| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Laboratory of Experimental Immunology, National Cancer Institute-Center for Cancer Research, Frederick, MD 21702; and
Department Microbiology, University of Nevada Medical School, Reno, NV 89557
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-galactosylceramide (GalCer). However, GalCer effectively stimulates and then diminishes the number of detectable NKT cells. It also exhibits a potent, indirect ability to activate NK cells. We have now discovered another ceramide compound, 
-galactosylceramide (GalCer) (C12), that efficiently diminishes the number of detectable mouse NKT cells in vivo without inducing significant cytokine expression or activation of NK cells. Binding studies using CD1d tetramers loaded with GalCer (C12) demonstrated significant but lower intensity binding to NKT cells when compared with GalCer, but both ceramides were equally efficient in reducing the number of NKT cells. However, GalCer (C12), in contrast to GalCer, failed to increase NK cell size, number, and cytolytic activity. Also in contrast to GalCer, GalCer (C12) is a poor inducer of IFN-
, TNF-, GM-CSF, and IL-4 gene expression. These qualitative differences in NKT perturbation/NK activation have important implications for delineating the unique in vivo roles of NKT vs NK cells. Thus, GalCer (which triggers NKT cells and activates NK cells) efficiently increases the resistance to allogeneic bone marrow transplantation while GalCer (C12) (which triggers NKT cells but does not activate NK cells) fails to enhance bone marrow graft rejection. Our results show GalCer (C12) can effectively discriminate between NKT- and NK-mediated responses in vivo. These results indicate the use of different TCR-binding ceramides can provide a unique approach for understanding the intricate immunoregulatory contributions of these two cell types. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
14-J18 variable region of the TCR and are either CD4+ or CD4-, CD8-. The responsiveness of these NKT cells is CD1d restricted, and these cells are primarily located in the thymus, liver, spleen, and bone marrow (1). The reactivity of this subset appears restricted to -galactosylceramide (GalCer)3 (1, 3) and represents a major component of the overall NKT cell population. This NKT subset coexpresses Ly49 receptors that are also found on NK cells (2). A second category of NKT cells is CD1d autoreactive and coexpresses a more distinctive TCR expression pattern (V3.2-J9/V8, V8) than type I (1). Although these NKT cells are CD1d autoreactive, they are not truly polyclonal. Like type I NKT cells, they are CD4+ or CD4-, CD8- and express predominately NK1.1. However they differ from type I NKT cells in both their location in vivo (liver) and the fact that they are nonreactive with GalCer (1). A third subset of NKT cells is not CD1d dependent, has a diverse TCR expression, and can be either CD4+, CD8+ or CD4-, CD8-. The Ag reactivity of this third subset is not known, and their in vivo distribution is similar to that of type I NKT cells. A fourth subset of NKT cells is defined by their expression of 2-integrin (CD49B), which is recognized by the NK-reactive Ab, DX5. This NKT population has little overlap with the type I subset and may have a very specialized role in the immune response (1).
Mouse studies have demonstrated that NKT cells can regulate autoimmune responses (4, 5, 6, 7, 8, 9, 10, 11) and can reverse some types of immune dysregulation (12, 13, 14, 15, 16). A better understanding of the characteristics and functions of NKT cells may provide new insights into treatment of autoimmune diseases, as well as novel strategies to increase reactivity against cancer and AIDS. In this context, many studies of NKT cells have centered on the functions of CD1-restricted, V14-J18 or V3, 2-J9/V8, V8-expressing NKT cell subsets that represent a majority of spleen-, liver-, and bone marrow-associated NKT cells (1, 3). The determination of NKT-mediated biological activities has been largely dependent on the ability of GalCer to selectively target and perturb these cells and thereby relate their apparent presence or absence to many facets of immune regulation. In this regard, administration of GalCer in vivo has been reported to rapidly deplete NKT cells through a TCR-dependent apoptotic process (17). However, during this process GalCer triggers production of IFN- from NKT cells and also activates NK cells that are often found in the spleen and liver. This secondary activation of NK cells can have confusing implications for delineating the role of NKT cells in various disease states and immune responses, because the activation of colocated NK cells could account for changes in biological activities that accompany loss of NKT cell detection. Thus, it is quite difficult to differentiate the baseline contribution of NKT cells to various immune responses from NKT-dependent or -independent amplification of other leukocyte responses.
In this study, we have shown that GalCer (C12) can be efficiently presented on CD1d and rapidly reduces the number of detectable NKT cells in vivo without inducing production of IFN-, activation of NK-mediated lysis, or NK-dependent rejection of bone marrow grafts.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mice were obtained from the Animal Production Area, National Cancer Institute-Frederick and used between 6 and 12 wk of age.
