SEARCH

SEARCH BY CITATION

Keywords:

  • DC;
  • IFN-γ;
  • NK cells;
  • Tumor immunity;
  • Vaccination

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

We vaccinated mice with DC loaded with or without invariant NKT-cell ligand α-galactosylceramide and evaluated long-term resistance against tumor challenge. When mice had been given either DC or DC/galactosylceramide and were challenged with tumor cells even 6–12 months later, both NK and NKT cells were quickly activated to express CD69 and produce IFN-γ. The NK cells could resist a challenge with several different tumors in vivo. The activated NK and NKT cells could be depleted with anti-NK1.1 treatment. In spite of this, the activated cells recovered, indicating that tumor-responsive NK and NKT cells were being generated continuously as a result of vaccination with DC and were not true memory cells. The NK and NKT antitumor response in DC-vaccinated mice depended on CD4+ T cells, but neither CD8+T cells nor CD4+CD25+ regulatory T cells. However, both vaccine DC and host DC were required for the development of long-term, tumor reactive innate immunity. These results indicate that DC therapy in mice induces long-lasting innate NK- and NKT-cell activation through a pathway that requires host DC and CD4+ T cells and that the continued generation of active NK cells resists the establishment of metastases in vivo.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

DC are specialized APC for initiating T-cell responses. Antigen-loaded DC induce antigen-specific T-cell immunity in an MHC-dependent manner in murine models and humans. In addition, DC can stimulate invariant Vα14+ NKT (iNKT) cells after being loaded with α-galactosylceramide (α-GalCer). Injection of α-GalCer-loaded DC (DC/Gal) leads to a more prolonged IFN-γ-producing iNKT-cell response in comparison with free α-GalCer administration 1–3. The IFN-γ-secreting iNKT cells that develop following injection of DC/Gal are evident 2 days after immunization but dissipate 2 wk later.

NK cells in the innate immune system represent a critical first line of defense against malignant transformation and infection. One way to activate NK cells involves NKT cells, which elicit cytokines from DC 2–5. Several groups have addressed these DC-derived NK-activating cytokines, such as IL-2, IL-12, IL-15 and type I IFN. For example, Granucci et al. 6 demonstrated that TLR-ligand-activated DC, e.g. with LPS, CpG or BCG produce IL-2 in a few hours, which then activates NK cells to produce IFN-γ. In addition, IL-12 is produced by iNKT-activated DC and is a primary cytokine for the production of IFN-γ and cytotoxicity by NK cells 7, 8. Other cytokines such as type I IFN, IL-15 and IL-18 released during the DC–NK-cell interaction can have synergistic effects on NK function 9, 10. In particular, a membrane-bound form of DC-derived IL-15 appears to be necessary to induce proliferation in NK cells. A role for endogenous type I IFN in restricting the growth of tumors, through NK-cell activation, has also been demonstrated 11.

Two groups documented the ability of DC even in the absence of α-GalCer loading, to activate NK cells to exert long-term resistance toward tumor cells. van den Broeke et al. 12 demonstrated that DC in the absence of exogenous glycolipids or antigen (here termed “unpulsed DC”) protect mice against tumor lung metastases in an NK-cell-dependent manner. Prins et al. 13 also reported that DC administration induced NK-cell-dependent protection against B16 melanoma in the brain. Mechanistically CD4+T cells were found to be necessary for NK-cell-mediated tumor inhibition 12, 13.

T cells also can contribute to antitumor resistance following administration of DC. Adam et al. 14 reported that the activation of NK cells by unpulsed DC leads to antitumor memory that is CD8+ CTL-mediated but CD4+ T-cell-independent. They found that unpulsed DC showed antitumor effects only against A20 (B-cell lymphoma), but not against MCP11 (B-cell lymphoma) and CT26 (colon carcinoma cell line). Dworacki et al. 15 demonstrated that unpulsed DC can induce noticeable CD4+ T-cell and CD8+ T-cell resistance to a variety of syngeneic transplantable mouse tumors, although a role of CD8+ T cells was more prominent. These publications indicate some T-cell involvement during NK-cell activation by DC.

In this paper, we compared α-GalCer-pulsed and non-pulsed DC to induce long-term NK- and NKT-cell activation at the single-cell level. To do so, we established single-cell assays to evaluate enhanced NK- and NKT-cell activity toward a challenge with tumor cells in DC-vaccinated mice. We will also show that DC therapy induces long-term NK-cell resistance to tumors through a combination of host DC and CD4+ T cells that sustains the continued activation rather than true-memory NK cells.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

DC induce long-term NK-cell-mediated resistance to B16 melanoma

We had previously studied the early NKT- and NK-cell responses to administration of DC/Gal 1. To assess the longevity of this response, mice were given DC or DC/Gal i.v. and then challenged with live B16 melanoma i.v. 2 wk to 12 months later. Two weeks after this challenge, we counted the number of tumor metastases in the lung as a measure of tumor resistance. Mice immunized with DC and DC/Gal were protected against a challenge of live tumor cells (Fig. 1A and B), and this protection was long lived. Fewer metastases developed in mice given DC/Gal versus DC only (Fig. 1A and B), but surprisingly, the protection induced by both lasted 12 months even though the mice had not been immunized with B16 melanoma.

