|
|
||||||||
Published online before print December 16, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||






,1
* Departments of Pathology and
Cell Biology and Physiology, University of Pittsburgh School of Medicine, and
University of Pittsburgh Cancer Institute, Pennsylvania
1 Correspondence: University of Pittsburgh Cancer Institute, Hillman Cancer Center, G.17d, 5117 Centre Avenue, Pittsburgh, PA 15213-1863. E-mail: vujanovicnl{at}msx.upmc.edu
| ABSTRACT |
|---|
|
|
|---|
by DC and NK cells, respectively. In contrast, specific, individual disruptions of the engagements of the corresponding TNfL-TNFfR pairs greatly impaired DC and NK cell abilities to reciprocally mediate the increases in cytokine secretion. These findings indicate that multiple TNFfLs mediate DC-NK cell interaction.
Key Words: NK cells interaction TNF family ligands
| INTRODUCTION |
|---|
|
|
|---|
DC sense pathological events in tissues via Toll-like receptors and other pattern recognition receptors, which directly recognize a wide array of molecular patterns associated with pathogens and affected cells [7 ]. The cognate process induces DC maturation and secretion of proinflammatory and immunoregulatory cytokines [i.e., interleukin (IL)-1ß, tumor necrosis factor (TNF), IL-12, IL-15, and IL-18]; enables efficient DC migration to secondary lymphoid tissues and antigen processing and presentation; and leads to initiation, polarization, and regulation of effective adaptive immune responses.
In contrast, NK cells sense pathological changes in tissues via balanced cognate activities of the NK inhibitory receptors (NKIR) and NK activating receptors (NKAR), which, respectively, recognize decreased expression of major histocompatibility complex (MHC) class I molecules and increased expression of MHC class I homologues (i.e., MICA and MICB) on affected cells [8
, 9
]. Consequently, NKIR are silenced, and NKAR are triggered; the receptor stimulation induces activation of NK cells and increases in NK cell perforin-mediated cytotoxicity and secretion of proinflammatory and immunoregulatory cytokines [i.e., IL-1ß, TNF, interferon-
(IFN-
), and granulocyte macrophage-colony stimulating factor (GM-CSF)]. The process leads to direct elimination of affected cells, induction of inflammation, and polarization and regulation of adaptive immune responses.
Recent studies have shown that DC and NK cells do not function independently but that they interact and regulate each other and mediate maturation and activation reciprocally [10
11
12
13
14
15
16
17
18
19
20
]. Fresh NK (fNK) cells induce in immature DC (iDC) increases in expression of maturation markers and secretion of IL-12. Reciprocally, iDC induce in fNK cells proliferation, IFN-
secretion, augmentation of perforin-mediated tumoricidal activity, and in vivo control of tumor growth. Furthermore, cocultures of iDC with fNK cells in the presence of DC maturation factors, mature DC with fNK cells, or iDC with activated NK (aNK) cells induce a more rapid and more pronounced DC-NK cell reciprocal stimulation.
In addition to the reciprocal stimulation, aNK cells [in humans, CD94/NKG2A+ killer immunoglobulin (Ig)-like receptor (KIR)] are able to kill iDC, which express low levels of the classical MHC class I molecules and/or human leukocyte antigen-E molecule [5 , 15 , 16 , 21 22 23 24 25 26 27 ]. This selective killing of DC by NK cells is induced in high effector:target ratios and could be mediated by triggering NKp30 NKAR and CD40L on NK cells and TNF-related apoptosis-inducing ligand (TRAIL) death receptors on DC [15 , 16 , 22 , 24 , 27 ]. The elimination of DC by NK cells is believed to be an important "control switch" of the immune system, which determines the quality of evolving immune responses by removing DC that are unsuited for efficient antigen presentation and initiation of effective adaptive immune responses. It is important that allogeneic NK cells, derived from KIR/MHC-mismatched bone marrow transplants, efficiently kill host DC and leukemia cells in vivo and prevent leukemia relapses and graft-versus-host and host-versus-graft reactions. These findings show that DC and NK cells can interact in vivo and that this interaction might occur in cell-to-cell contact. They also indicate that use of KIR/MHC-mismatched bone marrow transplants may be a promising strategy for treatment of leukemia [5 ].
