) were cultured in a 24-well plate (Corning, Corning,
NY) with ?-GalCer-pulsed iDCs (105/well) in the presence of rhIL-2 and
rhIL-15 for 24, 48, 72, or 96 h. Cell-free supernatants were collected from
each well and stored at ?80°C before analysis. IFN-?, IL-10, and IL-4
concentrations in culture supernatants were determined by ELISA kit (Bio-
source), according to the manufacturer’s recommended procedure.
MNCs (6 ? 104/well) were cultured in a 96-well round-bottom plate with
irradiated mature DCs (103/well) and PMA-stimulated (20 ng/ml, for 4 h)
or unstimulated NKT cells (3 ? 104/well, 1.2 ? 104/well, or 6 ? 103/well).
After 6 days, the cells were labeled with 1 ?Ci/well of [3H]thymidine (Life
Science Products, Boston, MA) for 16 h and harvested with an automated
cell harvester (Tomtec, Orange, CT). Incorporation of the radioisotope was
measured by a liquid scintillation counter (Pharmacia Wallac, Turku,
V?24?rhesus macaque NKT cells use the same CDR3 region,
but different V? TCR than human invariant NKT cells
We first asked whether rhesus macaque NKT cells express the
same V? and V? invariant TCR and CDR3 regions that are used
by human invariant NKT cells. Using anti-human V?24 and anti-
human V?11 mAbs that are cross-reactive with rhesus monkeys,
we examined expression in rhesus peripheral blood, spleen, and
culture-expanded cells. In fresh normal peripheral blood MNCs,
and in freshly obtained normal spleen cells, only a small percent-
age of cells exhibited either V?11 (MNC mean 0.3 ? 0.14 SD,
n ? 7; spleen mean 0.25 ? 0.1 SD, n ? 4) or V?24 expression
(PBL mean 0.27 ? 0.12 SD; spleen mean 0.15 ? 0.05 SD). The
percentage of V?11/V?24 double-positive T cells among the
V?24?fresh spleen cells was also low (mean 1.0 ? 0.7 SD, n ?
3). After 3-wk expansion in culture, purified V?24?spleen cells
2905The Journal of Immunology
expressed a high (65%) percentage of V?24?cells, but relatively
few (4%) V?11?cells. In longer term cultures (?2 mo), in which
the cells were stimulated weekly with ?-GalCer-pulsed iDC in the
presence of rhIL-2 and rhIL-15, expression of V?24 was nearly
homogenous (?98%), while V?11 staining was less than 1% (Fig.
1a). Most (89%) of these V?24?cells also stained with mAb
6B11, which is specific for the CDR3 region of V?24/J?Q TCR
(Fig. 1b) and binds to human invariant NKT cells (33). Rhesus
V?24?cells also specifically proliferated in response to ?-GalCer-
pulsed CD1d-transfected HeLa cells, but not to ?-GalCer-
pulsed, mock-transfected CD1d?control HeLa cells (data not
shown). These data indicate that cultured rhesus NKT cells are
semi-invariant, expressing V?24 and invariant CDR3 epitopes,
but not V?11.
The expression of T and NK cell markers by rhesus macaque
It has been previously demonstrated that most V?24?invariant
NKT cells display a DN or CD4?, but rarely a CD8?phenotype.
We found that rhesus V?24?invariant TCR was coexpressed with
CD3? (Fig. 2a). CD8?? and CD4 were essentially undetectable
(Fig. 2, c and d). However, a high frequency of these cells ex-
pressed CD8???(Fig. 2b, mean 53 ? 10 SD, n ? 4), which may
have been up-regulated due to culture in the presence of IL-2 and
IL-15 (18). Hereafter, we refer to these rhesus NKT cells as CD4/8
DN because they were negative for CD4 and CD8?.
Earlier studies have reported that most invariant NKT cells in
human and mouse coexpress NK cell markers, such as CD161 (7,
10, 34). CD161 is also expressed by many other human T cells (4).
Before testing rhesus spleen cells with anti-human CD161 re-
agents, the available mAbs were validated for binding activity to
freshly isolated human CD16?NK cells (mean 19.9 ? 0.3 SD;
n ? 3), CD3?T cells (mean 3.8 ? 1.9 SD; n ? 6), and long-term
(?1 mo) cultured CD3?T cells (mean 37.5 ? 3.8 SD; n ? 3).
Normal rhesus PBL stained with anti-CD161 FITC (clone DX12)
exhibited a small number (mean 0.4 ? 0.2 SD, n ? 6) of positive
cells, although both short- and long-term cultured NKT cells were
consistently negative. In contrast, as shown in Fig. 3, ?80% of
rhesus V?24?CD3?cells were positive for CD56, a molecule
commonly expressed on human NK cells (17). These results show
that rhesus macaque spleen-derived NKT cells are mostly CD4/8
DN and preferentially coexpress CD56.
Central and effector memory phenotype expression in NKT cells
derived from rhesus macaque spleen
The tissue distribution of invariant NKT cells has been reported to
affect the phenotype of these cells. Thymus- and liver-derived
cells as well as V?24?invariant NKT cells. Most cells were double pos-
itive for a, CD56 PE and V?24 FITC (NK-NKT phenotype) and also for
CD56 PE and fn18 SP/PerCP (NK-CD3 phenotype). Data represent three
Rhesus macaque spleen-derived cells are classical NKT
Spleen-derived NKT cells were positive for V?24 (FITC), but did not express
V?11 in any time point during long-term culture. b, Most of the NKT cells
were stained by 6B11 (PE), which recognizes CDR3 region of V?24/J?Q
TCR. Representative data from three independent experiments are shown.