thumbnail image

Figure 1. NK-cell-mediated antitumor activity many months after administration of DC or DC/Gal. (A and B) Mice were immunized with DC or DC/Gal and were challenged with B16 melanoma 2 wk and 1, 6 or 12 months later. Antitumor effects were evaluated 14 days after B16 challenge by counting the number of metastases in the lung. (C) Same as (A) and (B), but mice were treated with control rabbit serum or anti-asialo-GM1 serum just prior to B16 challenge. The depletion of NK1.1+ CD3 NK cells was verified by FACS analyses (left). (D) IFN-γ−/− or IL-4−/− mice were used as recipients of DC/Gal. Data show means obtained from three independent experiments. *p<0.001 for DC versus DC/Gal, p<0.001 for animals treated at 6 months, p<0.01 for animals treated at 12 months, p>0.05 for animals treated at 2 wk or 1 month.

Download figure to PowerPoint

To determine if NK cells were responsible for the observed protection against B16 melanoma, we treated mice with anti-asialo-GM1 Ab to deplete NK cells just prior to challenge with B16 tumor, but 1 month after DC immunization. FACS analysis confirmed the depletion of NK1.1+ CD3 NK cells (Fig. 1C, left). NK depletion, and also the use of IFN-γ−/− mice ablated the protection that had been induced by vaccination with either DC or DC/Gal (Fig. 1C and D). A similar observation has been reported by van den Broeke et al. 12 following vaccination with DC, i.e. the development of long-lived, NK-dependent resistance to tumors such as B16 melanoma and CT26 colon carcinoma.

Single-cell assays to monitor the development of long-lived active NK and NKT cells

To characterize the long-lived resistance to tumors at the single-cell level, we vaccinated mice with DC for 1 month, challenged with B16 melanoma, and at various time points, we examined the response of both NK (CD3 NK1.1+) and NKT (CD3+ NK1.1+) compartments using FACS assays. We then measured the up-regulation of the CD69 activation marker and the production of IFN-γ. We found that 16–24 h after B16 tumor challenge, these markers of NK and NKT activation were optimally detected (Fig. 2A). DC, but not B16 melanoma itself, could induce this heightened responsiveness of NK cells to B16 challenge (Fig. 2B). In addition, CD1d-GalCer dimer binding could be used to monitor the heightened reactivity of iNKT cells following B16 challenge (Fig. 2C). Then, to demonstrate that the innate immunity is not restricted in B16 melanoma, some syngeneic tumor cells (B16, EL4 and YAC-1) and allogeneic tumor cells, J558, were administered as targets in immunized mice with DC/Gal. NK and NKT cells in DC/Gal-immunized mice responded to these tumor cells and produced IFN-γ 16 h later (Fig. 2D). To verify the response in another strain of mice, BALB/c, we showed that DC or DC/Gal vaccination induced NK cells to produce IFN-γ 16 h after Colo26 challenge (Fig. 2E). In the lung tumor resistance model as seen in B16 protocol, the metastases in lung were not seen in DC or DC/Gal-immunized mice (Fig. 2F). Therefore, single-cell assays can be used to document the heightened reactivity of NK cells to several tumors in the spleens of DC-vaccinated mice.

thumbnail image

Figure 2. Single-cell assays for NK and NKT cells with increased reactivity to B16 melanoma. (A) DC/Gal-injected mice were challenged with B16 melanoma and at the indicated time points (y-axis, h), the activation of NK cells was analyzed by up-regulation of CD69 expression on NK cells (CD3 NK1.1+) or NKT cells (CD3+NK1.1+ ) (left panel). IFN-γ content also was measured by secretion assay (right panel). (B) IFN-γ production assay for B16-reactive NK cells was performed 16 h later in naïve mice and mice vaccinated with DC or DC/Gal. (C) To verify the activation of iNKT in DC or DC/Gal vaccinated, we analyzed IFN-γ production from CD1d/Gal-dimer+ cells by intracellular cytokine staining as described in the Materials and methods section. CD1d/Gal-dimer+ cells were stained using CD19-FITC (to gate out B cells), CD1d/Gal-dimer-PE and IFN-γ APC (i.e. CD19 CD1d/Gal-dimer+) after challenge with B16 melanoma cells in vivo. (D) Various tumor targets including syngeneic tumor cells (B16, EL4, YAC-1) and allogeneic tumor cells, J558 cells, were tested in mice immunized with DC/Gal. DC/Gal-immunized or non-immunized mice were challenged with each tumor 1 month later. Then, IFN-γ production by NK and NKT was analyzed 16 h later. (E) As shown in Fig. 2B, DC- or DC/Gal-injected BALB/c mice were challenged with Colo26 colon cancer cell line. IFN-γ production assay for Colo26-reactive NK cells was performed 16 h later. (F) As shown in Fig. 1, BALB/c mice that had been immunized with DC or DC/Gal were challenged with Colo26 1 month later. Antitumor effects were evaluated 21 days later by counting the number of metastases in the lung. Data are representative of two separate experiments (n≥4 per group).