Ceramide reagents
Various reagents including D-glucosyl-1-1' ceramide (C8), D-galactosyl-1-1' ceramide (C8), D-glucosyl-1-1' ceramide (C12), and D-galactosyl-1-1' ceramide (C12) were purchased from AvantiLipids (Alabaster, AL). GalCer (KRN7000) was graciously provided by Kirin Brewery (Tokyo, Japan). The ceramide reagents were first dissolved in DMSO, then diluted in PBS containing 0.5% Tween 20. Control diluent or PBS was used as a control for all studies.
NK cell isolation
NK cells were isolated from the livers or spleens of C57BL/6 (B6) mice and grown for 7–10 days in 1000 IU/ml rIL-2 (Hoffman-LaRoche, Nutley, NJ) as previously described (2). Liver-associated NK cells were isolated from IL-2-treated mice as previously described (2). Liver-associated mononuclear cells (35–70% CD3-, NK1.1+) were used immediately after isolation or after IL-2 expansion.
Proteose peptone method
As previously described (18), mice were injected i.p. with 1 ml of 3% proteose peptone (BD Biosciences, Sparks, MD). After 4 days, the peritoneal exudate cells were removed from the abdominal cavity using a needle and syringe. These cells were washed with PBS and used in the assay.
Flow cytometry reagents and Abs
NK1.1-PE (or allophycocyanin), DX-5-PE and CD3-PerCP (BD PharMingen, San Jose, CA), as well as CD69-FITC, were used for flow cytometric analysis as previously described (2). Intracellular detection of cytoplasmic IFN- was performed using kits purchased from BD PharMingen and used as per the manufacturer’s instructions. Annexin V staining was used to evaluate early events in apoptosis (BD PharMingen).
Flow cytometry analysis and lymphocyte sorting
Cells were stained as previously described (2) and analyzed on a FACSort or LSR System flow cytometer (BD Biosciences, San Jose, CA). Cells were directly stained using FITC-, PE-, PerCP-, and allophycocyanin-labeled primary Abs. Sorting experiments were performed on a MoFlo cell sorter (DAKO/Cytomation, Ft. Collins, CO) to isolate NK and NKT cells using NK1.1-PE and CD3biotin-SA.Tricolor (Caltag Laboratories, Burlingame, CA) as the defining immunological markers.
Cytokine measurement
IFN- and chemokine proteins were measured in culture supernatants or mouse serum using ELISA kits (R&D Systems, Minneapolis, MN) as previously described (2). Cell stimulations were performed at concentrations of 1–5 x 106 cells/ml, and variation within assays was always <5 pg/ml. In some experiments, B cell lines (A20 and A20/CD1d generously provided by M. Kronenberg, La Jolla Institute, San Diego, CA) were pretreated with various reagents for 30 min at 37°C, washed and mixed with sorted populations of NK or NKT cells (>98% pure), and supernatants were collected for analysis after 24 h of culture.
RNase protection assay
Multiprobe RNase protection assays were performed using the mck-1 or mck-5 probe sets (BD PharMingen). Total cellular RNA was extracted using TRIzol (Life Technologies, Gaithersburg, MD), and 1–5 µg of total mRNA was hybridized with a [33P]UTP-labeled RNA probe (1 x 106 cpm/sample) prepared according to the manufacturers directions (BD PharMingen) using the BD PharMingen RiboQuant in vitro transcription kit. Following hybridization, the samples were treated with RNase A and T1 according to the procedure provided by BD PharMingen. The RNase was inactivated and precipitated using a solution containing 200 µl of Ambion (Austin, TX) RNase inactivation reagent, 50 µl of ethanol, 5 µg of yeast tRNA, and 1 µl of Ambion GycoBlue coprecipitate per RNA sample. The samples were mixed well, incubated at -70°C for 15 min, and centrifuged at 14,000 rpm for 15 min at room temperature. The pellets were then suspended in 3 µl of BD PharMingen sample buffer and subjected to PAGE as recommended by the manufacturer (BD PharMingen).
Cytotoxicity assay
NK cytotoxicity was evaluated against the prototype NK target Yac-1 in a standard 4-h 51Cr release assay as previously described (2).
Bone marrow transplantation
The efficiency of reconstitution after bone marrow transplantation was evaluated using a new label-tracking procedure (19) that uses CFSE (Molecular Probes, Eugene, OR)-labeled bone marrow cells that can be evaluated in vivo after 1–4 days. Briefly, 10 x 106 CFSE-labeled bone marrow cells (20) were injected into irradiated recipient mice (C57BL/6, 900 rad; BALB/c, 800 rad). The spleens were then evaluated by flow cytometry for the total number of labeled cells in treated or untreated autologous vs allogeneic transplant groups. Mice were evaluated individually, and the mean and SD for each group was calculated.