Rhesus macaque NKT cells express semi-invariant TCR. a,
display DN phenotype. a, Most of NKT cells expressing V?24 (PE) were
positive for fn18 (FITC), later of which recognizes rhesus monkey CD3
Ag. b, More than half of V?24?cells (PE) expressed CD8?? (SP/PerCP).
c, Only 2% of gated V?24?cells were positive for both CD8?? (SP/
PerCP) and CD8? (PE) heterodimers. d, The V?24?(PE) NKT cells did
not express CD4 (FITC). Data are shown representative of three to five
Rhesus macaque spleen-derived V?24?NKT cells mainly
2906RHESUS MACAQUE NKT CELLS
NKT cells in mice have been described as memory phenotype-
positive cells, whereas spleen- and bone marrow-derived murine
NKT cells displayed a naive phenotype (20). Human PB and cord
blood NKT cells express an activated memory phenotype (22).
Freshly obtained rhesus NKT cells expressed a primed memory
phenotype, but with variable levels of expression. The uncultured
spleen cells, gated for the V?24?NKT cell population, strongly
expressed CD11a (mean 78.3 ? 11.4 SD, n ? 3). Two of three
rhesus were positive for CCR7 expression (mean 40.4 ? 17.6 SD),
but negative for CD62L (mean 0.7 ? 0.5 SD), CD28 (mean 4.9 ?
2.1 SD), CD45RA (mean 0.2 ? 0.2 SD), ?7integrin (mean 1.9 ?
1.9 SD), and CD95 (mean 2.3 ? 2.3 SD). One of three displayed
strong to moderate expression of CD62L, ?7integrin, CD45RA,
CD95, and CD28. The expression of CD45RO was consistently
positive, but of varying intensity. Two of three animals expressed
a high percentage of CD45RO?cells (27.3 and 75.8%), while one
expressed only 5.7% cells. None expressed CD69. In the same two
CD45RO?animals, both expressed high levels (?80%) of
CCR7?cells in the V?24?population, while the animal with low
CD45RO also expressed low CCR7 (3.7%). These results suggest
that most rhesus V?24?fresh spleen NKT cells have a primed,
central memory phenotype.
In the long-term cultures, there was relatively homogeneous ex-
pression of CD45RO, CD95, the adhesion molecule CD11a, and
the mucosal homing receptor ?7integrin, indicating retention of a
memory phenotype (Fig. 4a). In contrast, there was a consistent
lack of expression of the secondary lymphoid tissue migration
molecules, CD62L, CCR7, and also CD28, characteristic of the
naive phenotype (35). These results are consistent with expression
of the effector memory phenotype in long-term cultured spleen-
derived NKT cells of rhesus macaques.
Spleen-derived NKT cells exhibit semianergic phenotype in
regular culture conditions
CD69, an early activation marker of T and NK cells, is constantly
expressed on the NKT cell surface in mice and humans (17, 19) In
addition, a high expression of CD25 (IL-2R?) has been observed
in human cord blood NKT cells (22). Under regular culture con-
ditions, rhesus spleen-derived NKT cells, in contrast, did not ex-
press CD69, CD40L, but these markers were up-regulated after
PMA and ionomycin stimulation (Fig. 5, a–d). Intracellular pro-
duction of IL-2 (Fig. 5, e and f) and expression of CD25 (IL-2R?)
and CD122 (IL-2R?) failed to increase after stimulation with
PMA/ionomycin (data not shown). Moreover, the NKT cells grew
slowly, even though they were stimulated weekly with ?-GalCer-
pulsed DC in the presence of rhIL-2 and rhIL-15. This has previ-
ously been reported to be an optimal condition for expansion of
human NKT cells (18). These results suggest that rhesus spleen-
derived NKT cells are semianergic under regular culture
Th3/T regulatory-1 (Tr1)-polarized, cytokine profile
For functional analysis, we harvested NKT cells 4 days after
weekly stimulation with ?-GalCer-pulsed DC, rhIL-2, and rhIL-
15, typically a time of peak activation for NKT cells (36, 37).
Although the NKT cells displayed a semianergic phenotype,
?95% of unstimulated NKT cells expressed intracellular IL-10
(Fig. 6a), but no detectable IL-4 and IFN-? (data not shown). After
stimulation with PMA and ionomycin, ?90% of the NKT cells
coexpressed both IL-10 and IFN-? (Fig. 6b). However, IL-4 ex-
pression was still negligible (not shown). The possibility of non-
specific staining with anti-human IL-10 mAb was excluded by the
staining of positive and negative control cells and inhibition of
positive staining by rhIL-10 (Fig. 6, c and d). In supernatants of
NKT cells, large amounts of TGF-?, IL-13, and IL-6 were de-
tected, but only low levels of IL-4 and IL-10 were observed, using
ELISA reagents that efficiently detect rhesus IL-4 and IL-10 in our
experience (38) (Fig. 7). Culturing the NKT cells with ?-GalCer
increased the production of IL-13 and IL-10 by 1.4- to 2.6-fold,
compared with cultures without ?-GalCer (data not shown). The
cytokine production profile of anergic T cells following stimula-
tion with high doses of cross-linked anti-CD3 mAb is distinct from
Th1, Th2, or Th0 cells. Tr1/Th3 cells are typically anergic (39, 40),
and their most striking feature is low proliferative capacity, un-
usually high levels of IL-10 and TGF-?, and significant amounts of
IFN-?, IL-5, but no IL-4 or IL-2 (41–43). In this context, we
showed high to moderate expression of adhesion molecule CD11a, ?7integrin, and memory cell markers CD45RO and CD95. b, The absence of CD Ags
characteristic of naive cells is consistent with an effector memory phenotype. The cells were stained with appropriate mAb or isotype control Ab, described
in Materials and Methods, and analyzed by flow cytometry. Open histograms represent isotype control staining. Representative data from three independent
experiments are shown.
