Background
Dendritic cells (DCs) are professional antigen presenting cells (APCs) that initiate specific immune responses against pathogens [
1] and tumor cells [
2]. Immature DCs which locate in the tissues and the peripheral lymphoid organs persistently surveillance the environment and recognize the invading pathogens and cell debris [
3], and capture antigens by phagocytosis, micropinocytosis, and endocytosis. After the antigen recognition and uptake, the immature DCs undergo a series of maturation events, including the up-regulation of the major histocompatibility complex (MHC) II and the co-stimulatory molecules, the secretion of cytokines, the outgrowth of dendrites, and the modulation of chemokine receptor expression profile accompanied by the migration into the T-cell areas of the peripheral lymphoid organs [
4]. The antigen-loaded mature DCs can activate T-cells through the interaction between MHC II-peptide and T-cell receptor (TCR) complex, and can activate B-cells [
5] and NK-cells [
6] through specific ligands and cytokines expressed by DCs. Based on the differential expression of cell surface markers, DCs are grouped into two major classes including conventional DCs (cDCs) [
7] and plasmacytoid DCs (pDCs) [
8]. cDCs are further subdivided into different populations including the lymphoid tissue-resident DCs and the peripheral tissues-located migrating DCs [
1,
4].
DCs play critical roles in the initiation, programming and regulation of the anti-tumor immunity [
9,
10]. Nevertheless, as supported by both experimental studies and clinical observations, the immune responses against tumor cells are severely compromised in most, if not all, progressing solid tumors. The tumor infiltrating myeloid-derived suppressor cells (MDSCs), the tumor-associated macrophages (TAMs) [
11,
12] and the cytokines secreted by MDSCs and TAMs cooperatively create an immunosuppressive environment which leads to the suppression of DC functions and the induction of regulatory T-cells. Gerner et al reported that murine tumors were extensively infiltrated by partially activated tumor-infiltrating DCs (TIDCs) which had inefficient MHC II presentation due to poor intrinsic protein uptake capability, resulting in the inferior initiation of T-cell responses in the draining lymph nodes [
13]. These resting, non-activated, immature phenotypes of DCs have also been discovered in cancer patients [
14]. Moreover, DCs have been considered as a promising agent to generate effective anti-tumor immune therapies, because DCs can be generated in large numbers, and the cultured immature DCs could be converted into mature DCs through the antigen loading with peptides, recombinant proteins, tumor antigen-encoding mRNA, and whole tumor cell lysates. These DCs can be delivered to the tumor sites or the lymph nodes to activate T-cell responses against tumors [
15]. However, although the use of mature DCs as cellular vaccines showed promising anti-tumor effects in many mouse tumor models such as the B16 melanoma [
16], the Lewis lung cancer, the D2F2/E2 breast tumor and the EL4/E2 thymoma [
17], the application of mature DCs in phase III clinical trails in human cancer patients with prostate cancer [
18] or melanoma [
15] have largely failed. Therefore, the fully understanding of the molecular mechanisms regulating DC maturation and activation, which is still obscure, is a prerequisite for the DC-based anti-tumor therapies.
The Notch signaling pathway is an evolutionarily conserved pathway that regulates development by participating in cell fate determinations and cell proliferation, differentiation and apoptosis during embryonic and postnatal stages [
19]. There are four Notch receptors (Notch1-4) and five ligands (Jagged1, Jagged2, and Delta-like (Dll)1, 3, and 4) in mammals. After the triggering of the Notch receptors by the binding of the Notch ligands, the Notch intracellular domain (NIC) is cleaved by a proteinase complex containing γ-secretase. NIC then translocates into the nucleus, where it interacts with the transcription factor C promoter-binding factor 1/recombination signal-binding protein J/κ (RBP-J) [
20,
21]. This protein complex will recruit other transcription co-activators, and transactivate the transcription of the target genes such as the Hes family basic helix-loop-helix members [
22].
