Published online before print February 20, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Medical Microbiology and Immunology, University of Wisconsin Medical School, Madison, Wisconsin, USA; and
School of Biosciences, University of Birmingham, Birmingham, United Kingdom
1 Correspondence: Department of Medical Microbiology and Immunology, University of Wisconsin Medical School, 1300 University Ave., Madison, WI 53705, USA. E-mail: jegumperz{at}wisc.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: autoreactive • GM-CSF • IL-13
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The phenotypic and functional variations of DCs suggest they derive from multiple progenitor cell types. Myeloid lineage cells in the bone marrow are known to give rise to precursor DCs, which enter the bloodstream and renew immature DC populations continuously within the tissues [3 ]. Peripheral blood monocytes are an abundant leukocyte subset, which can also serve as a DC precursor population in vivo [3 ]. However, monocytes are developmentally plastic, and most appear to differentiate into macrophages rather than DCs [4 ]. The physiological signals that determine the course of monocyte differentiation are not fully characterized. Migration from tissues across an endothelial cell layer has been shown to lead to DC differentiation, whereas monocytes that remained in the subendothelial matrix became macrophages [5 ]. It is also known that exposure to particular cytokines can direct the course of monocyte differentiation: M-CSF stimulates macrophage differentiation, and GM-CSF, in combination with IL-4 or IL-13, leads to formation of immature DCs [6 7 8 9 10 ]. The physiological situations, in which monocytes are exposed to cytokines that direct DC differentiation, remain unclear.
Whereas DCs are specialized APC, which can potently activate naive T cells, monocytes lack high expression levels of costimulatory cell surface molecules and do not process and present peptide antigens efficiently and are therefore not usually effective stimulators of T cell responses. However, monocytes constitutively express CD1d, a nonclassical antigen-presenting molecule and can serve as efficient APC for CD1d-restricted NKT cells [11 , 12 ], which are a unique subpopulation of T lymphocytes that recognize lipids and glycolipids as antigens [13 ]. NKT cells have important regulatory functions and can promote proinflammatory as well as tolerizing immune responses [14 ]. Consistent with their pleiotropic functions, NKT cells efficiently produce Th1 and Th2 cytokines [14 ]. Stimulation of human NKT cells directly ex vivo led to the production of a wide variety of cytokines, including GM-CSF, IL-4, and IL-13 [15 ]. Hence, NKT cells may be one source that produces these cytokines in vivo.
NKT cells can be activated by recognition of specific microbial lipid antigens presented by CD1d and thus, may function as part of the antimicrobial defense system in vivo [16
, 17
]. However, NKT cells are unusual in that they can also be activated by self-lipids presented by CD1d [18
, 19
]. Their autoreactivity allows NKT cells to be activated in an innate-like manner during the course of microbial infections, as inflammatory cytokines, which are produced by DCs in response to microbial products, costimulate autoreactive NKT cell IFN-
secretion [16
, 20
]. It is not clear whether NKT cell autoreactivity also serves a physiological purpose when there is no immunological challenge.
We hypothesized that interactions with monocytes might activate NKT cells to produce factors that influence monocyte differentiation. Monocytes and NKT cells share expression of the chemokine receptors CCR5, CCR2, and CX3CR1, and both are recruited to sites of tissue inflammation [21 , 22 ]. In addition, monocytes and NKT cells are present in blood, lymphoid tissues, and bone marrow under noninflamed conditions [23 ]. Thus, NKT cells and monocytes are likely to colocalize in vivo under normal and inflammatory conditions and may undergo self- or foreign antigen-driven interactions in either of these contexts. Here, we investigate how NKT cells affect monocyte differentiation.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Generation and maintenance of T cell clones
Human NKT cell clones were generated from PBMC samples stained with 
-galactosylceramide (-GalCer)-loaded human CD1d tetramers, as described previously [20
, 24
]. NKT clones were maintained in T cell medium (RPMI-1640 medium supplemented with 2 mM L-glutamine, 100 µg/ml each penicillin and streptomycin, 10% FBS, 5% bovine calf serum, and 5% human AB serum) and stimulated periodically with irradiated, allogenic PBMC, 250 ng/ml PHA, and 400 U/ml recombinant human (rh)IL-2. Tetanus toxoid (TT)-specific clones were obtained by stimulating human PBMC with 0.01 flocculation U/ml TT antigen (Massachusetts Biological Laboratories, University of Massachusetts Medical School, Worcester, MA, USA) in T cell medium for 14 days, followed by limiting dilution cloning. Resulting clones were confirmed to be dependent on TT antigens for activation by testing their cytokine responses to autologous DCs in the presence and absence of TT antigen, and MHC Class II-restriction was confirmed by functional blocking with anti-MHC Class II mAb (L243 and IVA12).