Download figure to PowerPoint

Tumor-reactive NK cells are found for many months and systemically after administration of DC or DC/Gal

To characterize the innate reactivity in mice that had been injected with DC or DC/Gal, we evaluated the duration and tissue distribution of activated NK and NKT cells by measuring IFN-γ production following challenge with B16. Naïve mice did not respond to B16, but both types of innate cells were primed with DC or DC/Gal for long periods, 1–12 months (Fig. 3A). The priming of NKT and NK cells with DC/Gal was longer lasting than with DC (Fig. 3A). We also showed that activated NK cells were primed by DC to express IFN-γ in bone marrow, liver and lung as well as spleen (Fig. 3B). Apparent activation of NKT and NK cells in lymph nodes after challenge with B16 was not seen in Fig. 3B. To further analyze these responses in lymph nodes, immunized mice were challenged with B16 s.c. The activation of NK cells was weak, and NKT-cell activation was weak as well (Fig. 3C). The results in Figs. 2 and 3 indicate that mice receiving DC or DC/Gal several months earlier have heightened NKT and NK reactivity long term and systemically to challenge with various tumors.

thumbnail image

Figure 3. Tumor-reactive NK and NKT cells are found for many months and systemically after administration of DC or DC/Gal. (A) Mice were immunized with DC or DC/Gal and 1–12 months (M) later the animals were challenged or not challenged with B16 melanoma i.v. (+/− on y-axis) 16 h later, the NK and NKT cells were analyzed for IFN-γ content. (B) The activity of NK cells was determined in various organs, 1 month after vaccination with DC or DC/Gal. (C) To further analyze NKT cells in draining lymph nodes, immunized mice were challenged with B16 s.c. The activity of NK and NKT cells was determined by IFN-γ expression. Data are representative of two separate experiments (n≥4 per group).

Download figure to PowerPoint

DC are required to induce long-term, tumor-reactive NK cells

We first tested tumor cells and B blasts to induce tumor-reactive NK cells, but in contrast to DC, these other cell types could not elicit NK cells that would produce IFN-γ after challenge with B16 melanoma (Fig. 4A). We next generated DC under different conditions and injected them into mice, and 1 month later looked for NK responses to B16 melanoma (Fig. 4B, top row). The DC were active when they were generated in the presence of mouse rather than FBS, and when they had been exposed to recombinant GM-CSF (rather than the conditioned medium from GM-CSF-transfected J558 cells that is typically used to make DC) or recombinant TNF-α. Also, DC from IL-2-, IL-15- or IL-12-deficient mice were able to induce tumor-reactive NK cells (Fig. 4B, lower row). Not only BMDC but also mature splenic DC after 16 h culture improved the NK-cell response to B16 challenge (data not shown). In all these experiments, DC/Gal were more active than DC in inducing reactive NK cells (Fig. 4B). Therefore, DC are specialized APC for activating NK cells in mice, but surprisingly the DC do not need to produce either IL-2, -12 or -15 for this function.

thumbnail image

Figure 4. DC are required to induce long-term, tumor-reactive NK cells. (A) Several different tumor cells and B blasts, which had been generated from stimulating resting B cells for 2 days in the presence of LPS (25 μg/mL), were administered i.v. to mice. The mice were challenged with B16 melanoma i.v. and then analyzed for IFN-γ production as described in Fig. 2. (B) DC were generated under various conditions and injected into mice, and 1 month later NK responses to challenge with B16 melanoma was determined by IFN-γ production. Also, DC from IL-2-, -15- or -12-deficient mice were tested to activate NK cells against tumor cells. Data are representative of two separate experiments (n≥4 per group).

Download figure to PowerPoint

Recipient DC and T cells, but not NKT cells, maintain NK reactivity to DC vaccine

To assess the dependence on Vα14+ or CD1d-restricted NKT cells, DC or DC/Gal were injected into C57BL/6, CD1d-deficient or Jα18 knock-out mice lacking NKT cells. These mice were challenged with B16 melanoma 1 month later. The heightened IFN-γ secretion by NK cells in response to B16 challenge was apparent in CD1d−/− or Jα18−/− mice (Fig. 5A, left), indicating that NKT cells including iNKT cells in the recipient were not essential. Similarly, μMT−/− mice developed reactive NK cells, showing that B cells were dispensable, but Rag2−/− mice did not develop reactive NK cells to DC or DC/Gal immunization, indicating that T cells were indispensable (Fig. 5A, right).

thumbnail image

Figure 5. Recipient DC, T cells but not NKT cells are required to maintain long-term NK reactivity (A) DC or DC/Gal were injected into C57BL/6, CD1d-deficient (lacking iNKT cells and type II NKT cells), μMT knockout mice (lacking B cells) or Jα18 knockout mice (lacking iNKT cells). These mice were challenged with B16 1 month later and heightened IFN-γ secretion by NK cells in response to B16 challenge was analyzed. (B) To investigate whether DC are required in the recipient, we injected WT DC (upper) or DC/Gal (lower) into CD11c-DTR mice. One month later, the mice were injected with DT at the time points indicated to eliminate CD11chigh DC 16. The mice were evaluated for reactive NK cells following B16 injection. Data are representative of two separate experiments (n≥4 per group).

Download figure to PowerPoint

To look for the requirements for DC in the recipient, we injected WT DC or DC/Gal into CD11c-DTR mice. We then injected diphtheria toxin (DT) at the time points indicated in the top of Fig. 5B to eliminate CD11chigh DC 16, as we verified (data not shown). When we immunized DT-treated CD11c-DTR mice with DC or DC/Gal, the heightened NKT-cell and NK-cell responses disappeared (Fig. 5B lower row). Together these data indicate that the function of both DC and T cells in the recipient mice are required for heightened NKT and NK reactivity to develop.

CD4+ cells are needed for long-term NK reactivity following DC therapy

To assess the contribution of CD4+versus CD8+ cells in the recipient mice, we depleted these cells with antibodies from DC- or DC/Gal-immunized mice just prior to challenge with B16 melanoma (see Fig. 6A, experimental plan top). Removing CD4+ cells but not CD8+ cells just before B16 challenge abrogated the heightened IFN-γ secretion from NKT cells and NK cells upon B16 challenge (Fig. 6B).

thumbnail image

Figure 6. CD4+cells are needed for long-term NK reactivity following DC therapy. (A) Mice given DC or DC/Gal 1month before were challenged with B16. We analyzed IFN-γ production from CD3+ NK1.1+(NKT cells) (left) and CD3 NK1.1+(NK cells) (right) by intracellular staining after culturing 2 h in the presence of Golgi Plug 16 h after B16 challenge. (B) To assess the contribution of CD4+versus CD8+ cells in the recipient mice, we depleted these cells with Ab from DC- or DC/Gal-immunized mice just prior to challenge with B16 melanoma (see Fig. 6A, experimental plan top). Then, we analyzed IFN-γ production by intracellular staining as in (A). (C) To evaluate a role for activated CD25+ cells, including Foxp3+ Treg, in this system, CD25+ cells were depleted with PC61 mAb just prior to challenge with B16. Data are representative of two separate experiments (n≥4 per group).

Download figure to PowerPoint

We also considered a role for activated CD25+ cells, including Foxp3+ Treg, in this system by depleting CD25+ cells with PC61 mAb just prior to challenge with B16, but this depletion did not affect the heightened NKT and NK reactivity, i.e. high IFN-γ secretion, to the tumor (Fig. 6C). Taken together these data indicate that strong innate reactivity in DC- or DC/Gal-immunized mice requires CD4+, but not CD25+ or CD8+, cells.

Long-term NK activation in DC-vaccinated mice does not reflect true memory but rather continuous reactivation

To establish if the long-term reactivity of NK and NKT cells reflected a memory state in these innate lymphocytes, we treated the vaccinated mice with anti-asialo-GM1 (data not shown) or anti-NK1.1 Ab (Fig. 7) after injection with DC or DC/Gal. As shown in Fig. 1C, we again verified by FACS the depletion of appropriate cells by the treatment with anti-NK1.1 Ab, i.e. NKT (CD3+ NK1.1+) cells and NK (CD3 NK1.1+). Three months later, the mice were challenged with B16 melanoma (see experimental design, Fig. 7). Early treatment with anti-asialo-GM1 or anti-NK1.1 did not ablate the heightened NK reactivity to DC or DC/Gal 3 months (Fig. 7) or 6 months (data not shown) later. This indicates that reactive NK cells are continually being generated for many months in mice that are immunized with DC or DC/Gal.

thumbnail image

Figure 7. Long-term NK activation in DC-vaccinated mice does not reflect true memory but rather continuous reactivation. (A) To understand whether NK and NKT cells were needed for the establishment of NK reactivity, DC-vaccinated mice were treated with anti-asialo-GM1 Ab (data not shown) or anti-NK1.1 Ab. The depletion of NK1.1+ cells, including NKT (CD3+ NK1.1+) cells and NK (CD3 NK1.1+) cells were verified by FACS analysis in treated mice with anti-NK1.1 Ab 2 wk after injection with DC or DC/G, but 3 and 6 months (M) prior to challenge with B16 melanoma (see experimental design, top panel). (B) Then, IFN-γ secretion by NK cell and NKT cells in response to B16 challenge was analyzed. Data are representative of two separate experiments (n≥4 per group).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

This paper reports on a surprising response to immunization with DC or DC/Gal in mice. For prolonged periods (up to 12 months), the immunized mice have NKT and NK cells that are in a “primed” state, being able to produce IFN-γ rapidly upon challenge with B16 melanoma and to resist the metastases of B16 to the lung. After we had performed our experiments, we became aware of two publications that reported long-term NK-based resistance induced by DC vaccines 12, 13. Our paper extends this prior research in three ways: (i) by comparing DC with DC/Gal, (ii) by analyzing the increased NK- and NKT-cell reactivity at the single-cell level and (iii) by showing a role for recipient DC and recurrent NK activation rather than true memory.

A good deal of prior research documents the capacity of DC to activate NKT cells and NK cells particularly in mice. Kinjo et al. 17–19 or others showed that microbial-antigen mediated iNKT-cell stimulation. Nagarajan and Kronenberg 20 showed that LPS-matured DC independent of iNKT-cell ligands could produce IL-12 and IL-18 that activated iNKT cells. Brigl 21 used Salmonella-derived LPS to stimulate DC, and these directly stimulate NKT cells through both enhancement of endogenous ligand (glycosphingolipids) and IL-12 secretion. Paget et al. 22 recently demonstrated that TLR signals in unloaded myeloid DC can directly stimulate iNKT cells. They showed that CpG-oligonucleotides (CpG-ODN)-treated DC induced IFN-γ-producing iNKT cells and antitumor effects depending on type I IFN and charged glycosphingolipids. Similarly, several reports document the capacity of DC to activate NK cells in mice 23 or in cell cultures from human blood 24–26. However, these prior reports are different from the current study where we vaccinate mice with DC and then show a long-term (at least 1 year) increase in NK- and NKT-cell reactivity, which becomes evident only if one challenges the mice with B16 melanoma or other tumors (Fig. 2C, data not shown).

At first, the findings in Fig. 1 suggested that the injection of DC or DC/Gal was inducing “memory” NK cells. Depletion from mice of NK cells treated with the anti-asialo-GM1 Ab, which spares iNKT cells, completely abrogated the antitumor response elicited by injecting into mice either unloaded or α-GalCer-loaded DC (Fig. 1C), suggesting that NK cells were indispensable and sufficient for the antitumor response. However, we could deplete NK cells 2 wk after DC injection, and the mice developed new reactive NK cells 3 and 6 months later (Fig. 7A). This means that the DC-injected mice continued to generate reactive NK cells rather than developing true-memory NK cells. We also were able to find long-term NK-cell activation in multiple organs by establishing a single-cell method to detect the activation of NK cells without any culture. Further, we could show that the NK cells long after DC vaccination exerted rapid antitumor immune response (<10 h) when challenged with a variety of target tumor cells.

The mechanism for this generation of reactive NK and NKT cells requires DC (Fig. 5C) and CD4+ T cells (Fig. 6B) in the recipient. We hypothesize that the initial immunization with DC activates CD4+ T cells, which then act back on recipient DC to generate a stimulus triggering for continued activation of NKT and NK cells. As Granucci et al. 6 reported, DC-derived IL-2 can be a pivotal factor for activating NK cells. Surprisingly the molecular mechanism for the NK response to DC that we studied, which remains to be identified, does not require expression of IL-2, 12, or 15 by the injected DC, e.g. Fig. 4B. However, we cannot rule out the possibility that such cytokines can be released from effector cells or DC in the host, rather than the transferred DC. In particular, because of our current data on a role of CD4+ T cells, inflammatory cytokines from CD4+ T cells as well as DC in the host need to be further analyzed in promoting survival, proliferation, activation and enhanced cytotoxic activity of NK cells.

In contrast, memory T cells and memory B cells usually do not need any APC to maintain memory function. When encountering tumor cells, NK cells together with some danger signals may initiate the chain reaction of this DC- and CD4+T-cell-mediated immune response. Two other groups already ruled out the involvement of FBS response by measuring tumor growth in mice injected with DC exposed to mouse serum 12, 15, and we likewise confirmed IFN-γ secretion by NK cells in mice immunized with DC cultured in mouse serum.

In immunosurveillance, some types of CD4+ cells including CD4+CD25+ Treg cells 27, 28 and type II NKT cells have suppressive effects on tumor immunity 29. Ambrosino et al. 30 were able to suppress immune surveillance with selective stimulation of CD4+type II NKT cells, which are CD1d-dependent, but not iNKT cells. They found that Jα18−/− mice were more susceptible to tumor growth at early stage in contrast to WT mice due to the CD4+ type II NKT cells 30. They also demonstrated that tumor growth was inhibited by an injection of anti-CD4 Ab even in Jα18−/− mice. CD4+ type I invariant Vα14+ NKT cells also have less function due to the ICOS-ICOS ligand interaction under some conditions 31. Different from these studies, CD4+ T cells play a pivotal role in our study. We proved that not only NK cells but also NKT cells promptly responded to a challenge with tumor cells and this may explain the enhanced antitumor memory in mice immunized with DC/Gal relative to DC only. However, from the data of Fig. 5A, iNKT cells are apparently not essential players for the induction of long-term NK reactivity by DC. As discussed above, Paget et al. 22 demonstrated that CpG-ODN-treated, unloaded myeloid DC can directly stimulate iNKT cells. However, from their antitumor data, they showed some protection even in Jα18-deficient mice that would be consistent with our findings. That is, the antitumor effects induced by CpG-ODN-treated unloaded DC may also include some NK-cell-mediated antitumor protection independent of iNKT-cell function.

The data in this paper are potentially significant in two ways. First, as also proposed by van den Broeke 12 and Prins 13, the findings add a new dimension to DC therapy. DC therapy is usually considered as a means of inducing adaptive T-cell immunity, but it is evident that the therapy also induces prolonged heightened innate resistance. To elucidate the mechanism for the long-term NK-cell reactivity, host DC and CD4+ T cells should be further analyzed. Martin-Fontecha et al. 32 recently addressed the fact that in secondary lymphoid organs, DC can induce a chronic inflammatory condition that promotes autoimmune diseases; in this work, CD4+ effector memory T cells play a role, and the involvement of several inflammatory cytokines is suggested. Therefore, we will need to clarify in our experimental system the DC-CD4+T-cell interaction in detail, i.e. subsets of CD4+T cells and the requirement for costimulatory or adhesion molecules on DC, and also soluble factors for activating NK cells in vivo.

Studies in patients receiving DC therapy are required to extend this idea to humans, i.e. if the innate lymphocyte-mediated surveillance against tumor cells represents an advantage of DC therapy. A second implication of our work, given the role for recipient DC, is that the pathways in this paper are also taking place to some extent in non-vaccinated animals. DC, possibly during their normal antigen presentation function to T cells, might always be generating an environment to activate innate NK and NKT cells. This may be the reason as to why at least a fraction of these innate cells are able to work so quickly in biological defense upon an appropriate challenge.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

Mice and cell lines

We purchased pathogen-free C57BL/6 and BALB/c female mice at 6–8 wk from CLEA Japan, and IL-2−/−, and IL-4−/−, IFN-γ−/− mice from Jackson Laboratory or IL-15−/− mice from Taconic Farm. μMT−/−mice were kindly provided by Dr. T. Kurosaki (RIKEN, Japan). B6 CD11c-DTR/GFP Tg mice were a kind gift from Dr. D. Littman (New York University, NY) and were backcrossed 12 generations to C57BL/6. B6 CD1d−/− and Jα18−/− mice were provided by Dr. M. Taniguchi (RIKEN) and were backcrossed more than 9 generations to C57BL/6 mice 33, 34. The mice listed above and B6 Rag2−/− mice were maintained under specific pathogen-free conditions and studied in compliance with institutional guidelines. B16, YAC, EL4 and J558 cell lines were obtained from the American Type Culture Collection. MC38 cell line and Colo26 were kindly provided by Dr. M.T. Lotze (University of Pittsburgh, Pittsburgh, PA) and Dr. T. Saito (RIKEN) respectively.

Reagents

The following mAb were purchased from BD PharMingen: anti-mouse CD3 (145-2C11), -CD19 (1D3), -CD69 (H1.2F3), -NK1.1 (PK136), -IFN-γ (XMG1.2), -IL-4 (11B11) and mouse IgG1 (A85-1). For flow cytometry of iNKT cells, we used recombinant soluble dimeric mouse CD1d:Ig (BD PharMingen). For analysis, FACSCalibur™ instrument and CELLQuest™ (BD Biosciences) or FlowJo (Tree Star) software were used. For depletion of cell subsets, anti-NK1.1, anti-CD4 and anti-CD8 antibodies were prepared from hybridomas in our laboratory. Anti-asialo-GM1 was purchased from Wako Pure Chemical Industries.

Cell preparation

Primary cells were isolated from spleen, liver, lung, bone marrow and lymph node in C57BL/6 mice as previously described 34. Bone-marrow-derived DC were generated in the presence of GM-CSF as previously described 1, 35. On day 6, α-GalCer (100 ng/mL) was added to DC for 40 h. For DC maturation, cultures were pulsed for 16 h with 100 ng/mL of LPS or for 4 h with TNF-α (500 U/mL; R&D). In some experiments, DC were generated in mouse serum rather than in FBS. B-blast cells were generated from CD43 fraction of spleen cells, which were stimulated with 25 μg/mL LPS for 2 days as described previously 36.

Cytokine assays

IFN-γ release from NK or NKT cells was determined using a secretion assay detection kit according to the manufacturer's instructions (Miltenyi Biotec). Briefly, the cells were incubated on ice for 5 min and then diluted with warm RPMI with 5% FBS following incubation for 45 min at 37°C. After two washes, cells were resuspended in cold MACS buffer and incubated with PE-coupled IFN-γ detection reagent for 10 min at 4°C, following staining other surface marker, anti-CD3-FITC and anti-NK1.1-APC. For intracellular cytokine staining of NK or NKT cells by FACS, the cells were incubated in the presence of Golgi Plug (BD Bioscience) for 2 h and then preincubated with anti-CD16/32 Ab to block FcγR, washed, incubated with anti-CD1d-dimer-Gal followed by anti-mouse IgG1-PE and CD19-FITC for NKT cells or NK1.1-APC and CD3-FITC mAb for NK cells. After the cell surface was labeled with mAb, cells were permeabilized in Cytofix-Cytoperm Plus (BD Bioscience) and stained with anti-IFN-γ-APC or PE.

In vivo tumor studies

Mice were immunized i.v. with DC or DC/Gal (1×106) and challenged with 2×105 B16 melanoma at various time points. Mice were killed 14 days after tumor inoculation, the lungs were removed and individual surface lung metastases were counted with the aid of a microscope. In some experiments, mice were treated intraperitoneally with 50 μL of polyclonal Ab to asialo-GM1 (Wako Pure Chemical Industries) 3 days before the injection of B16 every other day until day 14.

In other experiments both NK and NKT cells were eliminated by treatment with anti-NK1.1 mAb.

Statistical analysis

Differences were analyzed using Mann–Whitney U-test. p<0.05 was considered statistically significant.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

We thank Yuri Kurosawa, Mikiko Fukui and Yohei Shozaki for providing technical assistance. α-GalCer was kindly provided from Dr. Yasuyuki Ishii (RIKEN). We thank Dr. Ralph M. Steinman for critical reading of the paper and helpful comments. This work is supported by grants from the Ministry of Education, Science, Sports, and Culture of Japan (to K.S.) and from Mitsubishi Pharma Research Foundation (to S.-i.F.).

Conflict of interest: The authors declare no financial or commercial conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  • 1
    Fujii, S., Shimizu, K., Kronenberg, M. and Steinman, R. M., Prolonged interferon-γ producing NKT response induced with α-galactosylceramide-loaded dendritic cells. Nat. Immunol. 2002. 3: 867874.
  • 2
    Fujii, S., Shimizu, K., Hemmi, H., Fukui, M., Bonito, A. J., Chen, G., Franck, R. W. et al., Glycolipid α-C-galactosylceramide is a distinct inducer of dendritic cell function during innate and adaptive immune responses of mice. Proc. Natl. Acad. Sci. USA 2006. 103: 1125211257.
  • 3
    Fujii, S., Shimizu, K., Hemmi, H. and Steinman, R. M., Innate Vα14+ natural killer T cells mature dendritic cells, leading to strong adaptive immunity. Immunol. Rev. 2007. 220: 183198.
  • 4
    Carnaud, C., Lee, D., Donnars, O., Park, S. H., Beavis, A., Koezuka, Y. and Bendelac, A., Cutting edge: Cross-talk between cells of the innate immune system: NKT cells rapidly activate NK cells. J. Immunol. 1999. 163: 46474650.
  • 5
    Smyth, M. J., Wallace, M. E., Nutt, S. L., Yagita, H., Godfrey, D. I. and Hayakawa, Y., Sequential activation of NKT cells and NK cells provides effective innate immunotherapy of cancer. J. Exp. Med. 2005. 201: 19731985.
  • 6
    Granucci, F., Zanoni, I., Pavelka, N., Van Dommelen, S. L., Andoniou, C. E., Belardelli, F., Degli Esposti, M. A. and Ricciardi-Castagnoli, P., A contribution of mouse dendritic cell-derived IL-2 for NK cell activation. J. Exp. Med. 2004. 200: 287295.
  • 7
    Andrews, D. M., Scalzo, A. A., Yokoyama, W. M., Smyth, M. J. and Degli-Esposti, M. A., Functional interactions between dendritic cells and NK cells during viral infection. Nat. Immunol. 2003. 4: 175181.
  • 8
    Moretta, L., Ferlazzo, G., Bottino, C., Vitale, M., Pende, D., Mingari, M. C. and Moretta, A., Effector and regulatory events during natural killer-dendritic cell interactions. Immunol. Rev. 2006. 214: 219228.
  • 9
    Zanoni, I., Granucci, F., Foti, M. and Ricciardi-Castagnoli, P., Self-tolerance, dendritic cell (DC)-mediated activation and tissue distribution of natural killer (NK) cells. Immunol. Lett. 2007. 110: 617.
  • 10
    Munz, C., Steinman, R. M. and Fujii, S., Dendritic cell maturation by innate lymphocytes: coordinated stimulation of innate and adaptive immunity. J. Exp. Med. 2005. 202: 203207.
  • 11
    Swann, J. B., Hayakawa, Y., Zerafa, N., Sheehan, K. C., Scott, B., Schreiber, R. D., Hertzog, P. and Smyth, M. J., Type I IFN contributes to NK cell homeostasis, activation, and antitumor function. J. Immunol. 2007. 178: 75407549.
  • 12
    van den Broeke, L. T., Daschbach, E., Thomas, E. K., Andringa, G. and Berzofsky, J. A., Dendritic cell-induced activation of adaptive and innate antitumor immunity. J. Immunol. 2003. 171: 58425852.
  • 13
    Prins, R. M., Vo, D. D., Khan-Farooqi, H., Yang, M. Y., Soto, H., Economou, J. S., Liau, L. M. and Ribas, A., NK and CD4 cells collaborate to protect against melanoma tumor formation in the brain. J. Immunol. 2006. 177: 84488455.
  • 14
    Adam, C., King, S., Allgeier, T., Braumuller, H., Luking, C., Mysliwietz, J., Kriegeskorte, A. et al., DC-NK cell cross talk as a novel CD4+T-cell-independent pathway for antitumor CTL induction. Blood 2005. 106: 338344.
  • 15
    Dworacki, G., Cicinnati, V. R., Beckebaum, S., Pizzoferrato, E., Hoffmann, T. K. and De Leo, A. B., Unpulsed dendritic cells induce broadly applicable anti-tumor immunity in mice. Cancer Biol. Ther. 2005. 4: 5056.
  • 16
    Jung, S., Unutmaz, D., Wong, P., Sano, G.-I., De los Santos, K., Sparwasser, T., Wu, S. et al., In vivo depletion of CD11c+ dendritic cells abrogation priming of CD8+ T cells by exogenous cell-associated antigens. Immunity 2002. 17: 211220.
  • 17
    Kinjo, Y., Wu, D., Kim, G., Xing, G. W., Poles, M. A., Ho, D. D., Tsuji, M. et al., Recognition of bacterial glycosphingolipids by natural killer T cells. Nature 2005. 434: 520525.
  • 18
    Kinjo, Y., Tupin, E., Wu, D., Fujio, M., Garcia-Navarro, R., Benhnia, M. R., Zajonc, D. M. et al., Natural killer T cells recognize diacylglycerol antigens from pathogenic bacteria. Nat. Immunol. 2006. 7: 978986.