Tetrameric GalCer-CD1d complexes
Tetramers of wild-type CD1d molecules were produced as described by Matsuda et al. (21). CD1d molecules were biotinylated and prepared by the National Institutes of Health tetramer facility. The production and biotinylation of murine CD1d (mCD1d) protocol was as previously reported (21). mCD1d was coexpressed with 2-microglobulin in insect cells (High 5) using a baculovirus expression system. The monomer was purified from media by concentration on a tangible filter device followed by standard nickle-agarose separation. The correct fractions were combined and exchanged twice in biotinylation buffer. The biotinylation reaction was conducted overnight at 25°C. Biotinylated mCD1d was separated from biotin and ATP on a S300 gel filtration column. Biotinylated CD1d molecules were then incubated overnight at room temperature with a 3-fold molar excess of or GalCer (solubilized in 0.5% Tween 20, 0.9% sodium chloride, hereafter called vehicle) or with an equal amount of vehicle alone. CD1d monomers were then tetramerized using a 1:4 molar ratio of allophycocyanin-conjugated streptavidin (BD PharMingen).
Tetramer staining
All staining and washes were performed in a buffer consisting of 2% FCS, 0.075% sodium bicarbonate, and 0.1% sodium azide in RPMI 1640 medium (Sigma-Aldrich, St. Louis, MO), as previous described (21). Cells were stained for 3 h at room temperature, then stained with subset Abs, washed two times, and fixed in 1% paraformaldehyde (Sigma-Aldrich) in PBS. The intensity of fluorescence on hybridoma cells was determined by flow cytometry analysis using a FACSort flow cytometer (BD Biosciences).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GalCer
To ultimately compare and contrast the immune-stimulating effects of different ceramide compounds, a series of in vivo studies was performed to define the direct effects of GalCer on NKT cells, as well as subsequent indirect effects on NK cells (Fig. 1). A rapid activation of NK cells was noted as defined by increased NK-mediated lysis (Fig. 1A), increased size of NK cells (Fig. 1B), increased expression of the CD69 activation Ag (Fig. 1C), increased total number (Fig. 1D), and percentage (Fig. 1F) of NK cells. As expected (1), administration of several doses of GalCer resulted in a rapid decrease in the number (Fig. 1E) and percentage (Fig. 1G) of detectable NKT cells. The highest doses of GalCer caused a 2- to 10-fold increase in total liver cellularity (Fig. 1H) and a rapid and potent increase in the amounts of IFN- in the serum that persisted for up to 48 h (Fig. 1I). Interestingly, a wide range of doses of GalCer resulted in potent and sustained (> 7 days) activation of NK cells (lytic function, Fig. 1A) and increase in NK cell numbers and frequency (Fig. 1, D and F). Although these effects were dose dependent, doses as low as 0.1 µg/mouse resulted in significantly increased activities. These findings clearly show that in vivo administration of GalCer results in a rapid decrease in a major NKT subset from the liver that coincides with a rapid and persistent activation of NK cells as defined by multiple parameters.
|
GalCer (Fig. 2). Administration of GalCer to SCID mice failed to increase NK cell size or cytolytic activity, whereas expected increases in both NK cell size and lytic functions were seen in control C57BL/6 mice. Other parameters of NK cell activation as shown in Fig. 1 were also not observed in SCID mice (data not shown). These results strongly implicate activation of NKT cells as a prerequisite for subsequent NK cell activation after administration of GalCer.
|
GalCer-induced NKT cell depletion and NK activation is not dependent on TH1 cytokines, perforin, or Fas ligand (FasL)
Because the decrease in NKT cell detection induced by GalCer is accompanied by a rapid burst of IFN- production and NK activation, we speculated that TH1 cytokines could be critical regulatory factors in these effects. Furthermore, because NKT cells rapidly become undetectable after GalCer administration, we speculated that the Fas or perforin pathways might contribute to this effect. Table I evaluates a series of GalCer treatments in selected gene disruption mice. However, the ability of GalCer to reduce the number of detectable NKT cells and activate NK cells were largely retained in mice with dysregulated TH1 cytokine, TNF superfamily, and perforin genes. Some mutant or knockout mice, such as FasL (gld) and CD40L (-/-), had low baseline percentages of NKT cells, but an additional loss of >50% of these cells was observed after GalCer treatment. Similarly, NK bystander activation as assessed by NK lysis, increased size, and up-regulation of CD69 was observed in all knockout mice (as expected perforin (PFP) mice lacked NK-mediated lysis). Interestingly, dysregulation of the FasL gene in gld mice impaired the ability of GalCer to substantially increase both NK cell number and function, suggesting a possible role for the Fas signaling pathway in the NK activation effect. It should be noted that 24 h is a relatively early time point for evaluation of these cellular increases, which are generally maximal at 3–5 days after GalCer administration, so additional studies will be needed to see if the FasL-dependent effect is due to a qualitative change in the response or to an altered kinetics of response. Perhaps of most interest was the observation that IFN- knockout mice demonstrate efficient NK bystander activation, showing that IFN- was not required for most of the bystander effects, and suggesting that the activation of NK cells may not be directly linked to the burst of IFN- that occurs coincident with NKT cell stimulation.
|
GalCer (C12)
The results described above re-emphasize the coordinate induction of two quite different GalCer-induced regulatory events in key innate immune parameters, a reduction in the number of detectable NKT cells and the activation of NK cells. This data set demonstrates the potency of GalCer for stimulating innate immune functions, but also illustrates the difficulties in using GalCer to discriminate the roles of these two cell types in various biological effects in vivo.