Spleen-derived NKT cells show effector memory phenotype. a, Histogram profiles gated on lymphocyte forward/side scatter properties
2907The Journal of Immunology
suggest that rhesus macaque spleen NKT cells mimic regulatory-
like cells in our culture conditions.
Spleen-derived NKT cells suppress allogeneic T cell responses
Next, we investigated how the resting and activated NKT cells
regulate T cell responses in allogeneic MLR. Both unstimulated/
resting NKT cells and also PMA-stimulated NKT cells consis-
tently suppressed allogeneic T cell responses in a dose-dependent
manner. To elicit strong MLR responses, we used DCs from MHC
class II DRB1-mismatched donors as a source of APC. As shown
in Fig. 8, unstimulated NKT cells strongly inhibited (43%) T cell
responses at a responder/NKT ratio of 2:1, while the inhibition was
decreased from 21 to 13% at a responder/NKT ratio of 5:1 and
10:1, respectively. In contrast, using PMA-stimulated NKT cells,
the strongest inhibition (47%) of T cell responses was found at a
responder/NKT ratio of 10:1, and was dose dependent, decreasing
from 36 to 28% at a responder/NKT ratio of 5:1 and 2:1, respec-
tively. Supernatants from both unstimulated and stimulated NKT
cells suppressed T cell proliferation to a lesser extent than the
strongest inhibition of NKT cells. Taken together, these results
indicate that rhesus macaque spleen-derived NKT cells can be in-
duced as regulatory cells with suppressive function.
NKT cells have been studied in humans, mice, and to a lesser
extent rats, but, to our knowledge, NKT cells have not been de-
fined in nonhuman primates. In this study, we obtained NKT cells
from the spleens of rhesus macaques, expanded them in vitro, and
characterized their phenotype and function. Due to the high ho-
mology between humans and rhesus macaques, many anti-human
mAbs cross-react with rhesus CD Ags, including those reported in
this work. The cross-reactivity of each mAb was confirmed on at
least three healthy monkeys before using them in additional ex-
periments. Our results indicate that rhesus spleen cells, positively
selected for V?24, consistently expressed the invariant CDR3
epitope, which is also expressed on human invariant NKT cells
(33). However, unlike their human counterparts, rhesus V?24?
NKT cells highly expressed CD56 and did not express V?11.
Discrimination of naive and memory phenotypic traits can pro-
vide insight into the biological function of NKT and conventional
cells. It is known that naive T cells express high levels of L-se-
lectin (CD62L) and CCR7 that promote their recruitment via high
endothelial venules into lymph nodes. In contrast, memory T cells
lose these two receptors, and instead express high levels of inte-
grins (CD11a and ?7) and chemokine receptors that promote re-
cruitment into nonlymphoid tissues (44). The central memory cells
recirculate through lymph nodes, where they encounter Ag.
Primed central memory cells (CD45RO?CCR7?) are thought to
represent a precursor population of further differentiated effector
memory cells (CD45RO?, CCR7?) that have polarized cytokine
production, reduced costimulatory requirements, and home to non-
lymphoid tissues. Pitcher et al. (35) recently reported an increased
ratio of memory (CD28?, CCR7?, CD62L?CD45RO?) to naive
(CD28?, CCR7?, CD62L?, CD45RO?) T cells in rhesus ma-
caque spleen compared with blood, confirming that rhesus spleen
is both a secondary lymphoid tissue (white pulp) and a potential
effector site (red pulp) (45). In this context, we expected both naive
and memory cells might coexist in our spleen NKT cells. How-
ever, we did not find evidence for naive cells in fresh rhesus spleen
V?24?cell population. Most fresh spleen cells gated for the
V?24?population expressed CCR7, and these cells consistently
coexpressed CD45RO, showing a central memory phenotype.
Consistent with this outcome, most V?24?cells were positive for
CD11a, the ?-chain of the adhesion molecule LFA-I. V?24 pos-
itively enriched NKT cells lost their expression of CCR7, CD62L,
and CD28, and developed elevated expression of CD45RO and
CD95 after short-term (5 days) culture (data not shown). Likewise,
in the long-term cultured NKT cells, the memory cell marker
CD45RO was moderate to strongly expressed throughout the cul-
ture period. However, further studies of the long-term cultured
NKT cells indicated a lack of expression of CD28, CD62L, and
CCR7, along with high expression of CD11a and ?7integrin, and
modest to high expression of CD95, confirming acquisition of an
Unstimulated/Resting NKT cells did not express activation and costimu-
latory markers CD69 (a) and CD40L (c) as well as intracellular cytokine
IL-2 (e), whereas PMA (20 ng/ml)- and ionomycin (1 ?g/ml)-stimulated
cells strongly up-regulated the expression of CD69 (b) and CD40L (d).