The Notch signaling plays an important role in the DC genesis. Both of the Notch ligands Jagged1 and Dll1 can activate Notch signaling in DCs, but their effect on DC differentiation is different. Dll1-expressing fibroblasts could induce DC differentiation, whereas Jagged1-expressing fibroblasts inhibit DC differentiation and promote the accumulation of immature myeloid cells [
23]. Similar results have also been reported by several other groups. For example, Weijzen et al have shown that Jagged1 is able to induce the maturation of the monocyte-derived human DCs [
24], while Ohishi et al have also indicated that Dll1 promotes the DC differentiation [
25,
26]. Cheng et al showed that the differentiation of DCs was significantly impaired in mice expressing Notch1 anti-sense RNA [
27] and in a system involving the Notch1 deficient embryonic stem (ES) cells [
28]. The deletion of RBP-J in DCs could result in the reduction of the conventional DCs in the spleens of the mice. This decrease is primarily limited to the CD8
- DC subset in the marginal zone of the spleens. As CD8
- DCs in the marginal zone were found to reside in close contact with Dll1-expressing cells, Dll1 could also be involved in the loss of CD8
- DCs [
29]. However, Sekine et al showed that the blocking of Dll1 alone had no significant effect on the maintenance of CD8
- DCs and the blocking of Dll1, Dll4, Jagged1 and Jagged2 significantly decreased CD8
- DCs [
30]. pDCs are different from cDCs in the phenotype and the function. Several studies have reported that the differentiation of pDCs could also be affected by Notch signaling. Oliver et al reported that Dll1 could increase the numbers of pDC through promoting the differentiation rather than affecting the proliferation. A γ-secretase inhibitor (GSI) could block this effect. But another group reported the opposite result, showing that Dll1 could block pDC development [
31].
We have recently shown that the RBP-J-mediated Notch signaling plays a critical role in the maturation of the LPS-induced DCs [
32]. In order to investigate whether the Notch signaling participates in the DC-mediated anti-tumor immunity, we established tumor-bearing mouse models by using several mouse tumor cell lines and RBP-J-deleted DCs [
33]. We found that the absence of RBP-J in DCs led to impaired DC-dependent anti-tumor immune responses. We further demonstrated that the RBP-J deficient DCs could not undergo a full activation process upon tumor antigen stimulation, which resulted in an inefficient T-cell activation and tumor progress.
Discussion and Conclusion
As the strongest professional antigen presenting cells, DCs play critical roles in the activation of the anti-tumor immunity. Much attention has been paid to the utilization of DC vaccines to initiate efficient tumor specific effector T-cells to inhibit tumor growth. Indeed, in 1990s, many investigators have attempted to use tumor antigen-loaded DCs in the clinical trials of established tumors after the initial success with the preventive and therapeutic DC vaccination in mouse models [
36]. Although the use of mature DCs as cellular vaccines had provided encouraging and exciting anti-tumor effects in many mouse tumor models [
16,
17], the applications of mature DCs in the phase III clinical trails in human cancer patients with prostate cancer [
18] or melanoma [
15] have largely failed. Obviously, the understanding of the molecular mechanisms regulating DC maturation is essential for the DC-based anti-tumor therapies.
Notch signaling pathway might influence tumor growth in multiple ways [
37]. In this study, we showed that the disruption of the transcription factor RBP-J, which is critical in the mediation of the canonical Notch signaling pathway, attenuated the DC-dependent anti-tumor immunities. By using co-inoculation of four types of tumor cells with DCs in mice, which mimic the injection of activated DCs directly into tumor tissues, we found that tumors co-inoculated with the RBP-J deficient DCs were significantly bigger than tumors containing the control DCs. The inefficient infiltration of the immune cells including T-cells, B-cells and NK cells into the tumor tissues containing RBP-J
-/- DCs further suggested that DCs with Notch signaling deficiency could not prime adaptive immune responses to tumor cells. Although previous studies have shown that Notch signaling affects the development, differentiation and function of DCs, the results reported here directly showed that the Notch signaling pathway is important for DCs to evoke efficient anti-tumor immune responses in mice.
The phenotypes of the tumor antigen-loaded RBP-J deficient DCs are reminiscent of immature DCs. DCs are derived from hematopoietic stem cells accommodated in BM. Immature DCs are generated through a series of complicated differentiation steps in BM, and are exported into the blood stream and thereafter enter into the secondary lymphoid organs and the peripheral tissues. Normally, upon the uptaking and processing of tumor antigens in the peripheral tissues, immature DCs mature into fully activated DCs, by the gain of migration capacity, the outgrowth of dendrites, and the up-regulated expression of the MHC molecules and the co-stimulatory molecules. Mature DCs subsequently migrate into the draining lymph nodes where they present the processed antigens to resting T lymphocytes in the shape of peptide-MHC complex, to generate immune responses against pathogens and tumors. In vitro and in vivo studies have shown that DCs loaded with tumor antigens could be activated to initiate anti-tumor immune responses. In our study reported here, DCs isolated from the spleen of mouse were activated by the incubation with the crude tumor antigens, and these DCs appeared to have the ability to activate anti-tumor immunities. While the RBP-J
+/- DCs showed the properties of mature DCs upon antigen loading, RBP-J
-/- DCs appeared immature in several aspects. First, the migration of the tumor antigen-loaded RBP-J
-/- DCs was impaired. We analyzed the expression of chemokine receptors including CCR7 and CXCR4, and found that although the CXCR4 level was almost the same in both RBP-J
-/- and RBP-J
+/- DCs after tumor antigen loading, the expression of CCR7 was significantly lower in RBP-J
-/- DCs. CCR7 has been shown to play important roles in the migration of mature DCs. It is likely that the lowered expression of CCR7 mediated the reduced migration of RBP-J deficient DCs from the tumor tissues to the draining lymph nodes, but more detailed molecular experiments are needed to clarify the mechanism controlling the chemokine receptor expression in DCs. Second, the expression of MHC II was weaker in the RBP-J deficient DCs as compared with the control DCs. The up-regulation of the MHC II expression is a critical marker of DC maturation. DCs in our experimental system appeared heterogenous concerning the MHC II expression. In the tumor antigen-unloaded and loaded RBP-J
-/- DCs, the level of MHC II was lower than the control DCs. Third, the level of co-stimulatory molecules, CD80 and CD86, was significantly lower on the tumor antigen-loaded RBP-J
-/- DCs than that on the control DCs. This is especially true for the expression of CD80. Lowered expression of the co-stimulatory molecules also supports that RBP-J deficient DCs are immature after tumor antigen loading. Last but not least, RBP-J deficient DCs showed lowered capacity in the stimulation of T-cells. Our data showed that both in vitro and in vivo, the tumor antigen-loaded DCs had attenuated capacity of stimulating the T-cell proliferation, cytokine production and cytotoxicity. Interestingly, in the draining lymph nodes of the mice bearing the B16 melanoma containing RBP-J
-/- DCs, higher proportion of apoptotic cells was observed, as compared with the controls. This suggested that RBP-J
-/- DCs might have immuno-repressive activity, but further studies are needed to access cell proliferation and apoptosis in different populations of cells. The molecular mechanisms, such as the components of crude tumor antigen preparations and the involvement of TLRs, are now under investigation, although the expression of PD-L1 and PD-L2 on DCs was not changed (data not shown). Because no difference was observed in the rate of tumor antigen phagocytosis between the RBP-J deficient and the control DCs (data not shown), the immature phenotypes of the RBP-J deficient DCs could be attributed to some intrinsic defects in the maturation of the tumor antigen-loaded DCs. However, we could not exclude a possibility that RBP-J deficient DCs represent another population of DCs, such as regulatory DCs. Further studies with DC-specific RBP-J deletion mouse models [
29] and genome-wide transcription profiling are needed to clarify this point.
B-cells and NK cells were also activated by DCs. Qi et al reported that lymph node B-cells could be activated by antigen-bearing DCs by two-photon intravital imaging [
38]. Kijima et al reported that NK-cells could be activated by DCs and that Jagged2-Notch interaction is very critical in this process [
6]. Our results also showed that the proportion and absolute number of B cells and NK cells were significantly lower in RBP-J
-/- DCs, indicating that Notch signaling is also important for DCs to activate B-cells and NK cells in the anti-tumor immunity.
Taken together, the Notch signaling pathway could affect the differentiation and the maturation of DCs, and the deficiency of Notch signaling pathway could impair DC-based anti-tumor immunity. Our results suggest the important role of Notch signaling pathway in the DC-based anti-tumor immunity. Therefore, the Notch signaling might be a potential target to interfere the DC-based anti-tumor immunotherapies. However, detailed studies about the mechanisms of the regulation of DCs by Notch signaling should be unveiled before the related therapies could be achieved.
Methods
Mice
Mice were maintained in the specific pathogen-free (SPF) conditions on the C57BL/6 background. The RBP-J-floxed (RBP-J
f) mice were as described [
33] and were crossed with the Mx-Cre transgenic mice to get the RBP-J
f/f-MxCre and RBP-J
+/f-MxCre (as controls) mice (hence, referred as RBP-J
-/- and RBP-J
+/-, respectively). The mice were genotyped by the polymerase chain reaction (PCR) [
33]. Four-week-old mice were injected intraperitoneally with 300 μg/100 μl poly(I)-poly(C) (Sigma, St. Louis, MO) for four times at 2-day intervals and were then injected with the same dosage of poly(I)-poly(C) for another eight times at 1-week intervals (twelve injections in total). All animal experiments were approved by the Animal Experiment Administration Commission of Fourth Military Medical University.
Cell lines and cell culture
Lewis lung carcinoma (LLC), mouse S180 sarcoma (S180), mouse H22 hepatocarcinoma (H22), and mouse melanoma (B16) cell lines were gifts from CH Shi. All of the tumor cell lines were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 50 IU/ml penicillin, and 50 μg/ml streptomycin sulfate.
DCs were separated by anti-CD11c magnetic beads (Miltenyi Biotec GmbH, Germany) from mouse spleen cell suspensions according to the instructions of the manufacturer, and were cultured in RPMI 1640 medium as described previously [
32].
Preparation of crude tumor antigens [39]
The cultured LLC, S180, H22 and B16 cells were harvested by using a cell scraper, washed in phosphate-buffered saline (PBS), and were resuspended in PBS at approximately 5 × 107/mL. The cells were disrupted by freeze and thaw for 5 times, followed by centrifugation in a benchtop microcentrifuge for 10 min at 10,000 rpm. The supernatants were transferred into sterilized Eppendorf tubes, and were stored at -80°C for later use. In some experiments, B16 cells were labeled with the Dio Cell-Labeling Solution (Invitrogen, Carlsbad, CA) according to the manufacturer's protocol, and were then used for the preparation of the Dio-labeled B16 tumor antigens.
Tumor-bearing mouse models
Tumor cells (5 × 106) were injected subcutaneously into the normal mice. Five days after the initial inoculation, tumor growth was monitored every 2 days by measuring the tumor length (L) and short (S) with a sliding caliper. The tumor size was calculated as L × S2 × 0.51. Seventeen days after the initial inoculation, the tumors were excised and the tumor weight was measured. The tumor tissues and the tumor draining lymph nodes were minced and were filtrated through a nylon filter for flow cytometry analysis.
Flow cytometry
Single cell suspensions were resuspended with PBS containing 2% fetal calf serum and 0.05% NaN3 and were counted. Cells (3-5 × 105) were stained with antibodies at 4°C for 30 min before being analyzed using a FACSCalibur (BD Immunocytometry Systems, San Jose, CA). Dead cells were excluded by the propidium iodide (PI) gating. Data were analyzed using the CellQuest software. PE-anti-CD4 (RM4-5), APC-anti-CD8 (53-6.7), APC-anti-CD19 (1D3), biotinylated anti-NK1.1 (PK136), APC-anti-CD3 (145-2C11), biotinylated anti-IAb (KH174), and FITC-anti-BrdU were purchased from BD PharMingen (San Diego, CA). FITC-anti-CD11c (N418), biotinylated anti-CD11c, FITC-anti-CD80 (16-110A1), PE-anti-CD86 (GL-1), and Streptavidin-APC were products of BioLegend (San Diego, CA), and Streptavidin-PE and biotinylated anti-CCR7 (4B12) were from eBioscience (San Diego, CA).
Migration assay
DCs (1 × 106) were separated by anti-CD11c magnetic beads from the mouse spleen cell suspensions, and were labeled with CFSE. DCs were co-cultured with the B16 tumor antigens at a ratio of 1:2 for 12 h at 37°C, and were then collected and resuspended in 200 μl PBS, and were injected subcutaneously at the left lower abdomen of mice. The mice were sacrificed 12 h later, and the draining lymph nodes were collected for FACS analysis.
T-cell proliferation assay
DCs (2 × 105) were seeded in 96-well plates and were co-cultured with B16 tumor antigen for 12 h, followed by medium exchanging. Congeneic T-cells (1 × 106) were isolated by negative selection using magnetic beads, labeled with carboxyfluorescein diacetate succinimidyl ester (CFSE), and were co-cultured with B16 tumor antigen-loaded DCs. Five days later, T-cell proliferation was detected by FACS.