Therefore, we speculated that perhaps other ceramide compounds might retain the ability to bind to CD1d on B cells and monocytes, but differ qualitatively from GalCer in their ability to directly deliver signals to NKT cells and indirectly activate NK cells. Therefore, studies were performed to evaluate if ceramides could be identified that would specifically stimulate NKT cells without causing downstream activation events (e.g., NK activation and IFN- production). The data presented in Table II show that GalCer (C12) potently reduces detectable NKT cells, while other ceramide compounds including GalCer (C8), -GlucCer (C12), and -GlucCer (C8) demonstrated no biological activity when compared with the diluent control. As shown in both Table II and Fig. 3, administration of GalCer (C12) resulted in a similar reduction in detectable NKT cells as seen for GalCer (Fig. 3D), but without the broad bystander NK activation obtained after administration of GalCer, as illustrated by a lack of increased percentage, lysis, and CD69 expression (Fig. 3, A–C, respectively). High doses of GalCer (C12) were used to ensure that maximal depletion and/or potential activation would be detected. Overall, it is clear that GalCer (C12) can very effectively reduce the number of detectable NKT cells, and this effect can be mediated in the absence of the NKT-dependent bystander NK activation that occurs after administration of GalCer.
|
|
GalCer (C12) reduces the number of detectable NKT cells in the absence of TH1 and TH2 cytokine induction
Because previous studies (1, 3, 22, 23), as well as our own results (Table I, Fig. 2), have demonstrated that GalCer activation and subsequent reduction in detectable NKT cells results in strong IFN- production, we evaluated the abilities of GalCer vs GalCer (C12) to induce cytokine production in vivo in C57BL/6 mice. Cytokine gene expression for IL-4, IL-13, or IFN- was evaluated in both liver and spleen 1, 2, or 4 h after administration of GalCer and GalCer (C12). As shown in Fig. 4, only GalCer induced IL-4, IL-13 (data not shown), or IFN-. Interestingly, the ability of GalCer to increase gene expression at 2 h and/or 4 h was demonstrated in both the liver, which contains large numbers of NKT cells (15–30% of all lymphocytes), as well as in the spleen, which generally contains only 1–3% NKT cells. Thus, although both GalCer and GalCer (C12) reduce NKT cell detection, only GalCer induces cytokine gene expression in spleen and liver. These results were extended by further studies where the ability of GalCer and GalCer (C12) to induce cytokine proteins in the serum was studied. The results shown in Fig. 5 demonstrate that GalCer potently increases serum levels of IFN- and GM-CSF (Fig. 5A) and IL-4 and TNF- (Fig. 5B) at 1, 2, or 4 h after treatment, whereas GalCer (C12) did not induce appreciable amounts of these cytokines. The rapid induction of cytokine genes and proteins by GalCer suggests that these effects must be closely linked to the binding and activation of GalCer to the TCR of NKT cells. However, the failure of GalCer (C12) to induce cytokine gene expression while effectively reducing the number of detectable NKT cells contrasts sharply with the effects of GalCer. This apparent dichotomy of activities for these two ceramides suggests on one hand a similar pathway for binding to and stimulating NKT cells, but in contrast a downstream divergence in signaling for cytokine gene expression. To investigate these questions, studies were performed to compare the abilities of GalCer vs GalCer (C12) to induce annexin V binding to NKT cells as an early measure of apoptosis induction. The results shown in Fig. 5C demonstrate that both GalCer and GalCer (C12) rapidly reduce the number of NKT cells (by 
50% within 1.5 h) also induce a rapid increase in annexin V binding on NKT but not NK cells (Fig. 5D). The simultaneous induction of both annexin V expression and loss of NKT detection for both GalCer and GalCer (C12) suggests a similar molecular mechanism for these effects by both agents. Examination of NK cell numbers over this time period demonstrated no increase in annexin V expression for either GalCer or GalCer (C12). Thus, GalCer (C12) can be used to dissociate NK-activating events from the process of NKT apoptosis and disappearance, suggesting that the quality of the signals induced by GalCer (C12) and GalCer binding to the TCR differ.