IL-2 expression (f) was not up-regulated. Positive controls show IL-2 ex-
pression in normal rhesus macaque peripheral blood MNCs with (h) and
without (g) stimulation with PMA/ionomycin. Stimulated and unstimulated
cells were incubated for 4 h at 37°C in complete medium containing brefel-
din A (10 ?g/ml) to prevent exocytosis. The samples were then washed and
stained with anti-human CD69 and anti-human CD40L, then fixed, per-
meabilized, and stained with anti-human IL-2 or isotype control Abs. Rep-
resentative data from three to four independent experiments are shown.
Rhesus macaque spleen-derived NKT cells are semianergic.
2908RHESUS MACAQUE NKT CELLS
effector memory phenotype in these spleen-derived rhesus ma-
caque NKT cells. From these findings, we speculate that rhesus
V?24?NKT cells home into the spleen as primed central memory
cells, and following Ag challenge they differentiate into effector
memory cells, able to home into nonlymphoid sites.
Although rhesus spleen NKT cells displayed the same effector
memory phenotype as mouse and neonatal and adult human NKT
cells (16, 17, 20, 21), they did not express activation markers in
regular culture conditions. Following activation with PMA and
ionomycin, CD69 and CD40L expression were up-regulated, but
other activation and costimulatory molecules, particularly CD25
(IL-2R?), CD122 (IL-2R?), and CD28, remained negative. Re-
gardless of weekly stimulation with ?-GalCer-pulsed DC in the
presence of rhIL-2 and rhIL-15, the growth of rhesus NKT cells
was slow, compared with that of human V?24?neonatal cord
blood NKT cells (46). In addition, rhesus NKT cells consistently
failed to produce intracellular IL-2 following stimulation with
PMA and ionomycin. Several factors may have contributed to this
outcome. For one, the lack of IL-2R on the rhesus NKT cells could
explain why they failed to proliferate, despite the presence of high
rhIL-2 concentrations. We speculate that the lack of costimulatory
molecules on iDC, added weekly to the NKT cultures, may have
also played a role. Except for conditions in which there is high
membrane Ag density, stimulation of the TCR with Ag is generally
insufficient to generate a productive conventional T cell response
with cytokine secretion and clonal expansion (47). Instead, sub-
optimal cross-linking of the TCR by Ag tends to lead to anergy,
which can be prevented by costimulation provided by accessory
molecules or the IL-2R (48, 49). Fujii et al. (50) showed that
?-GalCer stimulation alone causes NKT cell anergy. In our exper-
iments, frequent administration of free ?-GalCer every 2–3 days,
between the weekly stimulation with DC, might have driven the
rhesus NKT cells to become anergic. At the same time, the high
concentrations of supplemental rhIL-2 and rhIL-15, acting via low
density cytokine receptors, might have prevented them from be-
coming fully anergic, resulting in their long-term maintenance,
albeit slow growth in vitro. Overall, we suggest that a semianergic
state characterizes rhesus macaque spleen NKT cells in our regular
The semianergic NKT cells exhibited a regulatory cytokine pro-
file, common to that of Th2, Tr1, and Th3 regulatory cells. As with
conventional T cells, iDCs might induce NKT cells to differentiate
into regulatory cells (51). Roncarolo et al. (52) recently demon-
strated that CD4?Tr1 cell clones produce high levels of IL-10 and
TGF-?, moderate amounts of IFN-? and IL-5, but little or no IL-2
or IL-4, when activated via the TCR. In addition, Tr1 cell clones
rapidly produce IL-10 within 12–24 h after stimulation (53). In
contrast, T cells that are stimulated by the oral route primarily
secrete TGF-?, and are described as Th3 regulatory cells (54).
However, T cell lines generated from tolerant animals, following
oral administration of copolymer 1, produce high levels of IL-10
natants measured by ELISA. Culture supernatants were collected from the
cells at day 4 after weekly stimulation with ?-GalCer-pulsed DC, but with-
out PMA/ionomycin stimulation. Concentrations were calculated from the
mean of three wells for each cytokine. Representative data from three to
five independent experiments are shown.
Cytokine secretion of NKT cells in ambient culture super-
stained for intracellular cytokines. a, Unstimulated NKT
cells routinely produced intracellular IL-10, but minimal
IFN-?. b, PMA/ionomycin stimulation up-regulated in-
tracellular IFN-? production. The cells were harvested
at day 4 after weekly stimulation and incubated with or
without PMA/ionomycin for 4 h at 37oC in complete
medium containing brefeldin A (10 ?g/ml), then
washed, fixed, permeabilized, and stained with anti-
human IL-10 and anti-human IFN-? or isotype control
Abs. Intracellular anti-human IL-10 mAb was tested for
specificity using c, IL-10-transfected HEK cell line
(positive), and d, rhesus spleen-derived CD3?V?24?T
cells (negative). c, Anti-human IL-10 mAb was prein-
cubated with rhIL-10 (0.25 ?g/test) at room temperature
for 30 min. Then HEK cells were separately stained with
preblocked or unblocked anti-human IL-10 mAb. Neg-
ative cell surface IL-10 staining is shown on unperme-
abilized HEK cells, and positive staining for intracellu-
lar IL-10 is shown on permeabilized cells using anti-
CD3?V?24?T cells were stained with anti-human
IL-10 or isotype control mAbs after 4-h stimulation with
PMA/ionomycin, following fixation and permeabiliza-
tion. Representative data from three independent exper-
iments are shown.