The sites of in vivo DC-NK cell interactions have not been defined yet. However, DC and NK cells have been found in close proximity to each other in inflamed and tumor tissues as well as in draining lymph nodes. Therefore, they might interact in these tissues in vivo [28 ]. The interaction might induce enhancement of innate immunity, T helper cell type 1 (Th1) polarization, and amplification of antigen-specific immune responses. These immune processes might be essential for the effective elimination of invading pathogens and transformed cells.
The exact mechanisms and mediators that govern DC-NK cell interaction are largely unknown. The available published evidences clearly suggest that DC-NK cell interaction is only fully effective in direct cell-to-cell contact and is not functional or is functional at low levels across the transwell membrane [10 , 14 15 16 ]. Therefore, DC and NK cells might preferentially interact via cell membrane-bound molecules, although secreted molecules could also participate in this process but at a low level [14 , 15 ].
DC and NK cells express a variety of biologically active cell membrane-bound and secreted ligands and corresponding receptors, which cognate engagement could reciprocally mediate important biological responses. These molecules include the chemokines CC chemokine ligand 3 and CXC chemokine ligand 8, produced by DC, and IL-8 and macrophage-inflammatory protein 1, produced by NK cells [29
30
31
]; the cell adhesion molecules lymphocyte function-associated antigen 1 and intercellular adhesion molecule 1, expressed on both cell types [1
, 6
, 17
, 29
30
31
]; the costimulatory molecules CD80 and CD86, expressed on DC, and their receptor CD28, expressed on NK cells [21
, 32
]; the NKAR NKG2D, expressed on NK cells, and its inducible ligands, expressed on DC [18
, 33
]; and the cytokines IL-12, IL-15, and IL-18, secreted by DC, and IFN-
and GM-CSF, secreted by NK cells [1
, 6
]. We have shown that DC and NK cells also express the TNF family ligands (TNFfLs) TNF, lymphotoxin (LT)-
1ß2, FasL, and/or CD40L as well as the corresponding TNF family receptors (TNFfRs) TNF receptor 2 (TNFR2), LT-ßR, Fas, and CD40 [34
, 35
].
TNFfLs are pleiotropic, cell membrane-bound and secreted molecules, which mediate such diverse and important functions as apoptotosis, inflammation, and immunoregulation [36
37
38
39
]. Some of TNFfLs expressed by DC and NK cells, in particular, TNF, CD40L, and LT-
1ß2, are potent mediators of DC maturation and NK cell activation [36
37
38
39
]. These important biological responses are also induced in DC-NK cell interaction. In addition, it has been recently shown in humans that anti-TNF neutralizing antibodies can decrease the NK cell-mediated stimulation of CD86 expression by DC, indicating that TNF might be a mediator of DC maturation, which occurs in the DC-NK cell interaction [14
, 15
]. These findings indicate that TNF and some other members of TNFfLs might be important mediators of DC-NK cell interaction and reciprocal stimulation. In the present study, we provide the first experimental evidence supporting the hypothesis.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Reagents and antibodies
Low toxic rabbit complement was obtained from Accurate Chemical and Scientific Corp. (Westbury, NY). Mouse and human recombinant (hr) GM-CSF and IL-4 were purchased from the R&D Systems (Minneapolis, MN). hrIL-2 was provided by the Chiron-Cetus Corp. (Emeryville, CA). Escherichia coli lipopolysaccharides (LPS) was obtained from Sigma Chemical Co. (St. Louis, MO). Soluble recombinant mouse (srm)TNF was purchased from Alexis (San Diego, CA). Hybridomas producing rat anti-mouse CD4 (clone GK1.5, TIB207), CD8 (clone 2.43, TIB217), and CD45/B220 (clone RA3-3A1/6.1, TIB146) monoclonal antibodies (mAb) were obtained from American Type Culture Collection (Manassas, VA). Magnetic cell-sorter (MACS) NK cell isolation kit, which contains the biotin-antibody cocktail (anti-CD4, -CD5, -CD8
, -CD19, -Ly-6G, -Ter-119 mAb), antibiotin microbeads, and magnetic columns, was purchased from Miltenyi Biotec (Auburn, CA). Fluorochrome-conjugated anti-mouse NK1.1, CD3, CD14, CD19, and CD11c mAb and corresponding isotype-nonreactive control mAb were obtained from BD PharMingen (San Diego, CA). Rat anti-mouse TNF mAb (Pierce-Endogen, Rockford, IL), hamster anti-mouse CD40L mAb (BD PharMingen), hamster anti-mouse FasL mAb (MBL, Watertown, MA), and isotype- and species-matched, nonreactive control mAb (BD PharMingen) were used in blocking experiments. 5-Chloromethylfluorescein diacetate (CMFDA; green fluorescent dye) and 4-chloromethyl benzoyl amino tetramethyl rhodamine (CMTMR; orange fluorescent dye) were purchased from Molecular Probes (Eugene, OR). Mouse IL-12p70 and IFN-
enzyme-linked immunosorbent assay (ELISA) kits were obtained from R&D Systems.