|
|
GalCer and GalCer (C12) to NKT cells
Because the GalCer resulted in such a rapid and pronounced increase of NK cell activation and cytokine induction whereas GalCer (C12) failed to induce this potent stimulation, we evaluated the relative efficiency of binding of both ceramides to NKT cells. The results of a typical binding study are shown in Fig. 6. CD1d tetramers were either loaded with GalCer or GalCer (C12) and their binding to NKT-enriched liver lymphocytes was studied. Fig. 6A reveals the strong binding of GalCer-loaded CD1d to NKT cells (Fig. 6A, lower left), and this was not observed with either NK cells or NK1.1- T cells. NKT cells evaluated with CD1d tetramers without either GalCer or GalCer loading failed to demonstrate any significant binding (data not shown). Fig. 6B shows binding with GalCer (C12)-loaded tetramers where a similar percentage of NKT cells bound the CD1d, albeit at lower intensity, while both NK and NK1.1- T cells failed to demonstrate this binding (Fig. 6A and data not shown). When NKT subsets were evaluated, both CD4 and CD8 subsets bound GalCer with a strong intensity (Fig. 6B, left), while GalCer (C12) binding to CD4+ NKT cells was also similar to its binding to CD8+ cells (Fig. 6B, right). Overall, these results demonstrate strong binding of GalCer on both CD4 and CD8 NKT cells (Fig. 6B, left) with weaker or but quantitatively similar percentages for GalCer (Fig. 6B, right). The data shows that 9% and 10% of CD4+ and CD8+ cells, respectively, bind GalCer- and GalCer-loaded tetramers in fresh liver leukocyte populations. When NKT subsets were evaluated, both CD4 and CD8 subsets bound GalCer with a strong intensity (Fig. 6, C and D), while GalCer (C12) binding to CD4+ NKT cells was significantly stronger than it was to CD8+ cells (see Fig. 9D). Binding of unloaded tetramers was used as a control in Fig. 9, C and D. Similar experiments were performed with NKT cells that had been flow cytometrically sorted and cultured for 3–4 days in IL-2 (data not shown). These results demonstrated a similar intensity difference in binding of GalCer and GalCer (C12) to highly purified NKT cells and a lack of binding of GalCer to either NK or T cells.
|
|
GalCer or GalCer (C12) to directly interact with the TCR of NKT cells in C57BL/6 mice, we used an in vitro assay system in which control A20 cells or A20 cells transfected with CD1d were used to present GalCer or GalCer (C12) to isolated NKT vs NK cells (negative control). Cultures containing these enriched subsets of NK or NKT cells and A20 cells were stimulated with GalCer or GalCer (C12) for 24 h, and supernatants were evaluated for cytokine release (Fig. 7). IFN-, GM-CSF, and IL-4 production was strongly induced from NKT cells by GalCer, but not GalCer (C12), only when CD1d-transfected A20 cells were used. Only low levels of cytokine production were observed when transfected A20 cells were treated with GalCer (C12) or when NK cells were stimulated with either GalCer or GalCer (C12). These results clearly show that presentation of GalCer by CD1d is required for cytokine induction, while GalCer (C12) is either not efficiently presented to NKT cells by CD1d or the quality of the signal is insufficient to induce cytokine production.
|
GalCer and GalCer to activate NKT cells with ceramide presented by autologous peritoneal monocytes. Fig. 8 shows a representative experiment where liver lymphocytes (32% NKT) were mixed 1:1 with protease-peptone-induced peritoneal exudate cells after loading for 1 h at 37°C with varying doses of GalCer and GalCer (C12). Similar to the results with the CD1d-transfected cell line, GalCer rapidly induced activation and production of TNF-, IFN-, IL-4, and IL-13, while GalCer induced only low levels of TNF- and no IFN-, IL-4, and IL-13. Both GalCer and GalCer (C12) were able to induce IL-5 from liver lymphocytes. Overall, these results show that the biological effects of GalCer (C12) differ dramatically from those of GalCer in the context of CD1d and macrophage-mediated presentation to NK cells.
|
GalCer (C12) depletes NKT cells, but fails to induce rejection of allogeneic bone marrow
The accumulated results above demonstrate that GalCer (C12) is efficiently recognized by NKT cells but does not effectively activate NK cells. This conclusion implies that GalCer (C12) should not effectively induce cytokines or NK-dependent biological effects in vivo at doses that effectively reduce the number of detectable NKT cells.