Dot plot profiles of V?24?NKT cells
2909 The Journal of Immunology
and TGF-?, but little IL-2 or IL-4 (55). In our experiments, rhesus
NKT cells in resting conditions produced IL-10 at the single cell
level and secreted a large amount of TGF-? and IL-13. However,
the secretion of IL-10 was low at all time points tested (24–72 h;
data not shown), and production of both IL-4 and IL-2 was neg-
ligible. In this context, the cytokine profile of rhesus NKT cells
more closely resembles that of Th3 regulatory cells than Tr1 cells.
However, the high intracellular expression of IL-10 in the rhesus
cells leads us to speculate that they belong to a Tr1 lineage.
The discrepant pattern of positive intracellular IL-10 and neg-
ative extracellular IL-10 suggests that an inhibitory mechanism
may be interfering with IL-10 secretion in vitro. Yamauchi et al.
(57) recently demonstrated that thrombospondin-1 (TSP-1), an ex-
tracellular matrix glycoprotein synthesized by various types of
cells, including T cells (56), can inhibit IL-10 and enhance IL-6
release in vitro. They showed that TSP-1-induced activation of
TGF-?1, or exogenously added TGF-?1 inhibited IL-10 release.
Additionally, neutralizing Ab against TGF-?1 reversed TSP-1-in-
duced inhibition of IL-10 release. Collectively, these findings led
us to speculate that latent TGF-? might be activated by TSP-1 in
the rhesus NKT cell cultures, leading to inhibition of IL-10 release
into the supernatants. This mechanism, currently under examina-
tion, might explain the paradoxical high intracellular IL-10 and
low IL-10 secretion pattern that we have observed.
Both PMA-stimulated and resting rhesus NKT cells and their
culture supernatants inhibited T cell responses in allogeneic MLR.
This is analogous to and extends reports in human and mouse (13,
28). In view of their different cytokine and cell surface phenotypes,
we expected that the suppressive effects of rhesus NKT on MLR
might be different between PMA-stimulated and resting NKT
cells. Indeed, PMA-stimulated NKT cells suppressed MLR re-
sponses more strongly than did resting NKT cells, possibly related
to the augmented TGF-? produced by PMA-stimulated NKT cells.
However, the suppression by stimulated NKT cells occurred in an
inverse dose-response manner. The latter outcome might be ex-
plained by the heightened expression of CD40L and IFN-? after
PMA stimulation. Resting NKT cells expressed no detectable
CD40L or IFN-?, whereas ?90% of the stimulated cells expressed
both. In this context, it is conceivable that the PMA-stimulated
NKT cells enhanced the Ag-presenting activity of allogeneic DC
in MLR via CD40-CD40L interactions, promoting IL-12 and
IFN-? responses. In the MLR assays that contained the highest
numbers of NKT cells, this mechanism, along with the increased
expression of IFN-? by PMA-stimulated cells, could have defeated
the suppressive action of the PMA-stimulated NKT cells. Given
the striking conservation of NKT cells, another possibility is that
resting NKT cells responded to allogeneic CD1d?APC in MLR in
a manner distinct from that of pharmacologically (PMA/ionomy-
cin)-activated NKT cells. Taken together, our results suggest that
NKT cells from rhesus macaques have a novel invariant TCR?,
but V?11?CD56?CD161?and memory phenotype coupled to a
distinct regulatory suppressive function.
In conclusion, we have characterized spleen-derived rhesus ma-
caque NKT cells, which have been cultured and expanded in vitro.
This expansion is necessary because the numbers of NKT cells
present in normal rhesus tissues are too low to allow for extensive
analysis. Our results allow several conclusions regarding cultured
rhesus NKT cells. First, they are highly homologous with human
NKT cells and less so with murine counterparts. Second, they dis-
play a semi-invariant TCR (V?24?/J?Q?) and are CD4 CD8 DN,
CD161?, and CD56?cells with effector memory phenotype.
Third, rhesus NKT cells and their culture supernatants suppress
allogeneic MLR, and have a cytokine expression pattern most con-
sistent with Th3 cells. Although these properties might be skewed
by in vitro expansion, the studies provide original information that
will be useful for future characterization of uncultured NKT cells.
We thank Dr. T. Mise (Kirin Brewery) for providing the ?-GalCer, Martha
Wilson for caring for the monkeys and providing blood and tissue samples,
Stacie Jenkins for performing the ELISA, and Dr. Steven Porcelli for his
helpful critique of the manuscript.
1. Bendelac, A., O. Lantz, M. E. Quimby, J. W. Yewdell, J. R. Bennink, and
R. R. Brutkiewicz. 1995. CD1 recognition by mouse NK1?T lymphocytes. Sci-
2. Dellabona, P., E. Padovan, G. Casorati, M. Brockhaus, and A. Lanzavecchia.
1994. An invariant V?24-J?Q/V?11 T cell receptor is expressed in all individ-
uals by clonally expanded CD4?8?T cells. J. Exp. Med. 180:1171.