Short-term expansion of T cell lines from PBMC
Freshly isolated PBMC were stained with -GalCer-loaded CD1d tetramers and Alexa 647-conjugated anti-CD3, as described [15
]. CD3+ events, which were positive or negative for CD1d tetramer staining, were sorted separately and cultured for 14 days in T cell medium with irradiated, allogenic PBMC, PHA, and 400 U/ml rhIL-2.
Monocyte differentiation experiments
T cells were cultured in medium lacking IL-2 for 24–48 h prior to all experiments, which were carried out in RPMI-1640 medium supplemented with 2 mM L-glutamine, 100 µg/ml each penicillin and streptomycin, and 10% FBS. In coculture differentiation experiments, 1 x 106-purified monocytes were coincubated with 2 x 106 NKT cells in 2 ml culture medium, and differentiation of the monocytes was assessed after 3 days of culture. Transwell differentiation experiments were performed in 1.5 ml culture medium, and 1 x 106 T cells were coincubated with 0.5 x 106 "stimulator" monocytes in the transwell insert and 1 x 106 "responder" monocytes in the lower transwell. Unless otherwise noted, the monocytes in the lower well were exposed to the cells in the upper well inserts for 8 h, and then the inserts were removed, and the lower well monocytes were cultured for an additional 64 h to allow differentiation to occur. In both cases, negative control monocytes were incubated in medium alone, and positive control monocytes were incubated in medium containing 300 U/ml rhM-CSF and 200 U/ml rhIL-4. Where specified, 20 U/ml rhIL-2 was added to the medium, or the stimulator monocytes were pulsed for 2 h with 5 ng/ml -GalCer. The following antibodies were used at a final concentration of 10 µg/ml for blocking experiments: the anti-CD1d mAb CD1d42.1 (murine IgG1) [11
], the anti-GM-CSF mAb (Clone 3209, murine IgG1), the anti-IL-13 mAb (JES10-5A2, rat IgG1), the anti-CD40 ligand (anti-CD40L) mAb (24-31, murine IgG1), or isotype-matched, control mAb (MOPC21 murine IgG1 and RTK2071 rat IgG1). To induce maturation further, the immature DCs were treated with 250 ng/ml Salmonella typhimurium LPS for 2 days.
Flow cytometric analysis
Fluorescently labeled, commercially available antibodies against the following markers were used for flow cytometric staining: CD14 (Clone M5E2, mIgG2a), DC-specific ICAM-grabbing nonintegrin (DC-SIGN; DCN46, mIgG2b), CD40 (HB14, mIgG1), CD1a (HI149, mIgG1), CD1b (M-T101, mIgG1), CD86 (FUN-1, mIgG1), CD83 (HB15e, mIgG1), anti-CCR7 (150503, mIgG2a), anti-GM-CSF (BVD2-21C11, rat IgG2a), anti-IL-13 (JES10-5A2, rat IgG1), and negative control murine IgG1 (MOPC-21), murine IgG2a (G155-178), murine IgG2b (MPC-11), rat IgG2a (RTK2758), and rat IgG1 (RTK2071). Anti-CD3 mAb (SPVT3b, mIgG2a) and a negative control IgG2a mAb (UPC10) were conjugated to normal human serum (NHS)-activated Alexa 647, according to the manufacturer’s protocol (Invitrogen, Carlsbad, CA, USA). Staining was performed as described previously [15
]. Samples were analyzed on a Becton Dickinson LSRII flow cytometer using FACSDiva software (Becton Dickinson, San Jose, CA, USA) to collect the data, and final analysis was performed using Flowjo software (TreeStar Inc., Ashland, OR, USA).
Immunofluorescence
To obtain pure monocyte-derived cell populations for analysis, immunofluorescence was performed on lower well cells from the transwell culture system described above. The cells were fixed and stained using Cytofix-Cytoperm (BD Biosciences, San Jose, CA, USA), according to the manufacturer’s protocol, and then spun onto a slide for microscopic analysis. Staining was performed with anti-MHC Class II mAb (L243, murine IgG2a) and the negative control IgG2a mAb (UPC-10) conjugated to NHS-activated Alexa 633 (Invitrogen) and antilysosome-associated membrane protein-1 (anti-LAMP-1) mAb (H4A3, murine IgG1) and the negative control IgG1 mAb (MOPC-21) labeled with Alexa 488 using the anti-IgG1 Zenon reagent (Invitrogen). Images were taken using a Zeiss immunofluorescence microscope at x100 magnification. OpenLab software was used to take 0.2 µm Z-stack section pictures, and these stacks were used for three-dimensional reconstruction.