Because one of the most sensitive measures of NK activity in vivo is their ability to reject bone marrow, allogeneic transfer of C57BL/6 [H-2b] marrow into BALB/c [H-2d] mice was used as a model to confirm that GalCer (C12) was unable to induce NK-mediated functions in vivo. BALB/c mice are known to exhibit only relatively weak bone marrow rejection, and thus this model can easily detect even a small amount of NK cell activation. The transfer of C57BL/6 bone marrow cells into allogeneic BALB/c mice was >50% rejected by day 3 (Fig. 9A). As expected, this event was shown to be NK mediated because prior depletion of NK cells using anti-asGM1 rabbit serum in the BALB/c recipient mice (which lack NK1.1 expression) reduces the degree of allogeneic graft rejection. The treatment of mice with GalCer, which results in NK activation, further enhanced the rejection of the donor cells to 90%. In contrast, treatment with GalCer (C12) did not significantly enhance basal rejection. To determine whether clear dose-dependent differences could be established for these two agents, additional studies were performed. Results from these studies clearly demonstrated that GalCer doses from 0.1 to 0.01 µg/mouse retained NK activating abilities (Fig. 9B), while GalCer (C12) doses from 1 to 10 µg/mouse failed to induce any in vitro NK activation. The data shown in Fig. 9C confirm that both ceramides had the expected ability to reduce the detection of NKT cells at 24 h. Previous studies suggested that even doses of GalCer (C12) 
10 µg resulted in NKT depletion with minimal effects on NK cells. Figure 9D shows two experiments where several doses of GalCer and GalCer (C12) were evaluated for enhancement of marrow rejection. Whereas GalCer can potently activate marrow rejection at a dose as low as 0.1 µg/mouse, GalCer (C12) at a dose of 10 µg, which is effective at reducing the number of NKT cells, had no enhancing effect on bone marrow graft rejection. Interestingly, a very high dose of GalCer (20 µg), which also had no appreciable NK enhancing effects (Fig. 3), did have some ability to enhance rejection of bone marrow cells suggesting that at very high doses GalCer (C12) can either induce enough augmentation of NK activity (see Table II) to mediate some effects in vivo or induce some biological effects through a non-NK-mediated mechanism. However, it is clear that in all studies a dose of 10 µg of GalCer can be used in vivo to effectively reduce detectable NKT cells without appreciable NK-activating effects as assessed by NK number, activation stage, cytolytic activity, or very sensitive NK-dependent functions in vitro.
Overall, these results confirm that GalCer (C12) does directly bind to NKT cells, but that differences in the intensity and subset affinity of this binding may contribute to its dramatically different in vitro and in vivo biological effects.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GalCer and another somewhat different ceramide, GalCer (C12), can efficiently increase apoptosis and decrease detection in vivo of a major subset of NKT cells that are reactive with ceramide-specific TCR components. However, unlike GalCer, which also potently activates bystander NK cells and induces a storm of cytokines (IL-4, IFN-, GM-CSF, and TNF-), doses of GalCer (C12) that directly impact NKT cells fail to induce NK activation or cytokine production. Thus, GalCer is a novel tool for efficiently perturbing functional NKT cells in vivo without additional, indirect host immune activation. The mechanistic basis for reduced NKT detection by GalCer is not completely understood, although this effect does require interaction with the TCR (1, 3), which in turn results in some process that induces apoptosis in at least some cells (5, 17, 24, 25). It is also possible that at least some stimulated NKT cells might endocytose their NK1.1 and TCR as part of a process that results in apoptosis or that simply renders the cells nondetectable. This conclusion is consistent with recent report (26, 27) that endocytosis of NK1.1 on NKT cells follows finding of GalCer. The studies by van Kaer (26) also indicated a rapid loss of detection and subsequent return of NKT cells after a 5-µg dose in vivo, a result that differs somewhat from results in our in vivo treatment regimen, where a more gradual return to "the norm" after 6–7 days (Fig. 1) was seen. Our studies have shown that both GalCer and GalCer (C12) induced a loss of detectable NKT cells and an increase in NKT cell apoptosis in the liver within 4–6 h. In contrast, in vitro stimulation studies showed that CD1d presentation of GalCer does not result in the stimulation of NKT cells for direct or indirect production of cytokines, suggesting 1) qualitative differences in the signals induced in NKT cells by GalCer vs GalCer (C12) and 2) qualitative differences in the processes of apoptosis and cytokine production after stimulation of NKT cells through the TCR. In addition, CD1d tetramer binding was observed with both GalCer and GalCer, but a much weaker binding and different NKT subset affinity was seen with the GalCer-loaded tetramers. Interestingly, weak binding of Ag to the TCR during thymic development is well established as an anergy/death signal. Similar processes may be involved with GalCer (C12) binding to peripheral NKT cells that trigger their loss of detection without potent downstream gene activation. Thus, the use of GalCer vs GalCer (C12) may provide a novel approach for studying the biological functions of NKT and NK populations, and perhaps NKT subsets, in vivo.