For the in vivo assay, DCs (1 × 106) were sorted by using the anti-CD11c magnetic beads, and were loaded with the B16 tumor antigens for 12 h at 37°C. The tumor antigen-loaded DCs were injected subcutaneously into normal congenic mice at the left lower abdomen. On the 5th day, the mice were injected intraperitoneally with 1 mg/100 μl BrdU twice per day for three days, with PBS as a control. The mice were sacrificed on the 7th day and the draining lymph nodes were dissected. Single cell suspensions were prepared and were stained for CD3 and BrdU (anti-BrdU-FITC) according to the manual of the BrdU Flow kit provided by the manufacturer (BD Biosciences-Pharmingen).
Enzyme-linked immunosorbent assay (ELISA)
Culture supernatants were collected and were used to detect the level of cytokines using different assay kits (Bioresun, China). OD450 was recorded using a spectrophotometer, and was compared between groups.
CTL assay
DCs were loaded with the B16 tumor antigens for 12 h, and were then co-cultured with negatively selected T-cells (1 × 106) in 96-well plates for 72 h. The culture supernatants were discarded, and the B16 tumor cells were added into the wells with different ratios of 1:10, 1:20, 1:40, 1:80 to T-cells. After co-culturing for 36 h in the serum free RPMI 1640, the supernatants were collected for the examination of cell death by using a colorimetric lactate dehydrogenase (LDH) assay kit (Cayman Chemical Company, Ann Arbor, MI) according to the manufacturer's manuals.
For the in vivo assay, the B16 tumor antigen-loaded DCs (1 × 106) were injected subcutaneously into normal C57BL/6 mice at the left lower abdomen. On the 7th day after the injection, the mice were sacrificed and the draining lymph nodes were recovered. Single cell suspensions were prepared, and T-cells were negatively selected by using magnetic beads. The T-cells (1 × 106) were then co-cultured with the target B16 tumor cells at different ratios of 10:1, 20:1, 40:1, 80:1 in 96-well plates. The cells were co-cultured in the serum free RPMI 1640 for 36 h, and the supernatants were collected for the detection of the cell death by using the LDH assay kit (Cayman Chemical Company).
Analysis of DNA content
Cells were collected from the draining lymph nodes of the tumor-bearing mice or the mice inoculated with the tumor antigen-loaded DCs. The cells (1 × 106) were fixed with 70% ethanol for 20 min at room temperature, and were then washed once with PBS. After being resuspened in 500 μl of 20 μg/ml PI solution containing 0.1% (v/v) Triton X-100 (Sigma) and 0.2 mg/ml DNase-free RNase A (Sigma), the cells were analyzed by FACS.
Quantitative real time RT-PCR
Total RNA was extracted from SPDCs using TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. Complementary DNA was prepared using a reverse-transcription kit from TOYOBO (Osaka, Japan). Real-time reverse-transcription PCR (RT-PCR) was performed using a kit (SYBR Premix EX Taq, Takara) and the ABI PRISM 7300 real-time PCR system, with β-actin as an internal control. Primers used in real-time PCR were as follows: β-actin forward: CATCCGTAAAGACCTCTATGCCAAC, β-actin reverse: ATGGAGCCACCGATCCACA, HES1 forward: GCAGACATTCTGGAAATGACTGTGA, HES1 reverse: GAGTGCGCACCTCGGTGTTA; HES5 forward: AAAGACGGCCTCTGAGCAC, HES5 reverse: GGTGCTTCACAGTCATTTCCA.
Statistics
Statistical analysis was performed with the SPSS 12.0 program. Results were expressed as means ± SD. Comparisons between groups were undertaken using the unpaired Student's t-test. P < 0.05 was considered statistically significant.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
FF designed and performed all the animal and cell biological experiments, carried out data analysis and drafted the manuscript. WYC assisted in cell biological experiments. HXB assisted in immunological experiments. LXW assisted in in vivo animal experiments. JG assisted in FACS analysis. CYR participated in animal breed. WL assisted in cell culture. HF assisted in cell magnetic sorting. DGR assisted in data collection. LL participated in tumor antigen preparation. ZHW assisted in experiment design. HH, as Director of the department, coordinated its execution and design, and drafted and produced the final version of the manuscript. All authors read and approved the present version of the manuscript.