Generation of DC
Human iDC were generated by 7-day stimulation of purified peripheral blood monocytes with GM-CSF and IL-4, as described previously [35
].
Mouse iDC were produced from lineage marker-negative bone marrow cells, which were obtained by flashing bone marrow cavities of femurs and tibias. Bone marrow cells were depleted of erythrocytes by hypotonic lysis and of lineage marker-positive cells by treatment with anti-CD4, -CD8, and -B220 mAb (UPCI, Hybridoma Facility) and rabbit complement (Accurate Chemical and Scientific Corp.). After depletion of the mature hematopoietic cells, bone marrow cells were resuspended (0.5x106/ml) in complete cell culture medium (CM), which was constituted of RPMI-1640 medium supplemented with 0.1 mM nonessential amino acids, 2 mM sodium pyruvate, 1 mM L-glutamine, 100 µg/ml streptomycin, 100 U/ml penicillin, 20 mM HEPES buffer, 10% heat-inactivated fetal bovine serum (all from Life Technologies, Grand Island, NY), and 50 µM 2-mercaptoethanol (Bio-Rad, Hercules, CA) and cultured for 6 days in the presence of rmGM-CSF and IL-4 (R&D Systems, 15 and 7.5 ng/ml, respectively). Thus, generated iDC were 95% CD11c+lineage markers.
Purification and activation of NK cells
Purified human peripheral blood fNK cells were obtained as described previously [34
].
Mouse aNK cells were generated from purified, splenic fNK cells. Splenocytes were depleted of erythrocytes by hypotonic lysis, and their NK1.1+ cells were purified by negative immunoselection using MACS NK cell isolation kit (Miltenyi Biotec), as recommended by the manufacturer. Thus, purified fNK cells were consistently
75% NK1.1+CD3. Purified fNK cells were resuspended in CM (0.1x106/ml), supplemented with 6000 IU/ml IL-2, and cultured for 6 days. In this culture, aNK cells were generated, expanded ten- to 20-fold, and consistently
97% NK1.1+CD3.
DC-NK cell-binding assay
DC and NK cells were labeled with 1.25 µM CMFDA and CMTMR fluorescent dyes, respectively. The labeled cells were mixed in a 1:1 ratio, pelleted by centrifugation, and incubated for 10 min at 37°C. Next, the cell pellet was gently resuspended. The resulted cell suspension was added onto chamber slides and additionally incubated for 20 min at 37°C. The cells were then fixed with 1% paraformaldehyde and analyzed using a confocal microscope.