Although NKT cells represent a diverse population of lymphocytes, a great deal of information has been accumulated relating specifically to the immunological functions of the V14 and V24 NKT cell subsets. These cells have been identified as potent cytokine producers, and their appropriately controlled manipulation could be critical for initiating and sustaining immune responses against cancer or infectious diseases, or for modifying immunopathological responses. In this context, Hammond and Godfrey (7) recently summarized the implications of NKT cell function in diabetes, autoimmunity, sclerosis, systemic lupus erythematosus, rheumatoid arthritis, and Sjogren’s syndrome. The contributions of NKT cells to various types of beneficial and deleterious immune responses may be through production of cytokines and/or modification of the function of other cells such as NK cells. This NKT-NK interaction represents important cross-talk between various components of the innate immune system, and subtle disruptions of this cross-talk may have important qualitative or quantitative implications for the generation and persistence of adaptive immune responses. For example, nonobese diabetic mice have a deficiency in NKT and NK cell numbers and function and develop diabetes (4, 6, 8, 9, 11, 12, 15). The administration of GalCer protects nonobese diabetic mice from development of diabetes, but the mechanism for this effect is unclear, because presentation of GalCer in the context of CD1d directly activates NKT cells, and also indirectly stimulates NK cells. In the experimental autoimmune encephalomyelitis and multiple sclerosis models, SJL mice have a deficiency in NKT cells, but a recent report (28) has demonstrated that both 129/J and SJL mice also have NK signaling deficiencies in their activating Ly49 receptors that result in reduced cytokine production. Recent studies have shown that treatment of mice with GalCer protects mice from experimental autoimmune encephalomyelitis (13, 14, 16). As shown in the present study, both GalCer and GalCer (C12) can reduce the number of detectable NKT cells, and thus any in vivo differences in their biological effects could be due to factors or bystander activation of NK cells that are stimulated only by GalCer. A practical demonstration of the differential capabilities of these uniquely acting ceramides is illustrated in our studies by their disparate effects on bone marrow graft rejection.
The results of our studies with GalCer (C12) suggest that this agent provides a unique approach for delineating key roles of NKT cells in settings where their presence is specifically required for some therapeutic or disease-causing effect. Conversely, in settings where NK cells play a unique role, the induction of beneficial effects by GalCer (which activates NK cells) and the failure to induce such effects with GalCer (which does not activate NK cells) would provide new insights into the roles of this cell subset in innate and adaptive immune responses.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. John Ortaldo, National Cancer Institute, Center for Cancer Research, Laboratory of Experimental Immunology, Building 560, Room 31-93, Frederick, MD 21702-1201. E-mail address: ortaldo{at}mail.ncifcrf.gov
3 Abbreviations used in this paper: GalCer, -galactosylceramide; GalCer, -galactosylceramide; mCD1d, murine CD1d; FasL, Fas ligand.
Received for publication August 5, 2003. Accepted for publication November 4, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
TCR+CD4-CD8- T-cell deficiency and IDDM in NOD/Lt mice. Diabetes 46:572.[Abstract]
- or IL-12-treated mice: a major role for bone marrow in NKT cell homeostasis. Immunity 9:345.[Medline]
secretion of peripheral natural killer T cells in nonobese diabetic mice suggests new pathogenic mechanisms for insulin-dependent diabetes mellitus. J. Exp. Med. 190:963.-galactosylceramide prevents autoimmune diabetes in non-obese diabetic mice. Nat. Med. 7:1052.[Medline]
-galactosylceramide treatment prevents the onset and recurrence of autoimmune type 1 diabetes. Nat. Med. 7:1057.[Medline]
/-T cell receptor (TCR)+CD4-CD8- (NKT) thymocytes prevent insulin-dependent diabetes mellitus in nonobese diabetic (NOD)/Lt mice by the influence of interleukin (IL)-4 and/or IL-10. J. Exp. Med. 187:1047. by NK1.1+ T cells and natural killer cells is essential for the antimetastatic effect of -galactosylceramide. Blood 99:1259.-dependent natural killer cell protection from tumor metastasis. J. Exp. Med. 193:661.-galactosylceramide. Eur. J. Immunol. 30:1919.[Medline]
and NK cells, but not perforin-mediated cytotoxicity, to anti-metastatic effect of -galactosylceramide. Eur. J. Immunol. 31:1720.[Medline]
This article has been cited by other articles:
|
|
 |
|
  G. Lalazar, A. B. Ya'acov, D. M. Livovsky, M. El Haj, O. Pappo, S. Preston, L. Zolotarov, and Y. Ilan {beta}-Glycoglycosphingolipid-Induced Alterations of the STAT Signaling Pathways Are Dependent on CD1d and the Lipid Raft Protein Flotillin-2 Am. J. Pathol., April 1, 2009; 174(4): 1390 - 1399. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Lalazar, A. Ben Ya'acov, N. Eliakim-Raz, D. M. Livovsky, O. Pappo, S. Preston, L. Zolotarov, and Y. Ilan {beta}-Glycosphingolipids-mediated lipid raft alteration is associated with redistribution of NKT cells and increased intrahepatic CD8+ T lymphocyte trapping J. Lipid Res., September 1, 2008; 49(9): 1884 - 1893. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ebensen, C. Link, P. Riese, K. Schulze, M. Morr, and C. A. Guzman A Pegylated Derivative of {alpha}-Galactosylceramide Exhibits Improved Biological Properties J. Immunol., August 15, 2007; 179(4): 2065 - 2073. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Subleski, V. L. Hall, T. C. Back, J. R. Ortaldo, and R. H. Wiltrout Enhanced Antitumor Response by Divergent Modulation of Natural Killer and Natural Killer T Cells in the Liver. Cancer Res., November 15, 2006; 66(22): 11005 - 11012. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. R. Brutkiewicz CD1d Ligands: The Good, the Bad, and the Ugly J. Immunol., July 15, 2006; 177(2): 769 - 775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Kjer-Nielsen, N. A. Borg, D. G. Pellicci, T. Beddoe, L. Kostenko, C. S. Clements, N. A. Williamson, M. J. Smyth, G. S. Besra, H. H. Reid, et al. A structural basis for selection and cross-species reactivity of the semi-invariant NKT cell receptor in CD1d/glycolipid recognition J. Exp. Med., March 20, 2006; 203(3): 661 - 673. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Oki, C. Tomi, T. Yamamura, and S. Miyake Preferential Th2 polarization by OCH is supported by incompetent NKT cell induction of CD40L and following production of inflammatory cytokines by bystander cells in vivo Int. Immunol., December 1, 2005; 17(12): 1619 - 1629. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Ortaldo, R. T. Winkler-Pickett, E. W. Bere Jr, M. Watanabe, W. J. Murphy, and R. H. Wiltrout In Vivo Hydrodynamic Delivery of cDNA Encoding IL-2: Rapid, Sustained Redistribution, Activation of Mouse NK Cells, and Therapeutic Potential in the Absence of NKT Cells J. Immunol., July 15, 2005; 175(2): 693 - 699. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ota, K. Takeda, H. Akiba, Y. Hayakawa, K. Ogasawara, Y. Ikarashi, S. Miyake, H. Wakasugi, T. Yamamura, M. Kronenberg, et al. IFN-{gamma}-mediated negative feedback regulation of NKT-cell function by CD94/NKG2 Blood, July 1, 2005; 106(1): 184 - 192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Smyth, M. E. Wallace, S. L. Nutt, H. Yagita, D. I. Godfrey, and Y. Hayakawa Sequential activation of NKT cells and NK cells provides effective innate immunotherapy of cancer J. Exp. Med., June 20, 2005; 201(12): 1973 - 1985. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Bezbradica, A. K. Stanic, N. Matsuki, H. Bour-Jordan, J. A. Bluestone, J. W. Thomas, D. Unutmaz, L. Van Kaer, and S. Joyce Distinct Roles of Dendritic Cells and B Cells in Va14Ja18 Natural T Cell Activation In Vivo J. Immunol., April 15, 2005; 174(8): 4696 - 4705. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. V. Parekh, A. K. Singh, M. T. Wilson, D. Olivares-Villagomez, J. S. Bezbradica, H. Inazawa, H. Ehara, T. Sakai, I. Serizawa, L. Wu, et al. Quantitative and Qualitative Differences in the In Vivo Response of NKT Cells to Distinct {alpha}- and {beta}-Anomeric Glycolipids J. Immunol., September 15, 2004; 173(6): 3693 - 3706. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Gumperz CD1d-restricted "NKT" cells and myeloid IL-12 production: an immunological crossroads leading to promotion or suppression of effective anti-tumor immune responses? J. Leukoc. Biol., August 1, 2004; 76(2): 307 - 313. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Campos-Martin, M. Gomez del Moral, B. Gozalbo-Lopez, J. Suela, and E. Martinez-Naves Expression of Human CD1d Molecules Protects Target Cells from NK Cell-Mediated Cytolysis J. Immunol., June 15, 2004; 172(12): 7297 - 7305. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), 1 (
), or 0.1 (•) of 
), and their liver lymphocytes were evaluated for cell functions at various times (days). All points represent pooled cells from three to five mice per group. A, NK activity measured by 51Cr release from YAC-1 expressed in lytic units. B, NK cells based on flow cytometry forward scatter measurements. C measures CD69 expression on NK cells (NK1.1+, CD3-). D, Liver NK cell numbers. E, NKT cells numbers. F, The relative percentage NK cells in liver. G, The relative percentage NKT (NK1.1+, CD3+) cells in liver. H, Total leukocyte number in liver. I, Serum IFN-
), 
or 48 h 
, 1 µg/mouse, or 