3. Lantz, O., and A. Bendelac. 1994. An invariant T cell receptor ? chain is used by
a unique subset of major histocompatibility complex class I-specific CD4?and
CD4?8?T cells in mice and humans. J. Exp. Med. 180:1097.
4 after weekly stimulation. PMA/Ionomycin stimulation was in the absence of brefeldin A. R, responder T cell; DC, allogeneic mature DC; sup., NKT
culture supernatant; ratio of R:DC, 20:1. Ratio of R:NKT, 2:1, 5:1, and 10:1. Representative data of four independent, triplicate experiments are shown.
Suppressive effect of NKT cells on DC-stimulated allogeneic T cell proliferation. NKT cells and culture supernatants were harvested at day
2910 RHESUS MACAQUE NKT CELLS
4. Lanier, L. L., C. Chang, and J. H. Phillips. 1994. Human NKR-P1A: a disulfide-
linked homodimer of the C-type lectin superfamily expressed by a subset of NK
and T lymphocytes. J. Immunol. 153:2417.
5. Sato, N., T. Yahata, K. Santa, A. Ohta, Y. Ohmi, S. Habu, and T. Nishimura.
1996. Functional characterization of NK1.1?Ly-6C?cells. Immunol. Lett. 54:5.
6. Porcelli, S., C. E. Yockey, M. B. Brenner, and S. P. Balk. 1993. Analysis of T cell
antigen receptor (TCR) expression by human peripheral blood CD4?8??/? T
cells demonstrates preferential use of several V? genes and an invariant TCR ?
chain. J. Exp. Med. 178:1.
7. Exley, M., J. Garcia, S. P. Balk, and S. Porcelli. 1997. Requirements for CD1d
recognition by human invariant V?24?CD4?CD8?T cells. J. Exp. Med.
8. Spada, F. M., Y. Koezuka, and S. A. Porcelli. 1998. CD1d-restricted recognition
of synthetic glycolipid antigens by human natural killer T cells. J. Exp. Med.
9. Gritzapis, A. D., D. Dimitroulopoulos, E. Paraskevas, C. N. Baxevanis, and
M. Papamichail. 2002. Large-scale expansion of CD3?CD56?lymphocytes ca-
pable of lysing autologous tumor cells with cytokine-rich supernatants. Cancer
Immunol. Immunother. 51:440.
10. Makino, Y., R. Kanno, T. Ito, K. Higashino, and M. Taniguchi. 1995. Predom-
inant expression of invariant V?14?TCR ? chain in NK1.1?T cell populations.
Int. Immunol. 7:1157.
11. Kawano, T., Y. Tanaka, E. Shimizu, Y. Kaneko, N. Kamata, H. Sato, H. Osada,
S. Sekiya, T. Nakayama, and M. Taniguchi. 1999. A novel recognition motif of
human NKT antigen receptor for a glycolipid ligand. Int. Immunol. 11:881.
12. Kawano, T., J. Cui, Y. Koezuka, I. Toura, Y. Kaneko, K. Motoki, H. Ueno,
R. Nakagawa, H. Sato, E. Kondo, et al. 1997. CD1d-restricted and TCR-mediated
activation of V?14 NKT cells by glycosylceramides. Science 278:1626.
13. Exley, M. A., S. M. Tahir, O. Cheng, A. Shaulov, R. Joyce, D. Avigan,
R. Sackstein, and S. P. Balk. 2001. A major fraction of human bone marrow
lymphocytes are Th2-like CD1d-reactive T cells that can suppress mixed lym-
phocyte responses. J. Immunol. 167:5531.
14. Behar, S. M., T. A. Podrebarac, C. J. Roy, C. R. Wang, and M. B. Brenner. 1999.
Diverse TCRs recognize murine CD1. J. Immunol. 162:161.
15. Brossay, L., M. Chioda, N. Burdin, Y. Koezuka, G. Casorati, P. Dellabona, and
M. Kronenberg. 1998. CD1d-mediated recognition of an ?-galactosylceramide
by natural killer T cells is highly conserved through mammalian evolution.
J. Exp. Med. 188:1521.
16. Gumperz, J. E., S. Miyake, T. Yamamura, and M. B. Brenner. 2002. Functionally
distinct subsets of CD1d-restricted natural killer T cells revealed by CD1d tet-
ramer staining. J. Exp. Med. 195:625.
17. Lee, P. T., K. Benlagha, L. Teyton, and A. Bendelac. 2002. Distinct functional
lineages of human V?24 natural killer T cells. J. Exp. Med. 195:637.
18. Dunne, J., S. Lynch, C. O’Farrelly, S. Todryk, J. E. Hegarty, C. Feighery, and
D. G. Doherty. 2001. Selective expansion and partial activation of human NK
cells and NK receptor-positive T cells by IL-2 and IL-15. J. Immunol. 167:3129.
19. Godfrey, D. I., K. J. Hammond, L. D. Poulton, M. J. Smyth, and A. G. Baxter.
2000. NKT cells: facts, functions and fallacies. Immunol. Today 21:573.
20. Eberl, G., R. Lees, S. T. Smiley, M. Taniguchi, M. J. Grusby, and
H. R. MacDonald. 1999. Tissue-specific segregation of CD1d-dependent and
CD1d-independent NK T cells. J. Immunol. 162:6410.
21. Zeng, D., G. Gazit, S. Dejbakhsh-Jones, S. P. Balk, S. Snapper, M. Taniguchi,
and S. Strober. 1999. Heterogeneity of NK1.1?T cells in the bone marrow:
divergence from the thymus. J. Immunol. 163:5338.