DC-NK cell coculture and cytokine testing
Day 6 of their induction with cytokines, iDC and aNK cells were harvested, seeded individually, or mixed in a 1:1 ratio (0.5x106/ml each) in 24-well plates and additionally cultured for 48 h in the absence or presence of LPS (0.12 µg/ml) and/or IL-2 (6000 IU/ml). In reconstitution experiments, TNF/ DC and NK cells were cocultured in the absence or presence of rTNF (5 ng/ml). In blocking experiments, normal mouse DC and NK cells were cocultured in the presence of isotype control, anti-TNF, anti-CD40L, or anti-FasL mAb (20 µg/ml). Following the cultures, cell culture-conditioned media were collected and assessed for the presence and quantity of IL-12p70 and IFN-
using R&D Systems ELISA kits. Data were presented as means of triplicates of IL-12p70 and IFN-
pg/0.5 x 106 cells/ml. SD were <5% of means.
| RESULTS |
|---|
|
|
|---|
B p65. TNFfLs could be mediators of the NF-
B activation in interacting DC and NK cells, as triggering of TNFR2, LT-ßR, and CD40 has been shown to activate NF-
B [38
, 39
]. Next, 5 h coincubations of iDC and fNK cells in cell-to-cell contact but not separated by the transwell membrane reciprocally induced increases in expression of TNF, TNFR2, CD40L, CD40, FasL, Fas, and TRAIL mRNAs. In addition, 24 h cocultures of iDC and fNK cells or aNK cells in cell-to-cell contact induced augmentation of NK cell perforin-mediated and DC/NK cell TNFfL-mediated, tumoricidal activities (data not shown). In contrast, 4872 h coculture of iDC with aNK cells but not with fNK cells in direct cell-to-cell contact or separated by the transwell membrane reciprocally induced secretion of IL-12p70 and IFN-
, respectively. However, this stimulation was two- to threefold higher in the cell-to-cell contact than across the transwell membrane (data not shown). Furthermore, fNK cells and aNK cells were able to help iDC to prime naïve T cells and induce cytotoxic T lymphocytes to melanoma-associated peptides p53, gp100, and melanoma antigen recognized by T cells, and this immune function was also effectively mediated in direct cell-to-cell contact and was notably greater with aNK cells than fNK cells (data not shown). The data show that fNK cells interact with DC only in direct cell-to-cell contact, and aNK cells can do so mostly in cell-to-cell contact but also at low levels across the transwell membrane. Therefore, our studies show that DC communicate with fNK cells or aNK cells exclusively or mainly via cell membrane-bound molecules. This further indicates that the mediators of DC-NK cell interaction could be cell membrane-bound TNFfLs.
|
by NK cells, the two major Th1 cytokines, which are reciprocally induced in DC-NK cell interaction (data not shown). Using the gene knockout approach, we examined whether disruption of the interaction between TNF and TNFR2 could subsequently reduce DC-NK cell reciprocal stimulation. We compared the biological responses in cocultures of normal mouse and TNF/ mouse iDC and aNK cells. We found that TNF/ iDC and aNK cells were three- to tenfold less potent than normal iDC and aNK cells to stimulate IL-12p70 secretion by DC reciprocally (Table 1
) and killing of YAC-1 tumor cell targets (data not shown) and secretion of IFN-
by NK cells (Table 2
). Addition of srTNF to the coculture of iDC and aNK cells (Table 2)
but not during in vitro generation of iDC (data not shown) fully restored the ability of TNF/ DC to stimulate TNF/ NK cells to produce IFN-
. Therefore, the impairment of TNF/ DC and NK cells to collaborate was caused directly by the absence of TNF but not indirectly by potential maturation defects of DC, which could occur in the absence of TNF. Finally, using the antagonistic approach by applying specific blocking receptor-Fc constructs or antibodies, we examined the effects of individual or combined disruption of TNF-TNFR2, LT-
1ß2-LT-ßR, CD40L-CD40, and FasL-Fas ligand receptor pairs on DC-NK cell reciprocal activation. We found that the disruption of engagements of the each individual TNFfL-TNFfR pair notably inhibited DC-NK cell interaction and consequent reciprocal induction of expression of TRAIL and CD40L mRNAs (data not shown) as well as secretion of IL-12p70 and IFN-
(Table 3
, data not shown for LT-
1ß2-LT-ßR).