  • 19
    Mattner, J., Debord, K. L., Ismail, N., Goff, R. D., Cantu C., III, Zhou, D., Saint-Mezard, P. et al., Exogenous and endogenous glycolipid antigens activate NKT cells during microbial infections. Nature 2005. 434: 525529.
  • 20
    Nagarajan, N. A. and Kronenberg, M., Invariant NKT cells amplify the innate immune response to lipopolysaccharide. J. Immunol. 2007. 178: 27062713.
  • 21
    Brigl, M., Bry, L., Kent, S. C., Gumperz, J. E. and Brenner, M. B., Mechanism of CD1d-restricted natural killer T cell activation during microbial infection. Nat. Immunol. 2003. 4: 12301237.
  • 22
    Paget, C., Mallevaey, T., Speak, A. O., Torres, D., Fontaine, J., Sheehan, K. C., Capron, M. et al., Activation of invariant NKT cells by toll-like receptor 9-stimulated dendritic cells requires type I interferon and charged glycosphingolipids. Immunity 2007. 27: 597609.
  • 23
    Borg, C., Jalil, A., Laderach, D., Maruyama, K., Wakasugi, H., Charrier, S., Ryffel, B. et al., NK cell activation by dendritic cells (DCs) requires the formation of a synapse leading to IL-12 polarization in DCs. Blood 2004. 104: 32673275.
  • 24
    Ferlazzo, G., Tsang, M. L., Moretta, L., Melioli, G., Steinman, R. M. and Munz, C., Human dendritic cells activate resting NK cells and are recognized via the NKp30 receptor by activated NK cells. J. Exp. Med. 2002. 195: 343351.
  • 25
    Gerosa, F., Baldani-Guerra, B., Nisii, C., Marchesini, V., Carra, G. and Trinchieri, G., Reciprocal activating interaction between Natural Killer cells and dendritic cells. J. Exp. Med. 2002. 195: 327333.
  • 26
    Vitale, M., Della Chiesa, M., Carlomagno, S., Pende, D., Arico, M., Moretta, L. and Moretta, A., NK-dependent DC maturation is mediated by TNFalpha and IFNgamma released upon engagement of the NKp30 triggering receptor. Blood 2005. 106: 566571.
  • 27
    Yamaguchi, T., Hirota, K., Nagahama, K., Ohkawa, K., Takahashi, T., Nomura, T. and Sakaguchi, S., Control of immune responses by antigen-specific regulatory T cells expressing the folate receptor. Immunity 2007. 27: 145159.
  • 28
    Yamaguchi, T. and Sakaguchi, S., Regulatory T cells in immune surveillance and treatment of cancer. Semin. Cancer Biol. 2006. 16: 115123.
  • 29
    Berzofsky, J. A. and Terabe, M., NKT cells in tumor immunity: opposing subsets define a new immunoregulatory axis. J. Immunol. 2008. 180: 36273635.
  • 30
    Ambrosino, E., Terabe, M., Halder, R. C., Peng, J., Takaku, S., Miyake, S., Yamamura, T. et al., Cross-regulation between type I and type II NKT cells in regulating tumor immunity: a new immunoregulatory axis. J. Immunol. 2007. 179: 51265136.
  • 31
    Akbari, O., Stock, P., Meyer, E. H., Freeman, G. J., Sharpe, A. H., Umetsu, D. T. and DeKruyff, R. H., ICOS/ICOSL interaction is required for CD4+invariant NKT cell function and homeostatic survival. J. Immunol. 2008. 180: 54485456.
  • 32
    Martin-Fontecha, A., Baumjohann, D., Guarda, G., Reboldi, A., Hons, M., Lanzavecchia, A. and Sallusto, F., CD40L+CD4+memory T cells migrate in a CD62P-dependent fashion into reactive lymph nodes and license dendritic cells for T cell priming. J. Exp. Med. 2008. 205: 25612574.
  • 33
    Shimizu, K., Goto, A., Fukui, M., Taniguchi, M. and Fujii, S., Tumor cells loaded with α-galactosylceramide Induce innate NKT and NK cell-dependent resistance to tumor implantation in mice. J. Immunol. 2007. 178: 28532861.
  • 34
    Shimizu, K., Kurosawa, Y., Taniguchi, M., Steinman, R. M. and Fujii, S., Cross-presentation of glycolipid from tumor cells loaded with α-galactosylceramide leads to potent and long-lived T cell mediated immunity via dendritic cells. J. Exp. Med. 2007. 204: 26412653.
  • 35
    Inaba, K., Swiggard, W. J., Steinman, R. M., Romani, N. and Schuler, G., Isolation of dendritic cells. Curr. protoc. Immunol. 1998. Chapter 3. unit 3.7.13.7.15.
  • 36
    Inaba, K., Turley, S., Yamaide, F., Iyoda, T., Mahnke, K., Inaba, M., Pack, M. et al., Efficient presentation of phagocytosed cellular fragments on the MHC class II products of dendritic cells. J. Exp. Med. 1998. 188: 21632173.