Analysis of NKT cell supernatants
NKT cells were stimulated by exposure to APC or plate-bound, anti-CD3 mAb as indicated. CD1d-transfected 721.221 cells were generated as described [25
], and NKT cell responses were assessed by coculture for 24 h with the untreated or -GalCer (100 ng/ml)-pulsed transfectants as described previously [24
]. Freshly isolated monocytes were untreated or pulsed for 2 h with the indicated concentrations of -GalCer. Commercially available ELISAs were used to assess the concentrations of GM-CSF, IL-13, IFN-, and IL-4 in NKT + APC culture supernatants by comparison with standard curves generated using the appropriate recombinant cytokines (all from Peprotech, Rocky Hill, NJ, USA). For analysis of transwell culture supernatants (see Table 1
), the concentrations of GM-CSF, IL-4, IL-13, IL-6, IL-1ß, IFN-, TNF-, IFN-, IL-10, CD40L, IL-12p70, receptor activator of NF-
B ligand (RANK-L), Fas ligand (Fas-L), G-CSF IL-15, and TGF-ß were determined by a Multiplex ELISA assay (Pierce, Rockford, IL, USA).
|
View this table: [in a new window] |
Table 1. Correlation of Soluble Factors with the Induction of DC Differentiation
|
-GalCer-loaded CD1d tetramer, A647-conjuated, anti-CD3 mAb, and A700-conjugated, anti-CD4 in a PBS buffer containing 1% BSA and 0.05% NaN3. The samples were then washed and fixed with Cytofix/Cytoperm (BD Biosciences) and stained with PE-conjugated antibodies specific to GM-CSF or IL-13 or isotype-matched, negative control mAb. |
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-GalCer-loaded CD1d tetramers, as described previously [24
]. As a positive control to induce DC differentiation, the monocytes were cultured in medium containing GM-CSF and IL-4, and as a negative control, they were kept in medium alone. After 3 days, the phenotype of the myeloid cells was assessed by flow cytometric analysis of CD14, DC-SIGN, CD86, CD83, CD40, and CD1a, -b, and -c. As expected, monocytes, which were cultured with GM-CSF and IL-4, acquired a phenotype characteristic of immature DCs, including down-regulation of CD14, combined with up-regulation of DC-SIGN, the CD1 molecules, CD40, and some up-regulation of CD86 but little or no increase in expression of the mature DC marker CD83 (Fig. 1A
). Most of the monocytes, which were cultured with NKT cell clones, also expressed cell surface markers that are consistent with DC differentiation, although a minor fraction (ranging from 2% to 24% and generally less than 20%) failed to differentiate (Fig. 1A)
. In addition, the cell surface phenotype of the NKT cell-cocultured DCs differed slightly from those that were cultured with GM-CSF and IL-4. In particular, CD40 and CD86 expression tended to be somewhat higher, and there was little up-regulation of CD1a, although CD1b and CD1c were highly expressed (Fig. 1A)
. Similar results were obtained consistently with nine out of 10 NKT cell clones tested. In contrast, monocytes, which were maintained in culture medium alone, retained high expression of CD14 and did not up-regulate other markers, demonstrating that incubation in the culture medium alone did not induce monocyte differentiation (Fig. 1A)
. Thus, exposure to NKT cells appeared to induce most CD14+ peripheral blood monocytes to differentiate into cells resembling immature DCs, although a minor fraction appeared refractory to NKT cell differentiation signals.
 View larger version (33K): [in a new window] |
Figure 1. Monocytes cultured with NKT cells differentiate into immature DCs. (A) Flow cytometric analysis of monocytes cultured in medium alone (top row), with GM-CSF and IL-4 (middle row), or with a human NKT cell clone (bottom row) for 72 h. Anti-CD3 and 4',6-diamidino-2-phenylindole staining were used to gate out T cells and dead cells, respectively. The filled histograms show staining of cells with indicated marker-specific antibodies; open histograms show staining with the respective isotype-matched, negative control antibodies. Similar results were reproducibly observed for nine out of 10 NKT cell clones tested. (B) Immunofluorescence analysis of MHC Class II (red) and LAMP-1 (green) staining of monocytes cultured with medium alone (top row), with GM-CSF and IL-4 (middle row), or with a NKT cell clone (bottom row). Colocalization of MHC Class II and LAMP-1 in the overlaid (merged) images is indicated by orange or yellow color. The results are from one representative experiment out of three independent experiments. Similar results were obtained with three different NKT cell clones.
|
Another characteristic of immature DCs is that they are readily stimulated to mature by exposure to microbial products. Therefore, we tested the ability of the NKT cell-cocultured cells to mature further. The GM-CSF/IL-4 and NKT cell-stimulated myeloid cells were treated with 250 ng/ml LPS and after 48 h, assayed by flow cytometry for expression of the mature DC markers CD83, CCR7, and CD86. Both had up-regulated these markers, consistent with further differentiation into mature DCs (Fig. 2A ). Moreover, immunofluorescence microscopy revealed that in both cases, the MHC Class II staining had relocalized completely to the cell surface, as is typical of mature DCs (Fig. 2B) . Thus, exposure of freshly isolated monocytes to NKT cells resulted in differentiation into immature DCs, and these DCs could mature further upon stimulation from a microbial product.
 View larger version (20K): [in a new window] |
Figure 2. The immature DCs can be matured further by LPS. (A) Flow cytometric analysis of the immature DCs, which were generated by GM-CSF and IL-4 treatment (upper panel) or by culture with a NKT cell clone (lower row) after exposure to LPS. Filled histograms show staining with specific mAb; open histograms show staining with the respective isotype-matched, negative control antibodies. (B) Immunofluorescence analysis of MHC Class II (red) and LAMP-1 (green) staining of the DCs shown in A. The figure shows the overlay of MHC Class II and LAMP-1 staining. Results shown in each case are from one representative experiment out of three independent analyses.
|
 View larger version (27K): [in a new window] |
Figure 3. Induction of monocyte differentiation is dependent on CD1d recognition and mediated by factors secreted by activated NKT cells. (A) A transwell culture system was used to test the requirement for cell contact and CD1d recognition. Monocytes (monos) alone were placed in the lower transwells, and the cells indicated on the x-axis were placed in the transwell inserts. The plot shows CD14 (left y-axis) and DC-SIGN (right y-axis) expression [given as mean fluorescence intensity (MFI)] of the cells from the lower wells of transwell plates after 72 h of exposure to the transwell inserts. The CD14 and DC-SIGN staining of monocytes, which were cultured in parallel with GM-CSF and IL-4 or in medium alone, is shown for comparison. The results are from an experiment performed with NKT Clone J24L.17 and are representative of three independent experiments. Similar results were obtained with two different NKT cell clones. (B) A NKT cell clone (J3N.5) was stimulated for 24 h by culture with the indicated concentrations of plate-bound, anti-CD3 mAb, and the culture supernatants were filtered and added to cultures of purified monocytes at a 1:1 ratio with fresh culture medium. The plot shows the results of flow cytometric analysis of the monocytes after 3 days of culture with the NKT cell supernatants. (C) Kinetics of NKT cell:monocyte contact time required to induce DC differentiation. An autoreactive NKT cell clone (Clone J24L.17) was placed in transwell inserts with untreated monocytes (left panel) or monocytes, which were pulsed for 2 h with 5 ng/ml -GalCer (right panel), and untreated monocytes were placed in the lower transwells. The inserts were removed at the times shown on the x-axis, and the lower well cells were cultured for an additional 3 days to allow differentiation to occur. The plots show the CD14 (left y-axis) and DC-SIGN (right y-axis) expression of the resulting lower well cells. Similar results were obtained in three independent experiments.
|
To determine the length of monocyte contact time needed for NKT cell activation to occur, NKT cells were coincubated with stimulator monocytes in transwell inserts, and responder monocytes were placed in the lower transwells, and at varying time-points, the transwell inserts were removed. The responder monocytes in the lower wells were then cultured further to allow differentiation to occur, and after a total of 3 days, the cell surface phenotype of the responder monocytes was determined by flow cytometric analysis. This experiment revealed that 4–6 h of exposure to transwell inserts containing NKT cells with stimulator monocytes in the absence of added antigens was sufficient to induce DC differentiation of the responder monocytes (Fig. 3C
, left panel). If the stimulator monocytes were pulsed with -GalCer prior to culture with the NKT cells, only 2 h of exposure was required for differentiation of the responder monocytes to occur (Fig. 3C
, right panel). Hence, autoreactive activation of NKT cells by monocytes rapidly stimulated secretion of factors that induce monocyte differentiation, and NKT activation by -GalCer resulted in even faster production of the active factors.
The ability to induce monocyte differentiation appears specific to NKT cells
We tested whether MHC Class II-restricted, TT-specific T cell clones could induce DC differentiation in response to untreated or antigen-pulsed, autologous monocytes. When the TT clones were cultured with untreated, autologous monocytes in the upper wells of transwell plates, no DC differentiation of monocytes in the lower wells was observed (Fig. 4A
). Indeed, CD14 expression levels on the lower well monocytes tended to increase compared with monocytes that were kept in culture medium alone (Fig. 4A)
. When the TT clones were incubated with antigen-pulsed, autologous monocytes, slight up-regulation of DC-SIGN was observed on the responder monocytes, but CD14 was not down-regulated compared with untreated monocytes, suggesting that although some activation may have occurred, it did not result in DC differentiation (Fig. 4A)
. Similar results were obtained using two other MHC Class II-restricted TT clones derived from a different donor and using a polyclonal line derived by stimulation with toxic shock syndrome toxin-1 superantigen (data not shown). These results indicate that DC differentiation is not an intrinsic consequence of T cell interactions with monocytes.
 View larger version (34K): [in a new window] |
Figure 4. The ability to induce monocyte differentiation is specific to NKT cells. (A) MHC Class II-restricted, TT-specific T cell clones (clone names are shown on the x-axis) were tested for the ability to induce DC differentiation when stimulated by untreated, autologous monocytes or autologous monocytes that were pulsed with purified TT protein antigen (Ag). Monocytes alone were placed in the lower transwells, and the cells indicated on the x-axis were placed in the transwell inserts. The plot shows CD14 (left y-axis) and DC-SIGN (right y-axis) expression of the cells from the lower wells of transwell plates after 72 h of culture. The CD14 and DC-SIGN staining of control monocytes, which were cultured in parallel with GM-CSF and IL-4 or in medium alone, is shown for comparison. Figure shows 4 TT clones derived from an individual donor (SGTT-2, SGTT-3, SGTT-4, and SGTT-6). Similar results were obtained using two other TT clones. (B) Short-term polyclonal T cell lines were established from PBMC, which were stained and gated as shown in the left panel. The plots on the right show the CD14 and DC-SIGN expression by monocytes from the lower wells of transwell plates, which were exposed to inserts containing T cells alone, T cells with untreated monocytes in the presence or absence of IL-2, or T cells with monocytes, which were pulsed with -GalCer. Similar results were obtained in three independent experiments.
|
-GalCer-loaded CD1d tetramers and expanded by short-term culture in vitro. CD3+ cells, which were negative for CD1d tetramer staining, were also sorted and expanded. The resulting NKT-enriched and NKT-depleted T cell lines were tested for their ability to induce DC differentiation in the transwell system. The polyclonal NKT cells alone in the upper well failed to induce differentiation of the monocytes in the lower well, but partial differentiation was observed when the upper wells contained the NKT cells with autologous monocytes, and efficient differentiation occurred when 20 U/ml rIL-2 was added or when the stimulating monocytes were pulsed with -GalCer (Fig. 4B)
. In contrast, the NKT-depleted T cells did not induce monocyte differentiation under any of these conditions (Fig. 4B)
. These results show that the peripheral blood NKT cell population is enriched for cells that can promote monocyte differentiation. Moreover, although some classical T cells (e.g., Th2-biased T cells) might be able to stimulate monocyte differentiation into DCs in the presence of cognate antigens, T cells, which mediate this effect on the basis of autoreactive activation, appear to be infrequent or absent in the non-CD1d-restricted T cell population.
Role of NKT cell autoreactivity
Our results indicated that the ability of the NKT cells to induce DC differentiation was dependent on CD1d recognition but did not require the addition of glycolipid antigens. We have previously observed that human NKT cell clones vary in their autoreactive responses to CD1d+ APC [24
]. Autoreactive responses to CD1d-transfected 721.221 cells varied from less than 2% to greater than 80% of maximal cytokine secretion (defined as the response to the transfectant pulsed with a saturating concentration of -GalCer; Supplemental Fig. 1). We have also found previously that weak, autoreactive NKT cell cytokine secretion can be augmented significantly by costimulatory cytokines such as IL-12 and IL-2 [20
]. Thus, the ability to become autoreactively activated by monocytes might vary within the NKT cell population and might be enhanced by factors that provide additional NKT cell stimulation.
To investigate this, different NKT cell clones were stimulated in the upper wells of transwell plates by exposure to untreated monocytes in the presence or absence of IL-2 or by -GalCer-pulsed monocytes, and the differentiation of responder monocytes from the lower wells was assessed. Some of the NKT cell clones (J24L.10, J24L.17, and J3N.5) efficiently converted the lower well monocytes into immature DCs, when they were stimulated by untreated monocytes, and this was not improved significantly by the presence of IL-2 but was somewhat enhanced if the NKT cells were stimulated by -GalCer-pulsed monocytes (Fig. 5A
). Other NKT clones (J3N.4 and J24N.22) did not promote significant differentiation in response to untreated monocytes but stimulated partial differentiation in the presence of IL-2 and efficient differentiation in response to -GalCer-pulsed monocytes (Fig. 5B)
. Finally, some NKT cell clones (J3N.1 and J24N.70) failed to promote differentiation when stimulated with monocytes in the presence or absence of IL-2 but were efficiently able to induce differentiation in response to -GalCer-pulsed monocytes (Fig. 5C)
. These differences appeared to correlate with the ability of the NKT cell clones to become autoreactively activated to secrete GM-CSF in response to CD1d-transfected 721.221 cells (Fig. 5D)
. These results suggest that although highly autoreactive NKT cells may not require further stimulation, costimulation or presentation of foreign antigens may be necessary for other NKT cells to become sufficiently activated to induce monocyte differentiation.
 View larger version (33K): [in a new window] |
Figure 5. The stimulation requirements to induce monocyte differentiation vary among NKT cell clones. Flow cytometric analysis of monocytes cultured in the lower wells of transwell plates, which were exposed to inserts containing (A) NKT Clone J24L.17, (B) Clone J24N.22, and (C) Clone J3N.1 with monocytes in medium alone (left column), in medium containing 20 U/ml IL-2 (middle column), or -GalCer-pulsed monocytes (right column). (D) Clonal variation in CD1d-dependent autoreactivity. The maximal cytokine secretion by each clone is approximated by the GM-CSF response to CD1d-transfected 721.221 cells, which were pulsed with a saturating concentration of -GalCer (open bars), and the autoreactive response is shown by GM-CSF production stimulated by untreated CD1d/721.221 transfectants (solid bars). No significant cytokine secretion was observed in response to untransfected 721.221 cells, and the NKT cell responses to the CD1d transfectants were blocked by inclusion of anti-CD1d mAb but not negative control mAb (data not shown). Similar results were obtained in three independent experiments.
|
(Fig. 6A
). The NKT cell cytokine production was inhibited specifically by addition of an anti-CD1d mAb, demonstrating that it required CD1d recognition (Fig. 6A)
. When the autoreactive NKT cell clone was stimulated by -GalCer-pulsed monocytes, secretion of GM-CSF, IL-13, and IFN- increased, and IL-4 production was induced (Fig. 6B)
. Activation by -GalCer appeared to promote NKT cell secretion of IFN- and IL-4 disproportionately, as observed by the more sigmoidal dose-response curves for these cytokines, compared with the nearly linear curves for GM-CSF and IL-13 (Fig. 6B)
. Hence, autoreactive activation preferentially elicited NKT cell production of GM-CSF and IL-13, and -GalCer stimulation particularly enhanced secretion of IFN- and IL-4.
 View larger version (21K): [in a new window] |
Figure 6. Autoreactive activation of NKT cells by monocytes preferentially stimulates GM-CSF and IL-13 secretion. (A) An autoreactive NKT cell clone (J24L.17) was cultured for 24 h with freshly isolated, untreated monocytes in the presence of an anti-CD1d mAb (solid bars) or isotype-matched, negative control mAb (open bars), and supernatants were analyzed for the indicated cytokines by ELISA. Similar results were obtained in three independent analyses. (B) Monocytes were pulsed with the indicated concentrations of -GalCer and then cultured for 24 h with Clone J24L.17, and cytokine concentrations in the culture supernatants were determined by ELISA. The amounts of GM-CSF, IL-13, and IFN- (solid symbols) are shown according to the scale on the left y-axis, as the fold increase over the amount of each cytokine produced in response to untreated monocytes (i.e., the autoreactive response), and the IL-4 concentrations (open symbols) are shown on the scale on the right y-axis. The plots show the mean cytokine concentrations from triplicate samples, and the error bars show the standard deviations of the means.
|
 View larger version (22K): [in a new window] |
Figure 7. DC differentiation induced by autoreactive NKT cells is a result of GM-CSF and IL-13. (A) NKT cells (Clone J24L.17) were cultured with monocytes in the top wells of transwell plates, and differentiation of monocytes from the lower transwells was assessed by flow cytometric analysis of CD14 and DC-SIGN expression. Assays were performed in the presence of anti-GM-CSF and IL-13-blocking mAb or isotype-matched, negative control mAb, as shown on the x-axis. Results shown are from one representative experiment out of three independent analyses. (B–D) Freshly isolated monocytes were cultured with the indicated recombinant cytokines: (B) GM-CSF and IL-13 together; (C) GM-CSF alone; (D) IL-13 alone. Cytokine concentrations are shown on the x-axes and were chosen because they are similar to amounts detected in culture supernatants of autoreactively activated NKT cell clones. The plots show the CD14 (left y-axis) and DC-SIGN (right y-axis) expression after 3 days of culture. Similar results were obtained in three independent experiments.
|
-GalCer-loaded CD1d tetramers to detect CD1d-restricted NKT cells and then fixed, permeabilized, and stained with antibodies for intracellular GM-CSF and IL-13. Control samples were incubated without monocytes, with monocytes in the presence of 20 U/ml rIL-2, or with monocytes that had been pulsed with -GalCer. Intracellular cytokine staining for GM-CSF and IL-13 was near background levels in the NKT cell subset of samples, which were incubated without the addition of monocytes (Fig. 8
). In contrast, the NKT cells of samples that were exposed to monocytes showed five- to tenfold increases in the number of cells that stained positively for intracellular GM-CSF and IL-13, and addition of IL-2 augmented the staining (Fig. 8)
. These results show that fresh peripheral blood NKT cells can become activated by monocytes to produce cytokines that induce DC differentiation and thus provide the first evidence that such NKT cell autoreactive responses are likely to be a component of human immune function in vivo.
 View larger version (37K): [in a new window] |
Figure 8. CD1d-restricted NKT cells in fresh peripheral blood samples produce GM-CSF and IL-13 in response to autologous monocytes. PBMC were depleted of monocytes and B cells and cultured for 12 h alone (top row), with a 1:1 ratio of untreated monocytes (second row), with monocytes and 20 U/ml IL-2 (third row), or with monocytes that were pulsed with -GalCer (bottom row). The samples were stained with -GalCer-loaded CD1d tetramers and anti-CD3, then fixed and permeabilized, and stained with anti-GM-CSF, anti-IL-13, or matched isotype control mAb. The plots are gated on the CD3+/CD1d tetramer+ subset and show the intracellular cytokine staining. Results shown are from one representative experiment out of four independent analyses.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and IL-4 were only stimulated inefficiently. Thus, the autoreactive activation of NKT cells by monocytes appears to preferentially stimulate the secretion of cytokines that in turn induce monocytes to differentiate into DCs.
Previous studies have found that administration of -GalCer leads to the appearance of mature myeloid DCs in murine tissues, suggesting that NKT cells and myeloid lineage cells interact closely in vivo, and activated NKT cells promote DC maturation [30
, 31
]. However, it is not clear whether autoreactively activated NKT cells also affect DC differentiation in vivo. There do not appear to be gross defects in DC subsets in mice, which are genetically deficient for NKT cells, although there may be subtle abnormalities [32
]. However, as DCs are likely to arise from multiple progenitor cell types and differentiate via several pathways, it is likely that a defect in one differentiation pathway would not be readily detected. Our data show that human NKT cells can produce cytokines in response to autoreactive activation by monocytes directly ex vivo. Thus, we speculate that autoreactive NKT cells may continuously recruit monocytes into the DC lineage in vivo.
The potential for autoreactive T cells to cause tissue destruction in autoimmune disease is well established, but how the capacity for autoreactive activation contributes to the functions of regulatory T cells is not understood. NKT cells have been shown to protect against development of certain autoimmune diseases such as Type I diabetes and multiple sclerosis [33 , 34 ], but the mechanism by which they exert their effects is not clear. As immature DCs have been shown to tolerate naïve T cells [35 , 36 ], NKT cell autoreactivity may serve to constitutively promote peripheral tolerance by driving the formation of immature DCs. Whether the selective production of IL-13 over IL-4 by autoreactively activated NKT cells also contributes specifically to a tolerogenic outcome is not clear. It has been shown in a murine tumor rejection model that the suppressive effect of NKT cells on tumor-specific CTLs is connected to their production of IL-13 but not IL-4, supporting the possibility that IL-13 plays a unique role in NKT cell-mediated immune silencing [37 ]. However, previous studies of the functions of human monocyte-derived DCs, which were differentiated by exposure to GM-CSF and IL-13, have generally indicated that they were functionally similar to those that were differentiated by GM-CSF and IL-4 [10 ]. Thus, an important area of future investigation will be to determine whether DCs that differentiate in response to autoreactively activated NKT cells have particular functional properties.
The self-antigens, which physiologically activate NKT cells, are not well characterized. A lysosomally processed form of a globoside lipid has been identified recently as the antigen that is responsible for the thymic selection of most murine NKT cells [19 ]. However, NKT cells have been found to differ in their ability to recognize purified lipids presented by CD1d, suggesting that there is clonal variation in TCR specificity, and NKT cell clones are heterogeneous in their autoreactive responses to CD1d+ APC [18 , 24 , 38 , 39 ]. The functional impact of the variation in autoreactivity among NKT cells is not clear. Our findings indicate that only a fraction of NKT cells have strong enough autoreactive responses to drive monocyte differentiation in the absence of additional stimulation. Thus, an important consequence of the heterogeneity in NKT cell autoreactive responses is that it may result in an increased rate of DC differentiation under inflammatory conditions or during microbial infections that introduce highly activating NKT cell antigens, as these conditions may activate more of the NKT cell population to stimulate the differentiation of monocytes into immature DCs. This may also provide a critical mechanism for maintaining the numbers of DCs in inflamed tissues at sites of microbial challenge.
Received December 6, 2006; revised January 25, 2007; accepted January 29, 2007.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
J. Exp. Med. 179,1109-1118 24(+)V ß 11(+) NKT cell subsets with distinct cytokine-producing capacity Blood 100,11-16-galactosylceramide rapidly induces the full maturation of dendritic cells in vivo and thereby acts as an adjuvant for combined CD4 and CD8 T cell immunity to a coadministered protein J. Exp. Med. 198,267-27914-J281 transgenic nonobese diabetic mice against diabetes J. Exp. Med. 188,1831-1839 14-J 281 NKT cells naturally regulate experimental autoimmune encephalomyelitis in nonobese diabetic mice J. Immunol. 168,6007-6011This article has been cited by other articles:
 |
|
 |
|
  H.-J. Ko, J.-M. Lee, Y.-J. Kim, Y.-S. Kim, K.-A Lee, and C.-Y. Kang Immunosuppressive Myeloid-Derived Suppressor Cells Can Be Converted into Immunogenic APCs with the Help of Activated NKT Cells: An Alternative Cell-Based Antitumor Vaccine J. Immunol., February 15, 2009; 182(4): 1818 - 1828. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Wang, X. Chen, L. Rodenkirch, W. Simonson, S. Wernimont, R. M. Ndonye, N. Veerapen, D. Gibson, A. R. Howell, G. S. Besra, et al. Natural killer T-cell autoreactivity leads to a specialized activation state Blood, November 15, 2008; 112(10): 4128 - 4138. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Patterson, A. Chaidos, D. C. A. Neville, A. Poggi, T. D. Butters, I. A. G. Roberts, and A. Karadimitris Human Invariant NKT Cells Display Alloreactivity Instructed by Invariant TCR-CD1d Interaction and Killer Ig Receptors J. Immunol., September 1, 2008; 181(5): 3268 - 3276. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Mariotti, V. Sargentini, C. Marcantonio, E. Todero, R. Teloni, M. C. Gagliardi, A. R. Ciccaglione, and R. Nisini T-cell-mediated and antigen-dependent differentiation of human monocyte into different dendritic cell subsets: a feedback control of Th1/Th2 responses FASEB J, September 1, 2008; 22(9): 3370 - 3379. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Moreno, J. W. Molling, S. von Mensdorff-Pouilly, R. H. M. Verheijen, E. Hooijberg, D. Kramer, A. W. Reurs, A. J. M. van den Eertwegh, B. M. E. von Blomberg, R. J. Scheper, et al. IFN-{gamma}-Producing Human Invariant NKT Cells Promote Tumor-Associated Antigen-Specific Cytotoxic T Cell Responses J. Immunol., August 15, 2008; 181(4): 2446 - 2454. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