22. D’Andrea, A., D. Goux, C. De Lalla, Y. Koezuka, D. Montagna, A. Moretta,
P. Dellabona, G. Casorati, and S. Abrignani. 2000. Neonatal invariant V?24?
NKT lymphocytes are activated memory cells. Eur. J. Immunol. 30:1544.
23. Cui, J., T. Shin, T. Kawano, H. Sato, E. Kondo, I. Toura, Y. Kaneko, H. Koseki,
M. Kanno, and M. Taniguchi. 1997. Requirement for V?14 NKT cells in IL-12-
mediated rejection of tumors. Science 278:1623.
24. Toura, I., T. Kawano, Y. Akutsu, T. Nakayama, T. Ochiai, and M. Taniguchi.
1999. Cutting edge: inhibition of experimental tumor metastasis by dendritic cells
pulsed with ?-galactosylceramide. J. Immunol. 163:2387.
25. Apostolou, I., Y. Takahama, C. Belmant, T. Kawano, M. Huerre, G. Marchal,
J. Cui, M. Taniguchi, H. Nakauchi, J. J. Fournie, et al. 1999. Murine natural killer
T (NKT) cells contribute to the granulomatous reaction caused by mycobacterial
cell walls. Proc. Natl. Acad. Sci. USA 96:5141.
26. Kakimi, K., L. G. Guidotti, Y. Koezuka, and F. V. Chisari. 2000. Natural killer
T cell activation inhibits hepatitis B virus replication in vivo. J. Exp. Med.
27. Exley, M. A., N. J. Bigley, O. Cheng, S. M. Tahir, S. T. Smiley, Q. L. Carter,
H. F. Stills, M. J. Grusby, Y. Koezuka, M. Taniguchi, and S. P. Balk. 2001.
CD1d-reactive T-cell activation leads to amelioration of disease caused by dia-
betogenic encephalomyocarditis virus. J. Leukocyte Biol. 69:713.
28. Zeng, D., D. Lewis, S. Dejbakhsh-Jones, F. Lan, M. Garcia-Ojeda, R. Sibley, and
S. Strober. 1999. Bone marrow NK1.1?and NK1.1?T cells reciprocally regulate
acute graft versus host disease. J. Exp. Med. 189:1073.
29. Sonoda, K. H., M. Exley, S. Snapper, S. P. Balk, and J. Stein-Streilein. 1999.
CD1-reactive natural killer T cells are required for development of systemic
tolerance through an immune-privileged site. J. Exp. Med. 190:1215.
30. Seino, K. I., K. Fukao, K. Muramoto, K. Yanagisawa, Y. Takada, S. Kakuta,
Y. Iwakura, L. Van Kaer, K. Takeda, T. Nakayama, et al. 2001. Requirement for
natural killer T (NKT) cells in the induction of allograft tolerance. Proc. Natl.
Acad. Sci. USA 98:2577.
31. Hayakawa, Y., K. Takeda, H. Yagita, L. Van Kaer, I. Saiki, and K. Okumura.
2001. Differential regulation of Th1 and Th2 functions of NKT cells by CD28 and
CD40 costimulatory pathways. J. Immunol. 166:6012.
32. Kadowaki, N., S. Antonenko, S. Ho, M. C. Rissoan, V. Soumelis, S. A. Porcelli,
L. L. Lanier, and Y. J. Liu. 2001. Distinct cytokine profiles of neonatal natural
killer T cells after expansion with subsets of dendritic cells. J. Exp. Med.
33. Tahir, S. M., O. Cheng, A. Shaulov, Y. Koezuka, G. J. Bubley, S. B. Wilson,
S. P. Balk, and M. A. Exley. 2001. Loss of IFN-? production by invariant NK T
cells in advanced cancer. J. Immunol. 167:4046.
34. Exley, M., S. Porcelli, M. Furman, J. Garcia, and S. Balk. 1998. CD161 (NKR-
P1A) costimulation of CD1d-dependent activation of human T cells expressing
invariant V?24 J?Q T cell receptor ? chains. J. Exp. Med. 188:867.
35. Pitcher, C. J., S. I. Hagen, J. M. Walker, R. Lum, B. L. Mitchell, V. C. Maino,
M. K. Axthelm, and L. J. Picker. 2002. Development and homeostasis of T cell
memory in rhesus macaque. J. Immunol. 168:29.
36. Nieda, M., A. Kikuchi, A. Nicol, Y. Koezuka, Y. Ando, S. Ishihara, N. Lapteva,
T. Yabe, K. Tokunaga, K. Tadokoro, and T. Juji. 2001. Dendritic cells rapidly
undergo apoptosis in vitro following culture with activated CD4?V?24 natural
killer T cells expressing CD40L. Immunology 102:137.
37. Nieda, M., A. Nicol, Y. Koezuka, A. Kikuchi, N. Lapteva, Y. Tanaka,
K. Tokunaga, K. Suzuki, N. Kayagaki, H. Yagita, et al. 2001. TRAIL expression
by activated human CD4?V?24 NKT cells induces in vitro and in vivo apoptosis
of human acute myeloid leukemia cells. Blood 97:2067.
38. Thomas, J. M., J. L. Contreras, C. A. Smyth, A. Lobashevsky, S. Jenkins,
W. J. Hubbard, D. E. Eckhoff, S. Stavrou, D. M. Neville, Jr., and F. T. Thomas.
2001. Successful reversal of streptozotocin-induced diabetes with stable alloge-
neic islet function in a preclinical model of type 1 diabetes. Diabetes 50:1227.
39. Groux, H., M. Bigler, J. E. de Vries, and M. G. Roncarolo. 1996. Interleukin-10
induces a long-term antigen-specific anergic state in human CD4?T cells. J. Exp.
40. Khouri, S. J., W. W. Hancock, and H. L. Weiner. 1992. Oral tolerance to myelin
basic protein and natural recovery from experimental autoimmune encephalomy-
elitis are associated with down-regulation of inflammatory cytokines and differ-
ential up-regulation of transforming growth factor-?, interleukin-4, and prosta-
glandin E expression in the brain. J. Exp. Med. 176:1355.
41. Groux, H., and F. Powrie. 1999. Regulatory T cells and inflammatory bowel
disease. Immunol. Today 20:442.
42. Groux, H., A. O’Garra, M. Bigler, M. Rouleau, S. Antonenko, J. E. de Vries, and
M. G. Roncarolo. 1997. A CD4?T-cell subset inhibits antigen-specific T-cell
responses and prevents colitis. Nature 389:737.
43. Fukaura, H., S. C. Kent, M. J. Pietrusewicz, S. J. Khoury, H. L. Weiner, and
D. A. Hafler. 1996. Induction of circulating myelin basic protein and proteolipid
protein-specific transforming growth factor-?1-secreting Th3 T cells by oral ad-
ministration of myelin in multiple sclerosis patients. J. Clin. Invest. 98:70.
44. Tarazona, R., O. DelaRosa, C. Alonso, B. Ostos, J. Espejo, J. Pena, and
R. Solana. 2000. Increased expression of NK cell markers on T lymphocytes in
aging and chronic activation of the immune system reflects the accumulation of
effector/senescent T cells. Mech. Ageing Dev. 121:77.
45. Picker, L. J., and M. H. Siegelman. 1999. Lymphoid organs and tissues. In Fun-
damental Immunology, 4th Ed., Ch. 14. W. E. Paul, ed. Lippincott-Raven, Phil-
adelphia, p. 479.
46. Gansuvd, B., M. Hagihara, Y. Yu, H. Inoue, Y. Ueda, T. Tsuchiya, A. Masui,
K. Ando, Y. Nakamura, N. Munkhtuvshin, et al. 2002. Human umbilical cord
blood NK T cells kill tumors by multiple cytotoxic mechanisms. Hum. Immunol.
47. Weiss, A., and D. R. Littman. 1994. Signal transduction by lymphocyte antigen
receptors. Cell 76:263.
48. Schwartz, R. H. 1997. T cell clonal anergy. Curr. Opin. Immunol. 9:351.
49. Beverly, B., S. M. Kang, M. J. Lenardo, and R. H. Schwartz. 1992. Reversal of
in vitro T cell clonal anergy by IL-2 stimulation. Int. Immunol. 4:661.
50. Fujii, S., K. Shimizu, M. Kronenberg, and R. M. Steinman. 2002. Prolonged
IFN-?-producing NKT response induced with ?-galactosylceramide-loaded DCs.
Nat. Immun. 3:867.
51. Roncarolo, M. G., M. K. Levings, and C. Traversari. 2001. Differentiation of T
regulatory cells by immature dendritic cells. J. Exp. Med. 193:F5.
52. Roncarolo, M. G., R. Bacchetta, C. Bordignon, S. Narula, and M. K. Levings.
2001. Type I T regulatory cells. Immunol. Rev. 182:68.
53. Bacchetta, R., M. Bigler, J. L. Touraine, R. Parkman, P. A. Tovo, J. Abrams,
R. de Waal Malefyt, J. E. deVries, and M. G. Roncarolo. 1994. High levels of
interleukin 10 production in vivo are associated with tolerance in SCID patients
transplanted with HLA mismatched hematopoietic stem cells. J. Exp. Med.
54. Chen, Y., V. K. Kuchroo, J. Inobe, D. A. Hafler, and H. L. Weiner. 1994. Reg-
ulatory T cell clones induced by oral tolerance: suppression of autoimmune en-
cephalomyelitis by oral administration of copolymer 1. Science 265:1237.
55. Teitelbaum, D., R. Arnon, and M. Sela. 1999. Immunomodulation of experimen-
tal autoimmune encephalomyelitis by oral administration of copolymer 1. Proc.
Natl. Acad. Sci. USA 96:3842.
56. Li, S. S., A. Ivanoff, S. Bergstro ¨m, A. Sandstro ¨m, B. Christensson, J. van Nerven,
J. Holgersson, D. Hauzenberger, I. Arencibia, and K. Sundqvist. 2002. T lym-
phocyte expression of thrombospondin-1 and adhesion to extracellular matrix
components. Eur. J. Immunol. 32:1069.
57. Yamauchi, Y., M. Kuroki, T. Imakiire, H. Abe, H. Uchida, R. Beppu,
Y. Yamashita, M. Kuroki, and T. Shirakusa. 2002. Thrombospondin-1 differen-
tially regulates release of IL-6 and IL-10 by human monocytic cell line U937.
Biochem. Biophys. Res. Commun. 290:1551.
2911The Journal of Immunology