|
|
|
| DISCUSSION |
|---|
|
|
|---|
1ß2-LT-ßR, and FasL-Fas ligand receptor pairs. It also indicates that the multiple TNFfL-TNFfR pairs mediate DC-NK cell interaction by simultaneous engagements in which all the involved ligand receptor pairs have a similar, important role. A possible explanation for these findings is as follows: TNFR2, LT-ßR, and CD40 possess specific cytoplasmic domains of homology [PXQX(T/S)], and upon their ligation, the receptors recruit and interact with adaptor molecules, TNFR-associated factors (TRAFs). TNFR2 interacts with TRAF1 and TRAF2, LT-ßR interacts with TRAF3 and TRAF5, and CD40 interacts with TRAF2, TRAF3, TRAF5, and TRAF6 [41
] (Fig. 2
). In addition, triggering of TNFR1 or Fas recruits the apoptosis signaling molecules and TRAF1 and TRAF2. Furthermore, reverse signaling via TNFfLs (e.g., FasL and CD40L) induces inflammatory responses and might also stimulate TRAF signaling pathway [22
, 41
]. It is important that four TRAFs are shared by the TNFfRs, and it appears that their recruitments initiate the common TRAF signaling pathway, which leads to the induction of NIK, IKK, and I
B phosphorylation and consequently, activation and translocation into the nucleus of NF-
B, expression of cytokine genes, production of cytokines, and inflammation (Fig. 2)
. Therefore, the simultaneous engagements of multiple TRAF-recruiting TNFfL-TNFfR pairs expressed on DC and NK cells may induce reciprocally simultaneous multifocal recruitments of multiple TRAFs and simultaneously mediate multiple, similar signaling events, which converge into a streamline of a strong signal (Fig. 2)
. A reason for the necessity of the simultaneous engagements of each of the multiple heterologous TNFfL-TNFfR pairs to induce significant biological responses of interacting DC and NK cells might be a low cell-surface expression of TNFfLs and TNFfRs on these cells [34
, 35
, 40
], which individual expression levels and large molecular distances unable their homologous but enable their heterologous polymerization and cross-linking. The simultaneous, heterologous engagements and cross-linking of multiple TNFfMs induce an appropriate and strong signal, which mediates significant biological responses. The hypothesis is under our current investigation.
|
| ACKNOWLEDGEMENTS |
|---|
Received November 19, 2004; accepted November 24, 2004.
| REFERENCES |
|---|
|
|
|---|
stimulated dendritic cells in NK cell activation: impairment in chronic hepatitis C virus infection J. Immunol. 170,1249-1256
1ß2, Fas ligand and TRAIL J. Immunol. 168,1831-1839This article has been cited by other articles:
![]() |
R. Winkler-Pickett, H. A. Young, J. M. Cherry, J. Diehl, J. Wine, T. Back, W. E. Bere, A. T. Mason, and J. R. Ortaldo In Vivo Regulation of Experimental Autoimmune Encephalomyelitis by NK Cells: Alteration of Primary Adaptive Responses J. Immunol., April 1, 2008; 180(7): 4495 - 4506. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Tian, S. Grimaldo, M. Fugita, J. Cutts, N. L. Vujanovic, and L.-Y. Li The Endothelial Cell-Produced Antiangiogenic Cytokine Vascular Endothelial Growth Inhibitor Induces Dendritic Cell Maturation J. Immunol., September 15, 2007; 179(6): 3742 - 3751. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Xu, A. K. Chakrabarti, J. L. Tan, L. Ge, A. Gambotto, and N. L. Vujanovic Essential role of the TNF-TNFR2 cognate interaction in mouse dendritic cell-natural killer cell crosstalk Blood, April 15, 2007; 109(8): 3333 - 3341. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Huang, M. Kucia, F. Rezzoug, J. Ratajczak, M. K. Tanner, M. Z. Ratajczak, C. L. Schanie, H. Xu, I. Fugier-Vivier, and S. T. Ildstad Flt3-Ligand-Mobilized Peripheral Blood, but Not Flt3-Ligand-Expanded Bone Marrow, Facilitating Cells Promote Establishment of Chimerism and Tolerance Stem Cells, April 1, 2006; 24(4): 936 - 948. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |