Induction of Impaired Antitumor Immunity by Fusion of MHC Class II-Deficient Dendritic Cells with Tumor Cells
http://www.100md.com
免疫学杂志 2005年第1期
Abstract
To dissect the role of Ag presentation through MHC class I and/or II pathways by dendritic cell (DC)-tumor fusion cells, we have created various types of DC-tumor fusion cells by alternating fusion cell partners. Fusions of MC38/MUC1 carcinoma cells with DC from wild-type (WT-DC), MHC class I knockout (IKO-DC), class II knockout (IIKO-DC), or class I and II knockout (I/IIKO-DC) mice created WTDC-fusion cells (FC), IKO-FC, IIKO-FC, and I/IIKO-FC, respectively. MHC class II- and MUC1-positive fusion cells were constructed by fusion of B16/MUC1 melanoma cells with IKO-DC (IKO/B16-FC). Immunization of MUC1 transgenic mice with 5 x 105 WTDC-FC, IKO-FC, IIKO-FC, or I/IIKO-FC provided 100, 91.7, 61.5, and 15.4% protection, respectively, against tumor challenge with MC38/MUC1 cells. In contrast, all mice immunized with irradiated MC38/MUC1 tumor cells or WT-DC developed tumors. One group of mice was immunized with 5 x 105 IKO/B16-FC and then challenged with B16/Ia+/MUC1 on one flank and MC38/MUC1 on the other flank. Immunization of these mice with IKO/B16-FC resulted in 100 and 78.6% protection against B16/Ia+/MUC1 and MC38/MUC1 tumor challenge, respectively. The antitumor immunity induced by immunization with IKO/B16-FC was able to inhibit the growth of MHC class II-negative tumor. In addition, in vivo results correlated with the induction of Ag-specific CTL. Collectively, the data indicate that MHC class II Ag presentation targeting activation of CD4 T cells is indispensable for antitumor immunity.
Introduction
Effective Ag processing and presentation are crucial to antitumor immunity. Tumor cells express tumor Ags, but seldom induce effective antitumor immune response. This observation is probably related to the fact that tumor cells do not express essential costimulatory molecules and seldom express MHC class II molecules. Therefore, effort has been made to enhance tumor cell-based vaccine using genetically modified tumor cells. Immunization of mice with autologous class II- and/or B7-transfected tumor cells provides protection against challenge with wild-type tumor (1, 2) and mediates regression of established tumor (3). Alternatively, a strategy has been developed based on dendritic cells (DC),4 which are bone marrow-derived leukocytes bearing a characteristic veiled morphology that excel in Ag presentation and the initiation of primary immune responses (4, 5, 6). DC reside at the port of entry, take up exogenous Ags, and migrate to draining lymph nodes, where the Ags are presented to CD4 T cells through MHC class II pathways. In addition, DC are capable of initiating CD8 T cell response through a cross-presentation pathway. Exogenous Ags from tumor cells can be delivered to the cytosol, processed, and presented through an endogenous pathway (7, 8, 9, 10).
Another evolving strategy is the use of DC fused to tumor cells (11). In this approach, tumor Ags are delivered to DC, processed, and presented through both MHC class I and II pathways in the context of costimulatory molecules. The fusion cells function like APCs with the ability to migrate to draining lymph nodes, where they interact with CD4 and CD8 T cells and induce potent antitumor immunity (12, 13). Coculture of human peripheral blood monocytes with DC-tumor fusion cells induces both CD4 and CD8 T cells (14, 15). However, the role of MHC class I-restricted or class II-restricted Ag presentation and the activation of CD4 and CD8 T cells in the antitumor immune responses are not well defined. In the present study, we created various types of DC-tumor fusion cells with intact or deficient expression of MHC class I or II molecules by using several kinds of DC and tumor fusion partners. The fusion cells were used in the prevention and treatment of tumors in MUC1 transgenic mice (MUC1.Tg). We observed differential impairment of antitumor immunity induced by fusions of DC from MHC class I and/or II knockout mice. Immunization with MHC class II-deficient DC-tumor fusion cells abolished the IFN- production of CD4 and CD8 T cells and the induction of CTL, and severely impaired antitumor immunity. These results indicate that MHC class II Ag presentation targeting activation of CD4 T cells is indispensable in antitumor immunity.
Materials and Methods
DC generation
DC were obtained from bone marrow cultures of C57BL/6 wild-type (WT), MHC class I knockout (2-microglobulin (2m)–/–) (16), MHC class II knockout (Abb–/–) (17), and MHC class I and II double-knockout(2m–/–/Abb–/–) (18) mice from Taconic Farms, using the method previously described (19). Briefly, bone marrow cells were flushed from long bones and further depleted of lymphocytes, granulocytes, B cells, and APCs by incubation with anti-CD8 (2.43), anti-CD4 (GK1.5), anti B220/CD45R (RA3-3A1/6.1), anti-Ia (B21-2), and anti-Gr-1 (RB6-8C5) (American Type Culture Collection (ATCC)) Abs, followed by rabbit complement. The cells were cultured in RPMI 1640 medium supplemented with 5% heat-inactivated FCS, 50 μM 2-ME, 1 mM HEPES (pH 7.4), 2 mM glutamine, 10 U/ml penicillin, 100 μg/ml streptomycin, and 20 ng/ml murine rGM-CSF (Sigma-Aldrich). On day 5 of culture, DC were purified by multiple steps of plating and collection of the loosely adherent population. After overnight culture, DC were harvested for phenotype analysis and fusion to MC38/MUC1 cells.
Tumor cells
Murine MC38 colon adenocarcinoma and B16 melanoma cells were stably transfected with a MUC1 cDNA, resulting in MC38/MUC1 (20, 21) and B16/MUC1 tumor cells (22). The B16/MUC1 cells positive for MUC1, but negative for both MHC class I and II were selected and fused to B16 positive for MHC class II (B16/Ia+; a kind gift from S. Ostrand-Rosenberg, University of Maryland, Baltimore, MD) to create B16/Ia+/MUC1 tumor cells. The human breast carcinoma cell line, MCF-7, was obtained from ATCC. Cells were maintained in DMEM supplemented with 10% heat-inactivated FCS, 2 mM L-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin.
DC-Tumor fusion process
DCs were fused to MC38/MUC1 carcinoma cells using a previously described method (11). Briefly, DC generated from WT, 2m–/–, Abb–/–, and 2m–/–Abb–/– mice were designated as WT-DC, class I knockout (IKO)-DC, class II knockout (IIKO)-DC, and I/IIKO-DC, respectively, and were mixed with MC38/MUC1 tumor cells at a 10:1 ratio. The fusion process was conducted with 50% polyethylene glycol (Sigma-Aldrich) in prewarmed Dulbecco’s PBS without Ca2+ or Mg2+ at pH 7.4 for 5 min. The polyethylene glycol solution was then diluted by slow addition and mixing of 1, 2, 4, 8, and 16 ml of warm serum-free medium. The cell pellets obtained after centrifuge at 10,000 rpm were resuspended in RPMI 1640 medium supplemented with 10% heat-inactivated FCS, 2 mM glutamine, 10 mM nonessential amino acids, 1 mM sodium pyruvate, 10% NCTC 109, 10 U/ml penicillin, 100 μg/ml streptomycin, and 10 ng/ml murine rGM-CSF. The fused cells were plated in 24-well culture plates for 5–7 days. The fusions of WT-DC, IKO-DC, IIKO-DC, and I/IIKO-DC with MC38/MUC1 created WTDC-fusion cells (FC), IKO-FC, IIKO-FC, and I/IIKO-FC, respectively. The fusion of IKO-DC with B16/MUC1 resulted in IKO/B16-FC. On day 5–7 of culture, the unfused tumor cells grew firmly attached to the tissue culture flask, while the loosely attached DC-tumor fusion cells were dislodged by gentle pipetting. In the in vitro studies, the fusion cells were further purified with cell sorting.
Flow cytometry
MC38/MUC1 tumor cells, WT-DC, IKO-DC, IIKO-DC, and I/IIKO-DC, and WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC were stained with FITC-conjugated HMPV (anti-MUC1; BD Pharmingen) for 30 min on ice. After being washed with PBS, the cells were incubated with PE-conjugated KH95 (anti-MHC class I), M5/114 (anti-MHC class II), and GL1 (anti-B7-2) (BD Pharmingen) for an additional 30 min on ice. The cells were washed, fixed, and analyzed by FACScan (BD Biosciences) with CellQuest analysis software. To determine the fusion efficiency, MUC1 and MHC class II and/or B7-2 were used as tumor or DC marker, respectively.
Mixed leukocyte reaction
WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC were stained with FITC anti-MUC1 and PE anti-B7-2 mAbs and selected by cell sorting (DakoCytomation) with Summit v3.0 analysis software. The fusion cells were exposed to ionizing radiation (30 Gy) and then cocultured with allogeneic (BALB/c) T cells at various ratios (1:20–1:540) in 96-well flat-bottom culture plates for 5 days. The responding T cells ranged from 2 x 105 to 5.4 x 106 in 200 μl/well. WT-DC, IKO-DC, IIKO-DC, and I/IIKO-DC and irradiated MC38/MUC1 and T cells alone were used as controls. Cells were pulsed with 1 μCi of [3H]thymidine (New England Nuclear) per well for 12 h and then collected on filters with a semiautomatic cell harvester. Tritium incorporation was quantified by liquid scintillation. All determinations were conducted in triplicate and expressed as the mean ± SD.
RT-PCR
Lymph node cells (LNC) were isolated from mice immunized twice with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC or treated with PBS, and were sorted into CD4 and CD8 subsets (purity 97–98%). RNA from 1 x 106 sorted CD4 or CD8 T cells was extracted by TRIzol reagent (Invitrogen Life Technologies). Total RNA to cDNA was reverse transcribed using a poly(dT) oligonucleotide and SuperScript (Invitrogen Life Technologies). PCR was performed by amplifying cDNA with the following oligonucleotide primer (23): murine IL-2 (5'-TCCACTTCAAGCTCTACAG-3' and 5'-GAGTCAAATCCAGAACATGCC-3'); IFN- (5'-CATTGAAAGCCTAGAAAGTCTG-3' and 5'-CTCATGGAATGCATCCTTTTTCG-3'); -actin (5'-TGTGATGGTGGGAATGGGTCAG-3' and 5'-TTTGATGTCACGCACGATTTCC-3') (Stratagene). PCR-amplified products were analyzed on a 2% agarose gel.
51Cr cytotoxicity assay
Spleens from MUC1.Tg mice immunized twice with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC were removed, and T cells were isolated into single cell suspensions for use as effector cells. The targets included MC38, MC38/MUC1, B16, B16/MUC1, and MCF-7 tumor cells. Briefly, tumor cells (1–2 x 106 cells) were labeled with 100–200 μCi of Na251CrO4 for 60 min at 37°C, followed by thorough washing to remove unincorporated isotope. T cells and tumor targets were resuspended in culture medium and then combined at various E:T ratios in 96-well V-bottom plates. The plates were centrifuged at 100 x g for 5 min to initiate cell contact and incubated for 5 h at 37°C with 5% CO2. After incubation, supernatants were collected and radioactivity was quantified in a gamma counter. Spontaneous release of 51Cr was determined by incubation of targets in the absence of effectors, and maximum or total release of 51Cr by incubation of targets in 0.1% Triton X-100. The percentage of specific release of 51Cr was calculated using the following: percentage of specific release = ((experimental – spontaneous)/(maximum – spontaneous)) x 100.
In vivo tumor prevention
C57BL/6 MUC1.Tg mice (a kind gift from Dr. S. J. Gendler, Mayo Clinic, Scottsdale, AZ) that express MUC1 at a level similar to that found in humans (24) were used. Seven-week-old MUC1.Tg mice were immunized s.c. on days 0 and 7 with 5 x 105 WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC calculated on the basis of fusion efficiency. WT-DC were used as a control. On day 14, the mice were challenged by s.c. injection in the flank with 5 x 105 syngeneic MC38/MUC1 (left side) and MC38 (right side) tumor cells.
To determine the antitumor immunity induced by MHC class II-expressing, but class I-deficient fusion cells, groups of MUC1.Tg mice were immunized twice s.c. with 5 x 105 IKO/B16-FC. Control mice were immunized with IKO-DC or irradiated tumor cells. Seven days after the second vaccination, the mice were challenged with 5 x 105 MC38/MUC1 tumor cells (MHC class I and MUC1 positive) on the left flank and B16/Ia+/MUC1 tumor cells (MUC1 and MHC class II positive, but class I negative) on the right flank.
All of the mice were followed for 30 days. The size of tumor was determined by measuring perpendicular dimensions with a vernier caliper every 2–3 days. Tumors with a diameter of 3 mm were designated as positive. The mice were maintained in microisolator cages under specific pathogen-free conditions.
Treatment of established pulmonary metastases
Pulmonary metastases were established by i.v. injection of 1 x 106 MC38/MUC1 tumor cells through the tail vein in 7-wk-old MUC1.Tg mice. Two days after the tumor inoculation, mice were immunized with 1 x 106 WTDC-FC, IKO-FC, IIKO-FC, I/IIKO-FC, or WT-DC. The immunization was repeated on day 8. The mice were sacrificed 20 days after the last immunization. Pulmonary metastases were enumerated by counting after staining the lungs with india ink (25).
Statistical analysis
Statistical significance was analyzed using 2 and Student’s t tests.
Results
Characterization of various types of DC-tumor fusion cells
Constitutive deletion of the gene for 2m results in loss of expression of MHC class I in all nucleated cells (16), whereas mutation in the Ab gene abrogates MHC class II expression in all APCs (17). Therefore, DC from the 2m–/–/Abb–/– double-knockout mice are devoid of expression of MHC class I and II molecules (18). To confirm these findings, the phenotype of various types of DC, DC-tumor fusion cells, and tumor cells was assessed by FACS analysis. MC38/MUC1 tumor cells expressed MUC1 and MHC class I molecules (Fig. 1A). DC isolated from WT, 2m–/–, and Abb–/– mice expressed MHC class I and II, class II, and class I molecules (Fig. 1B), respectively. 2m–/–/Abb–/– mice expressed no MHC molecules. However, MHC class I and/or II deficiency did not affect the expression of B7-2 on DC. Fusion of MC38/MUC1 with WT-DC resulted in coexpression of MUC1 with MHC class I and II molecules or B7-2 (Fig. 1C). Similar results were obtained with IKO-DC fused to MC38/MUC1, indicating that tumor-derived MHC class I molecules were expressed. In contrast, IIKO-DC or I/IIKO-DC fused with MC38/MUC1 led to the expression of MUC1 and MHC class I or B7-2 molecules, but not MHC class II molecules (Fig. 1C). These data indicate that the properties of fusion cells are dictated by their parent cells and that DC-tumor fusion cells deficient in MHC class II expression can be created by using DC from Abb–/– or 2m–/–/Abb–/– mice as fusion cell partners.
Stimulation of T cell proliferation compromised in IIKO-FC
To determine the ability of fusion cells to stimulate T cell proliferation, the MLR assay was used. T cells from splenocytes of BALB/c mice cocultured with WT-DC or WTDC-FC proliferated vigorously (Fig. 2A, left upper panel). Coculture of these T cells with IKO-DC or IKO-FC also resulted in proliferation of T cells, albeit at a lower level (Fig. 2A, left lower panel). In contrast, impaired T cell proliferation was observed when T cells were cocultured with IIKO-DC or I/IIKO-DC or their fusion counterparts (Fig. 2A, right upper and lower panels). These data indicate that MHC molecules on DCs or fusion cells are vital for T cell proliferation in MLR.
IFN- abolished in CD4 and CD8 T cells primed by IIKO-FC
Cytokine production is the hallmark of T cell activation. To determine whether cytokine expression was affected in T cells primed by various types of DC-tumor fusion cells, we used RT-PCR to assess the cytokine mRNA levels of LNC isolated 7 days after the second immunization. Whereas sorted CD4 T cells from mice immunized with WTDC-FC or IKO-FC expressed IL-2 and IFN- (Fig. 2B), the expression of IFN- was abolished in CD4 T cells primed by IIKO-FC or I/IIKO-FC. The expression of IL-2 was also abolished in CD4 T cells primed by I/IIKO-FC. Similarly, IFN- was detected in sorted CD8 T cells from mice immunized with WTDC-FC or IKO-FC. However, IFN- was not detected in CD8 T cells primed by IIKO-FC or I/IIKO-FC. These results indicate that MHC class II-restricted Ag presentation affects downstream cytokine production of both CD4 and CD8 T cells. The secretion of IFN- was abolished in T cells primed by MHC class II-deficient fusion cells, indicating the impairment of T cell activation.
Presentation of MHC class II Ag required for effective induction of CTL
CTL response against tumor cells was evaluated by standard 51Cr release assays to assess the effectiveness of immunization with various types of DC-tumor fusion cells. Immunization with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC resulted in 52, 37, 26, and 16% CTL activity, respectively, against MC38/MUC1 tumor cells (Fig. 3A). CTL elicited by WTDC-FC or IKO-FC showed moderate killing against MUC1-positive B16/MUC1 melanoma cells (Fig. 3A). Interestingly, immunization with WTDC-FC or IKO-FC induced 40 and 27% CTL activity, respectively, against MC38 (Fig. 3A), the parent tumor cell of MC38/MUC1, indicating that fusion cells elicited CTL not only against MUC1, but also against unidentified tumor Ag in MC38. In contrast, there were no CTL induced by WTDC-FC against unrelated B16 melanoma cells or MUC1-positive human breast carcinoma cells (Fig. 3A). Similar results were obtained in a separate experiment (Fig. 3B). These results indicate that immunization with DC-tumor fusion cells induces Ag-specific polyclonal
CTL. In addition, CTL activity was almost abolished when mice were immunized with DC-tumor fusion cells deficient in MHC class II molecules.
Suboptimal protection from tumor challenge resulted from lack of MHC class II Ag presentation
To determine the requirement for MHC class I and/or II Ag presentation in antitumor immunity induced by the fusion cells, MUC1.Tg mice were immunized twice with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC or DC, and then challenged with MC38/MUC1 tumor cells on one flank and MC38 tumor cells on the other flank. Immunization with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC or WT-DC resulted in 100, 91.7, 61.5, 15.4, and 0% protection, respectively, against MC38/MUC1 tumor challenge (Fig. 4A). The protection against MC38 was slightly lower. Immunization with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC or WT-DC provided 100, 66.7, 38.5, 7.7, and 0% protection, respectively, against MC38 tumor challenge (Fig. 4B). The findings were consistent with the CTL activity against MC38/MUC1 and MC38 targets (Fig. 4, C and D). The results indicate that both MHC class I and II Ag presentation contributes to antitumor immunity induced by DC-tumor fusion cells. However, antitumor immunity was more compromised when mice were immunized with fusion cells deficient in MHC class II than with fusion cells deficient in MHC class I molecules.
MHC class I and II Ag presentation required for maximal antitumor immunity to eliminate established tumor
To assess whether immunization with fusion cells can eliminate established tumor, MUC1.Tg mice were injected i.v. with MC38/MUC1 tumor cells and then treated with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC or WT-DC on days 2 and 8. Treatment with WTDC-FC or IKO-FC rendered 100 and 90% of mice, respectively, free of pulmonary metastasis (Fig. 5A). In contrast, 20 and 0% mice were free of tumors when treated with IIKO-FC or I/IIKO-FC (Fig. 5A). All mice treated with I/IIKO-FC or DC alone had tumor growth in the lung, although fewer tumor nodules were found in mice treated with I/IIKO-FC than those treated with DC alone. To assess the CTL status of mice immunized with various types of DC-tumor fusion cells, splenocytes from the immunized mice were isolated and the standard 51Cr release assay was performed. CTL activities against MC38/MUC1 and, to a lesser extent, MC38 or B16/MUC1 were observed in mice immunized with WTDC-FC or IKO-FC (Fig. 5B). In contrast, minimal CTL induction occurred in mice immunized with IIKO-FC or I/IIKO-FC, and very little induction in mice immunized with WT-DC alone. These results suggest that immunization with DC-tumor fusion cells deficient in MHC class II Ag presentation impairs the induction of CTL activity and compromises antitumor immunity. Maximal antitumor immunity can be achieved with DC-tumor fusion cells having both MHC class I and II Ag presentation.
In the in vivo study, all of the mice were immunized with 5 x 105 fusion cells. However, the numbers of unfused DC and tumor cells varied due to variation of fusion efficiency. Can these unfused DC and tumor cells contribute to the antitumor immunity? Table I shows the numbers of fused DC/tumor cells and unfused DC or tumor cells in the vaccines. In both experiments, IIKO-FC and I/IIKO-FC contained greater numbers of unfused DC and tumor cells than did WTDC-FC and IKO-FC; yet, vaccination with IIKO-FC and I/IIKO-FC provided less protection against challenge with tumor cells than vaccination with WTDC-FC and IKO-FC. These results indicate that the unfused DC and tumor cells in the vaccine have minimal impact on antitumor immunity in our system.
MHC class II-expressing vaccine provided protection against MHC class I- or II-positive tumor challenge
The previous results indicate the differential antitumor immunity induced by IKO-FC and IIKO-FC. Immunization with IIKO-FC affected antitumor immunity much more acutely than that with IKO-FC. One explanation for the differential antitumor immunity induced by IKO-FC and IIKO-FC is that IKO-FC express MHC class I molecules of tumor origin as demonstrated in FACS analysis (Fig. 1, B and C); thus, the deficiency of MHC class I in DC has been compensated for, at least in part, by MHC class I molecules derived from tumor cells. Therefore, the impact of IKO-FC on induction of CTL and antitumor immunity is not so severe as that of IIKO-FC. To address this concern, we created DC-tumor fusion cells devoid of MHC class I expression.
DC isolated from MHC class I knockout mice (IKO-DC) express MHC class II, but not MHC class I molecules, whereas B16/MUC1 express MUC1, but are negative for MHC molecules (Fig. 6A). Fusion of IKO-DC and B16/MUC1 cells created IKO/B16-FC that expressed MUC1and MHC class II, but not MHC class I, molecules (Fig. 6A). Groups of MUC1.Tg mice were vaccinated twice with IKO/B16-FC. Seven days after the second vaccination, the mice were challenged with MC38/MUC1 tumor cells on the left flank and B16/Ia+/MUC1 tumor cells on the right flank. Mice vaccinated with IKO/B16-FC had 100% protection against challenge by B16/Ia+/MUC1 tumor cells (Fig. 6B). Interestingly, the vaccination also prevented by 78.6% MC38/MUC1 tumor growth (Fig. 6B). CTL induction of splenocytes from the vaccinated mice confirmed these observations (Fig. 6C). The data indicate that MHC class II-expressing vaccine can induce antitumor immunity against both MHC class I- and II-positive tumors.
Discussion
In the present study, we created various DC-tumor fusion cells that deliberately targeted the activation of CD4 and/or CD8 T cells. We have demonstrated: 1) the fusion of DC with tumor cells is a versatile approach to create tumor vaccine; 2) DC-tumor fusion vaccine deficient in MHC class II Ag presentation severely affects the downstream involvement of CD8 and CD4 T cells; and 3) maximal antitumor immune responses require both MHC class I and II Ag presentation, although Ag presentation through MHC class II plays a more important role in the antitumor immunity.
MHC class I and II molecules have a critical function in the selection of CD8 and CD4 T cells in the thymus (26, 27). It has been reported that mice with a disrupted 2m gene express few MHC class I molecules and are virtually devoid of CD8+ T cells (16), and that mice with a disrupted Ab gene lack the expression of class I-A molecules on class II-expressing cells and the development of CD4+ T cells (17). In addition, MHC class I and II molecules are important for Ag processing and presentation, and subsequent activation of CD8 and CD4 T cells, respectively. Splenocytes from MHC-deficient mice were poor stimulators in MLR (18) and in allogeneic CTL generation (16). DC from MHC class I- or II-deficient mice were able to prime only CD8 or CD4 T cells, respectively (28). These results are in line with our findings that lack of MHC class I and/or II molecules in DC or DC-based vaccines compromises the activation of T cells. We show, however, the differential magnitude of antitumor immunity effects with IKO-FC or
IIKO-FC vaccines: whereas immunization with IKO-FC resulted in slightly decreased CTL induction, tumor prevention, and tumor treatment compared with immunization with WTDC-tumor fusion cells, immunization with IIKO-FC abolished IFN- production of T cells, significantly impaired CTL induction, and severely compromised the immunotherapeutic effect of T cells in the prevention and treatment experiments. These results indicate that MHC class II Ag presentation targeting CD4 T cells is essential for successful elimination of tumor challenge or established tumors.
CD8 T cells are the focus of study in antitumor immunity because most nonhemopoietic tumors are positive for MHC class I, but negative for MHC class II, and CD8 CTL are the predominant tumoricidal effector cells. Therefore, development of vaccine has been directed toward activation and amplification of CD8 T cells. However, there is increasing evidence that CD4 T cells play a broader role in antitumor immunity (29). Unlike CD8 T cells, CD4 T cells contribute to antitumor immunity through diverse mechanisms. It is well documented that CD4 T cells provide help to CD8 T cells by activating APC through CD40-CD40L interaction (30, 31, 32) and/or IL-2 production (33). In addition to providing help in the priming phase, CD4 T cells are also needed in the effector phase (34, 35), in which they are required for the maintenance of CTL in vivo and the infiltration of CD8 T cells at the tumor site (36, 37). More important, CD4 T cells have been found to participate in MHC class II-negative tumor destruction through the
MHC-independent pathway. CD4 T cells are implicated in the activation of innate arms of the immune response by recruiting macrophages and/or eosinophils (34). Adoptive transfer of CD4 T cells can control tumor growth (38, 39). Taken together, all of these observations underscore the importance of CD4 T cells in antitumor immunity and strongly argue for the design of tumor vaccine targeting activation of both CD4 and CD8 T cells.
How can MHC class I-expressing but class II-deficient DC-tumor fusion cells impair the induction of CTL and antitumor immunity, while MHC class II-expressing and class I-deficient DC-tumor fusion cells promote CTL induction and antitumor immunity against both MHC class I- and II-positive tumors? The conventional explanation is that the lower CTL induction and antitumor immunity by IIKO-FC are due to lack of help from MHC class II-restricted CD4 T cells in the priming phase, whereas the induction of CTL and antitumor immunity by IKO-FC is through cross-priming (40) by the host DC. One caveat of this interpretation is that cross-priming by host DC should be equally functional in the IIKO-FC vaccination. Theoretically, cross-priming by host DC should be more effective in activation of CD4 T cells through an exogenous pathway. The difficulty in reconciling these results with current knowledge suggests that a novel mechanism of antitumor immunity mediated by CD4 T cells may be involved.
Acknowledgments
We are indebted to Dr. Suzanne Ostrand-Rosenberg from University of Maryland for the gift of B16/Ia+ melanoma cells.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Cancer Institute Grant R01 CA87057; U.S. Department of Defense Breast and Ovarian Cancer Research programs, Grants DAMA170010220 and DAMA170010572; and the Leukemia & Lymphoma Society, Grant 667501.
2 Y.T. and S.K. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Jianlin Gong, Boston University School of Medicine, 650 Albany Street, Room 309, Boston, MA 02118. E-mail address: jgong@bu.edu
4 Abbreviations used in this paper: DC, dendritic cell; 2m, 2-microglobulin; FC, fusion cell; IKO, class I knockout; IIKO, class II knockout; LNC, lymph node cell; Tg, transgenic; WT, wild type.
Received for publication March 19, 2004. Accepted for publication November 12, 2004.
References
Ostrand-Rosenberg, S., A. Thakur, V. Clements. 1990. Rejection of mouse sarcoma cells after transfection of MHC class II genes. J. Immunol. 144:4068.
Townsend, S. E., J. P. Allison. 1993. Tumor rejection after direct costimulation of CD8+ T cells by B7-transfected melanoma cells. Science 259:368.
Baskar, S., L. Glimcher, N. Nabavi, R. T. Jones, S. Ostrand-Rosenberg. 1995. Major histocompatibility complex class II+B7-1+ tumor cells are potent vaccines for stimulating tumor rejection in tumor-bearing mice. J. Exp. Med. 181:619.
Steinman, R. M.. 1991. The dendritic cell system and its role in immunogenicity. Annu. Rev. Immunol. 9:271
Steinman, R. M.. 2001. Dendritic cells and the control of immunity: enhancing the efficiency of antigen presentation. Mt. Sinai J. Med. 68:106.
Young, J. W., K. Inaba. 1996. Dendritic cells as adjuvants for class I major histocompatibility complex-restricted antitumor immunity. J. Exp. Med. 183:7.
Berard, F., P. Blanco, J. Davoust, E. M. Neidhart-Berard, M. Nouri-Shirazi, N. Taquet, D. Rimoldi, J. C. Cerottini, J. Banchereau, A. K. Palucka. 2000. Cross-priming of naive CD8 T cells against melanoma antigens using dendritic cells loaded with killed allogeneic melanoma cells. J. Exp. Med. 192:1535.
Jenne, L., J. F. Arrighi, H. Jonuleit, J. H. Saurat, C. Hauser. 2000. Dendritic cells containing apoptotic melanoma cells prime human CD8+ T cells for efficient tumor cell lysis. Cancer Res. 60:4446.
Hoffmann, T. K., N. Meidenbauer, G. Dworacki, H. Kanaya, T. L. Whiteside. 2000. Generation of tumor-specific T-lymphocytes by cross-priming with human dendritic cells ingesting apoptotic tumor cells. Cancer Res. 60:3542.
Hoffmann, T. K., N. Meidenbauer, J. Muller-Berghaus, W. J. Storkus, T. L. Whiteside. 2001. Proinflammatory cytokines and CD40 ligand enhance cross-presentation and cross-priming capability of human dendritic cells internalizing apoptotic cancer cells. J. Immunother. 24:162.
Gong, J., D. Chen, M. Kashiwaba, D. Kufe. 1997. Induction of antitumor activity by immunization with fusions of dendritic and carcinoma cells. Nat. Med. 3:558.
Phan, V., F. Errington, S. C. Cheong, T. Kottke, M. Gough, S. Altmann, A. Brandenburger, S. Emery, S. Strome, A. Bateman, et al 2003. A new genetic method to generate and isolate small, short-lived but highly potent dendritic cell-tumor cell hybrid vaccines. Nat. Med. 9:1215.
Koido, S., Y. Tanaka, D. Chen, D. Kufe, J. Gong. 2002. The kinetics of in vivo priming of CD4 and CD8 T cells by dendritic/tumor fusion cells in MUC1-transgenic mice. J. Immunol. 168:2111.
Gong, J., D. Avigan, D. Chen, Z. Wu, S. Koido, M. Kashiwaba, D. Kufe. 2000. Activation of antitumor cytotoxic T lymphocytes by fusions of human dendritic cells and breast carcinoma cells. Proc. Natl. Acad. Sci. USA 97:2715.
Gong, J., N. Nikrui, D. Chen, S. Koido, Z. Wu, Y. Tanaka, S. Cannistra, D. Avigan, D. Kufe. 2000. Fusions of human ovarian carcinoma cells with autologous or allogeneic dendritic cells induce antitumor immunity. J. Immunol. 165:1705.
Zijlstra, M., M. Bix, N. E. Simister, J. M. Loring, D. H. Raulet, R. Jaenisch. 1990. 2-microglobulin deficient mice lack CD4–8+ cytolytic T cells. Nature 344:742.
Grusby, M. J., R. S. Johnson, V. E. Papaioannou, L. H. Glimcher. 1991. Depletion of CD4+ T cells in major histocompatibility complex class II-deficient mice. Science 253:1417.
Grusby, M. J., H. Auchincloss, Jr, R. Lee, R. S. Johnson, J. P. Spencer, M. Zijlstra, R. Jaenisch, V. E. Papaioannou, L. H. Glimcher. 1993. Mice lacking major histocompatibility complex class I and class II molecules. Proc. Natl. Acad. Sci. USA 90:3913.
Inaba, K., M. Inaba, N. Romani, H. Aya, M. Deguchi, S. Ikehara, S. Muramatsu, R. M. Steinman. 1992. Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony-stimulating factor. J. Exp. Med. 176:1693.
Siddiqui, J., M. Abe, D. Hayes, E. Shani, E. Yunis, D. Kufe. 1988. Isolation and sequencing of a cDNA coding for the human DF3 breast carcinoma-associated antigen. Proc. Natl. Acad. Sci. USA 85:2320.
Akagi, J., J. W. Hodge, J. P. McLaughlin, L. Gritz, G. Mazzara, D. Kufe, J. Schlom, J. A. Kantor. 1997. Therapeutic antitumor response after immunization with an admixture of recombinant vaccinia viruses expressing a modified MUC1 gene and the murine T-cell costimulatory molecule B7. J. Immunother. 20:38.
Mukherjee, P., A. R. Ginardi, T. L. Tinder, C. J. Sterner, S. J. Gendler. 2001. MUC1-specific cytotoxic T lymphocytes eradicate tumors when adoptively transferred in vivo. Clin. Cancer Res. 7:848s.
Reiner, S. L., S. Zheng, D. B. Corry, R. M. Locksley. 1993. Constructing polycompetitor cDNAs for quantitative PCR. J. Immunol. Methods 165:37.
Rowse, G. J., R. M. Tempero, M. L. VanLith, M. A. Hollingsworth, S. J. Gendler. 1998. Tolerance and immunity to MUC1 in a human MUC1 transgenic murine model. Cancer Res. 58:315.
Wexler, H.. 1966. Accurate identification of experimental pulmonary metastases. J. Natl. Cancer Inst. 36:641.(Yasuhiro Tanaka, Shigeo K)
To dissect the role of Ag presentation through MHC class I and/or II pathways by dendritic cell (DC)-tumor fusion cells, we have created various types of DC-tumor fusion cells by alternating fusion cell partners. Fusions of MC38/MUC1 carcinoma cells with DC from wild-type (WT-DC), MHC class I knockout (IKO-DC), class II knockout (IIKO-DC), or class I and II knockout (I/IIKO-DC) mice created WTDC-fusion cells (FC), IKO-FC, IIKO-FC, and I/IIKO-FC, respectively. MHC class II- and MUC1-positive fusion cells were constructed by fusion of B16/MUC1 melanoma cells with IKO-DC (IKO/B16-FC). Immunization of MUC1 transgenic mice with 5 x 105 WTDC-FC, IKO-FC, IIKO-FC, or I/IIKO-FC provided 100, 91.7, 61.5, and 15.4% protection, respectively, against tumor challenge with MC38/MUC1 cells. In contrast, all mice immunized with irradiated MC38/MUC1 tumor cells or WT-DC developed tumors. One group of mice was immunized with 5 x 105 IKO/B16-FC and then challenged with B16/Ia+/MUC1 on one flank and MC38/MUC1 on the other flank. Immunization of these mice with IKO/B16-FC resulted in 100 and 78.6% protection against B16/Ia+/MUC1 and MC38/MUC1 tumor challenge, respectively. The antitumor immunity induced by immunization with IKO/B16-FC was able to inhibit the growth of MHC class II-negative tumor. In addition, in vivo results correlated with the induction of Ag-specific CTL. Collectively, the data indicate that MHC class II Ag presentation targeting activation of CD4 T cells is indispensable for antitumor immunity.
Introduction
Effective Ag processing and presentation are crucial to antitumor immunity. Tumor cells express tumor Ags, but seldom induce effective antitumor immune response. This observation is probably related to the fact that tumor cells do not express essential costimulatory molecules and seldom express MHC class II molecules. Therefore, effort has been made to enhance tumor cell-based vaccine using genetically modified tumor cells. Immunization of mice with autologous class II- and/or B7-transfected tumor cells provides protection against challenge with wild-type tumor (1, 2) and mediates regression of established tumor (3). Alternatively, a strategy has been developed based on dendritic cells (DC),4 which are bone marrow-derived leukocytes bearing a characteristic veiled morphology that excel in Ag presentation and the initiation of primary immune responses (4, 5, 6). DC reside at the port of entry, take up exogenous Ags, and migrate to draining lymph nodes, where the Ags are presented to CD4 T cells through MHC class II pathways. In addition, DC are capable of initiating CD8 T cell response through a cross-presentation pathway. Exogenous Ags from tumor cells can be delivered to the cytosol, processed, and presented through an endogenous pathway (7, 8, 9, 10).
Another evolving strategy is the use of DC fused to tumor cells (11). In this approach, tumor Ags are delivered to DC, processed, and presented through both MHC class I and II pathways in the context of costimulatory molecules. The fusion cells function like APCs with the ability to migrate to draining lymph nodes, where they interact with CD4 and CD8 T cells and induce potent antitumor immunity (12, 13). Coculture of human peripheral blood monocytes with DC-tumor fusion cells induces both CD4 and CD8 T cells (14, 15). However, the role of MHC class I-restricted or class II-restricted Ag presentation and the activation of CD4 and CD8 T cells in the antitumor immune responses are not well defined. In the present study, we created various types of DC-tumor fusion cells with intact or deficient expression of MHC class I or II molecules by using several kinds of DC and tumor fusion partners. The fusion cells were used in the prevention and treatment of tumors in MUC1 transgenic mice (MUC1.Tg). We observed differential impairment of antitumor immunity induced by fusions of DC from MHC class I and/or II knockout mice. Immunization with MHC class II-deficient DC-tumor fusion cells abolished the IFN- production of CD4 and CD8 T cells and the induction of CTL, and severely impaired antitumor immunity. These results indicate that MHC class II Ag presentation targeting activation of CD4 T cells is indispensable in antitumor immunity.
Materials and Methods
DC generation
DC were obtained from bone marrow cultures of C57BL/6 wild-type (WT), MHC class I knockout (2-microglobulin (2m)–/–) (16), MHC class II knockout (Abb–/–) (17), and MHC class I and II double-knockout(2m–/–/Abb–/–) (18) mice from Taconic Farms, using the method previously described (19). Briefly, bone marrow cells were flushed from long bones and further depleted of lymphocytes, granulocytes, B cells, and APCs by incubation with anti-CD8 (2.43), anti-CD4 (GK1.5), anti B220/CD45R (RA3-3A1/6.1), anti-Ia (B21-2), and anti-Gr-1 (RB6-8C5) (American Type Culture Collection (ATCC)) Abs, followed by rabbit complement. The cells were cultured in RPMI 1640 medium supplemented with 5% heat-inactivated FCS, 50 μM 2-ME, 1 mM HEPES (pH 7.4), 2 mM glutamine, 10 U/ml penicillin, 100 μg/ml streptomycin, and 20 ng/ml murine rGM-CSF (Sigma-Aldrich). On day 5 of culture, DC were purified by multiple steps of plating and collection of the loosely adherent population. After overnight culture, DC were harvested for phenotype analysis and fusion to MC38/MUC1 cells.
Tumor cells
Murine MC38 colon adenocarcinoma and B16 melanoma cells were stably transfected with a MUC1 cDNA, resulting in MC38/MUC1 (20, 21) and B16/MUC1 tumor cells (22). The B16/MUC1 cells positive for MUC1, but negative for both MHC class I and II were selected and fused to B16 positive for MHC class II (B16/Ia+; a kind gift from S. Ostrand-Rosenberg, University of Maryland, Baltimore, MD) to create B16/Ia+/MUC1 tumor cells. The human breast carcinoma cell line, MCF-7, was obtained from ATCC. Cells were maintained in DMEM supplemented with 10% heat-inactivated FCS, 2 mM L-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin.
DC-Tumor fusion process
DCs were fused to MC38/MUC1 carcinoma cells using a previously described method (11). Briefly, DC generated from WT, 2m–/–, Abb–/–, and 2m–/–Abb–/– mice were designated as WT-DC, class I knockout (IKO)-DC, class II knockout (IIKO)-DC, and I/IIKO-DC, respectively, and were mixed with MC38/MUC1 tumor cells at a 10:1 ratio. The fusion process was conducted with 50% polyethylene glycol (Sigma-Aldrich) in prewarmed Dulbecco’s PBS without Ca2+ or Mg2+ at pH 7.4 for 5 min. The polyethylene glycol solution was then diluted by slow addition and mixing of 1, 2, 4, 8, and 16 ml of warm serum-free medium. The cell pellets obtained after centrifuge at 10,000 rpm were resuspended in RPMI 1640 medium supplemented with 10% heat-inactivated FCS, 2 mM glutamine, 10 mM nonessential amino acids, 1 mM sodium pyruvate, 10% NCTC 109, 10 U/ml penicillin, 100 μg/ml streptomycin, and 10 ng/ml murine rGM-CSF. The fused cells were plated in 24-well culture plates for 5–7 days. The fusions of WT-DC, IKO-DC, IIKO-DC, and I/IIKO-DC with MC38/MUC1 created WTDC-fusion cells (FC), IKO-FC, IIKO-FC, and I/IIKO-FC, respectively. The fusion of IKO-DC with B16/MUC1 resulted in IKO/B16-FC. On day 5–7 of culture, the unfused tumor cells grew firmly attached to the tissue culture flask, while the loosely attached DC-tumor fusion cells were dislodged by gentle pipetting. In the in vitro studies, the fusion cells were further purified with cell sorting.
Flow cytometry
MC38/MUC1 tumor cells, WT-DC, IKO-DC, IIKO-DC, and I/IIKO-DC, and WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC were stained with FITC-conjugated HMPV (anti-MUC1; BD Pharmingen) for 30 min on ice. After being washed with PBS, the cells were incubated with PE-conjugated KH95 (anti-MHC class I), M5/114 (anti-MHC class II), and GL1 (anti-B7-2) (BD Pharmingen) for an additional 30 min on ice. The cells were washed, fixed, and analyzed by FACScan (BD Biosciences) with CellQuest analysis software. To determine the fusion efficiency, MUC1 and MHC class II and/or B7-2 were used as tumor or DC marker, respectively.
Mixed leukocyte reaction
WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC were stained with FITC anti-MUC1 and PE anti-B7-2 mAbs and selected by cell sorting (DakoCytomation) with Summit v3.0 analysis software. The fusion cells were exposed to ionizing radiation (30 Gy) and then cocultured with allogeneic (BALB/c) T cells at various ratios (1:20–1:540) in 96-well flat-bottom culture plates for 5 days. The responding T cells ranged from 2 x 105 to 5.4 x 106 in 200 μl/well. WT-DC, IKO-DC, IIKO-DC, and I/IIKO-DC and irradiated MC38/MUC1 and T cells alone were used as controls. Cells were pulsed with 1 μCi of [3H]thymidine (New England Nuclear) per well for 12 h and then collected on filters with a semiautomatic cell harvester. Tritium incorporation was quantified by liquid scintillation. All determinations were conducted in triplicate and expressed as the mean ± SD.
RT-PCR
Lymph node cells (LNC) were isolated from mice immunized twice with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC or treated with PBS, and were sorted into CD4 and CD8 subsets (purity 97–98%). RNA from 1 x 106 sorted CD4 or CD8 T cells was extracted by TRIzol reagent (Invitrogen Life Technologies). Total RNA to cDNA was reverse transcribed using a poly(dT) oligonucleotide and SuperScript (Invitrogen Life Technologies). PCR was performed by amplifying cDNA with the following oligonucleotide primer (23): murine IL-2 (5'-TCCACTTCAAGCTCTACAG-3' and 5'-GAGTCAAATCCAGAACATGCC-3'); IFN- (5'-CATTGAAAGCCTAGAAAGTCTG-3' and 5'-CTCATGGAATGCATCCTTTTTCG-3'); -actin (5'-TGTGATGGTGGGAATGGGTCAG-3' and 5'-TTTGATGTCACGCACGATTTCC-3') (Stratagene). PCR-amplified products were analyzed on a 2% agarose gel.
51Cr cytotoxicity assay
Spleens from MUC1.Tg mice immunized twice with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC were removed, and T cells were isolated into single cell suspensions for use as effector cells. The targets included MC38, MC38/MUC1, B16, B16/MUC1, and MCF-7 tumor cells. Briefly, tumor cells (1–2 x 106 cells) were labeled with 100–200 μCi of Na251CrO4 for 60 min at 37°C, followed by thorough washing to remove unincorporated isotope. T cells and tumor targets were resuspended in culture medium and then combined at various E:T ratios in 96-well V-bottom plates. The plates were centrifuged at 100 x g for 5 min to initiate cell contact and incubated for 5 h at 37°C with 5% CO2. After incubation, supernatants were collected and radioactivity was quantified in a gamma counter. Spontaneous release of 51Cr was determined by incubation of targets in the absence of effectors, and maximum or total release of 51Cr by incubation of targets in 0.1% Triton X-100. The percentage of specific release of 51Cr was calculated using the following: percentage of specific release = ((experimental – spontaneous)/(maximum – spontaneous)) x 100.
In vivo tumor prevention
C57BL/6 MUC1.Tg mice (a kind gift from Dr. S. J. Gendler, Mayo Clinic, Scottsdale, AZ) that express MUC1 at a level similar to that found in humans (24) were used. Seven-week-old MUC1.Tg mice were immunized s.c. on days 0 and 7 with 5 x 105 WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC calculated on the basis of fusion efficiency. WT-DC were used as a control. On day 14, the mice were challenged by s.c. injection in the flank with 5 x 105 syngeneic MC38/MUC1 (left side) and MC38 (right side) tumor cells.
To determine the antitumor immunity induced by MHC class II-expressing, but class I-deficient fusion cells, groups of MUC1.Tg mice were immunized twice s.c. with 5 x 105 IKO/B16-FC. Control mice were immunized with IKO-DC or irradiated tumor cells. Seven days after the second vaccination, the mice were challenged with 5 x 105 MC38/MUC1 tumor cells (MHC class I and MUC1 positive) on the left flank and B16/Ia+/MUC1 tumor cells (MUC1 and MHC class II positive, but class I negative) on the right flank.
All of the mice were followed for 30 days. The size of tumor was determined by measuring perpendicular dimensions with a vernier caliper every 2–3 days. Tumors with a diameter of 3 mm were designated as positive. The mice were maintained in microisolator cages under specific pathogen-free conditions.
Treatment of established pulmonary metastases
Pulmonary metastases were established by i.v. injection of 1 x 106 MC38/MUC1 tumor cells through the tail vein in 7-wk-old MUC1.Tg mice. Two days after the tumor inoculation, mice were immunized with 1 x 106 WTDC-FC, IKO-FC, IIKO-FC, I/IIKO-FC, or WT-DC. The immunization was repeated on day 8. The mice were sacrificed 20 days after the last immunization. Pulmonary metastases were enumerated by counting after staining the lungs with india ink (25).
Statistical analysis
Statistical significance was analyzed using 2 and Student’s t tests.
Results
Characterization of various types of DC-tumor fusion cells
Constitutive deletion of the gene for 2m results in loss of expression of MHC class I in all nucleated cells (16), whereas mutation in the Ab gene abrogates MHC class II expression in all APCs (17). Therefore, DC from the 2m–/–/Abb–/– double-knockout mice are devoid of expression of MHC class I and II molecules (18). To confirm these findings, the phenotype of various types of DC, DC-tumor fusion cells, and tumor cells was assessed by FACS analysis. MC38/MUC1 tumor cells expressed MUC1 and MHC class I molecules (Fig. 1A). DC isolated from WT, 2m–/–, and Abb–/– mice expressed MHC class I and II, class II, and class I molecules (Fig. 1B), respectively. 2m–/–/Abb–/– mice expressed no MHC molecules. However, MHC class I and/or II deficiency did not affect the expression of B7-2 on DC. Fusion of MC38/MUC1 with WT-DC resulted in coexpression of MUC1 with MHC class I and II molecules or B7-2 (Fig. 1C). Similar results were obtained with IKO-DC fused to MC38/MUC1, indicating that tumor-derived MHC class I molecules were expressed. In contrast, IIKO-DC or I/IIKO-DC fused with MC38/MUC1 led to the expression of MUC1 and MHC class I or B7-2 molecules, but not MHC class II molecules (Fig. 1C). These data indicate that the properties of fusion cells are dictated by their parent cells and that DC-tumor fusion cells deficient in MHC class II expression can be created by using DC from Abb–/– or 2m–/–/Abb–/– mice as fusion cell partners.
Stimulation of T cell proliferation compromised in IIKO-FC
To determine the ability of fusion cells to stimulate T cell proliferation, the MLR assay was used. T cells from splenocytes of BALB/c mice cocultured with WT-DC or WTDC-FC proliferated vigorously (Fig. 2A, left upper panel). Coculture of these T cells with IKO-DC or IKO-FC also resulted in proliferation of T cells, albeit at a lower level (Fig. 2A, left lower panel). In contrast, impaired T cell proliferation was observed when T cells were cocultured with IIKO-DC or I/IIKO-DC or their fusion counterparts (Fig. 2A, right upper and lower panels). These data indicate that MHC molecules on DCs or fusion cells are vital for T cell proliferation in MLR.
IFN- abolished in CD4 and CD8 T cells primed by IIKO-FC
Cytokine production is the hallmark of T cell activation. To determine whether cytokine expression was affected in T cells primed by various types of DC-tumor fusion cells, we used RT-PCR to assess the cytokine mRNA levels of LNC isolated 7 days after the second immunization. Whereas sorted CD4 T cells from mice immunized with WTDC-FC or IKO-FC expressed IL-2 and IFN- (Fig. 2B), the expression of IFN- was abolished in CD4 T cells primed by IIKO-FC or I/IIKO-FC. The expression of IL-2 was also abolished in CD4 T cells primed by I/IIKO-FC. Similarly, IFN- was detected in sorted CD8 T cells from mice immunized with WTDC-FC or IKO-FC. However, IFN- was not detected in CD8 T cells primed by IIKO-FC or I/IIKO-FC. These results indicate that MHC class II-restricted Ag presentation affects downstream cytokine production of both CD4 and CD8 T cells. The secretion of IFN- was abolished in T cells primed by MHC class II-deficient fusion cells, indicating the impairment of T cell activation.
Presentation of MHC class II Ag required for effective induction of CTL
CTL response against tumor cells was evaluated by standard 51Cr release assays to assess the effectiveness of immunization with various types of DC-tumor fusion cells. Immunization with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC resulted in 52, 37, 26, and 16% CTL activity, respectively, against MC38/MUC1 tumor cells (Fig. 3A). CTL elicited by WTDC-FC or IKO-FC showed moderate killing against MUC1-positive B16/MUC1 melanoma cells (Fig. 3A). Interestingly, immunization with WTDC-FC or IKO-FC induced 40 and 27% CTL activity, respectively, against MC38 (Fig. 3A), the parent tumor cell of MC38/MUC1, indicating that fusion cells elicited CTL not only against MUC1, but also against unidentified tumor Ag in MC38. In contrast, there were no CTL induced by WTDC-FC against unrelated B16 melanoma cells or MUC1-positive human breast carcinoma cells (Fig. 3A). Similar results were obtained in a separate experiment (Fig. 3B). These results indicate that immunization with DC-tumor fusion cells induces Ag-specific polyclonal
CTL. In addition, CTL activity was almost abolished when mice were immunized with DC-tumor fusion cells deficient in MHC class II molecules.
Suboptimal protection from tumor challenge resulted from lack of MHC class II Ag presentation
To determine the requirement for MHC class I and/or II Ag presentation in antitumor immunity induced by the fusion cells, MUC1.Tg mice were immunized twice with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC or DC, and then challenged with MC38/MUC1 tumor cells on one flank and MC38 tumor cells on the other flank. Immunization with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC or WT-DC resulted in 100, 91.7, 61.5, 15.4, and 0% protection, respectively, against MC38/MUC1 tumor challenge (Fig. 4A). The protection against MC38 was slightly lower. Immunization with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC or WT-DC provided 100, 66.7, 38.5, 7.7, and 0% protection, respectively, against MC38 tumor challenge (Fig. 4B). The findings were consistent with the CTL activity against MC38/MUC1 and MC38 targets (Fig. 4, C and D). The results indicate that both MHC class I and II Ag presentation contributes to antitumor immunity induced by DC-tumor fusion cells. However, antitumor immunity was more compromised when mice were immunized with fusion cells deficient in MHC class II than with fusion cells deficient in MHC class I molecules.
MHC class I and II Ag presentation required for maximal antitumor immunity to eliminate established tumor
To assess whether immunization with fusion cells can eliminate established tumor, MUC1.Tg mice were injected i.v. with MC38/MUC1 tumor cells and then treated with WTDC-FC, IKO-FC, IIKO-FC, and I/IIKO-FC or WT-DC on days 2 and 8. Treatment with WTDC-FC or IKO-FC rendered 100 and 90% of mice, respectively, free of pulmonary metastasis (Fig. 5A). In contrast, 20 and 0% mice were free of tumors when treated with IIKO-FC or I/IIKO-FC (Fig. 5A). All mice treated with I/IIKO-FC or DC alone had tumor growth in the lung, although fewer tumor nodules were found in mice treated with I/IIKO-FC than those treated with DC alone. To assess the CTL status of mice immunized with various types of DC-tumor fusion cells, splenocytes from the immunized mice were isolated and the standard 51Cr release assay was performed. CTL activities against MC38/MUC1 and, to a lesser extent, MC38 or B16/MUC1 were observed in mice immunized with WTDC-FC or IKO-FC (Fig. 5B). In contrast, minimal CTL induction occurred in mice immunized with IIKO-FC or I/IIKO-FC, and very little induction in mice immunized with WT-DC alone. These results suggest that immunization with DC-tumor fusion cells deficient in MHC class II Ag presentation impairs the induction of CTL activity and compromises antitumor immunity. Maximal antitumor immunity can be achieved with DC-tumor fusion cells having both MHC class I and II Ag presentation.
In the in vivo study, all of the mice were immunized with 5 x 105 fusion cells. However, the numbers of unfused DC and tumor cells varied due to variation of fusion efficiency. Can these unfused DC and tumor cells contribute to the antitumor immunity? Table I shows the numbers of fused DC/tumor cells and unfused DC or tumor cells in the vaccines. In both experiments, IIKO-FC and I/IIKO-FC contained greater numbers of unfused DC and tumor cells than did WTDC-FC and IKO-FC; yet, vaccination with IIKO-FC and I/IIKO-FC provided less protection against challenge with tumor cells than vaccination with WTDC-FC and IKO-FC. These results indicate that the unfused DC and tumor cells in the vaccine have minimal impact on antitumor immunity in our system.
MHC class II-expressing vaccine provided protection against MHC class I- or II-positive tumor challenge
The previous results indicate the differential antitumor immunity induced by IKO-FC and IIKO-FC. Immunization with IIKO-FC affected antitumor immunity much more acutely than that with IKO-FC. One explanation for the differential antitumor immunity induced by IKO-FC and IIKO-FC is that IKO-FC express MHC class I molecules of tumor origin as demonstrated in FACS analysis (Fig. 1, B and C); thus, the deficiency of MHC class I in DC has been compensated for, at least in part, by MHC class I molecules derived from tumor cells. Therefore, the impact of IKO-FC on induction of CTL and antitumor immunity is not so severe as that of IIKO-FC. To address this concern, we created DC-tumor fusion cells devoid of MHC class I expression.
DC isolated from MHC class I knockout mice (IKO-DC) express MHC class II, but not MHC class I molecules, whereas B16/MUC1 express MUC1, but are negative for MHC molecules (Fig. 6A). Fusion of IKO-DC and B16/MUC1 cells created IKO/B16-FC that expressed MUC1and MHC class II, but not MHC class I, molecules (Fig. 6A). Groups of MUC1.Tg mice were vaccinated twice with IKO/B16-FC. Seven days after the second vaccination, the mice were challenged with MC38/MUC1 tumor cells on the left flank and B16/Ia+/MUC1 tumor cells on the right flank. Mice vaccinated with IKO/B16-FC had 100% protection against challenge by B16/Ia+/MUC1 tumor cells (Fig. 6B). Interestingly, the vaccination also prevented by 78.6% MC38/MUC1 tumor growth (Fig. 6B). CTL induction of splenocytes from the vaccinated mice confirmed these observations (Fig. 6C). The data indicate that MHC class II-expressing vaccine can induce antitumor immunity against both MHC class I- and II-positive tumors.
Discussion
In the present study, we created various DC-tumor fusion cells that deliberately targeted the activation of CD4 and/or CD8 T cells. We have demonstrated: 1) the fusion of DC with tumor cells is a versatile approach to create tumor vaccine; 2) DC-tumor fusion vaccine deficient in MHC class II Ag presentation severely affects the downstream involvement of CD8 and CD4 T cells; and 3) maximal antitumor immune responses require both MHC class I and II Ag presentation, although Ag presentation through MHC class II plays a more important role in the antitumor immunity.
MHC class I and II molecules have a critical function in the selection of CD8 and CD4 T cells in the thymus (26, 27). It has been reported that mice with a disrupted 2m gene express few MHC class I molecules and are virtually devoid of CD8+ T cells (16), and that mice with a disrupted Ab gene lack the expression of class I-A molecules on class II-expressing cells and the development of CD4+ T cells (17). In addition, MHC class I and II molecules are important for Ag processing and presentation, and subsequent activation of CD8 and CD4 T cells, respectively. Splenocytes from MHC-deficient mice were poor stimulators in MLR (18) and in allogeneic CTL generation (16). DC from MHC class I- or II-deficient mice were able to prime only CD8 or CD4 T cells, respectively (28). These results are in line with our findings that lack of MHC class I and/or II molecules in DC or DC-based vaccines compromises the activation of T cells. We show, however, the differential magnitude of antitumor immunity effects with IKO-FC or
IIKO-FC vaccines: whereas immunization with IKO-FC resulted in slightly decreased CTL induction, tumor prevention, and tumor treatment compared with immunization with WTDC-tumor fusion cells, immunization with IIKO-FC abolished IFN- production of T cells, significantly impaired CTL induction, and severely compromised the immunotherapeutic effect of T cells in the prevention and treatment experiments. These results indicate that MHC class II Ag presentation targeting CD4 T cells is essential for successful elimination of tumor challenge or established tumors.
CD8 T cells are the focus of study in antitumor immunity because most nonhemopoietic tumors are positive for MHC class I, but negative for MHC class II, and CD8 CTL are the predominant tumoricidal effector cells. Therefore, development of vaccine has been directed toward activation and amplification of CD8 T cells. However, there is increasing evidence that CD4 T cells play a broader role in antitumor immunity (29). Unlike CD8 T cells, CD4 T cells contribute to antitumor immunity through diverse mechanisms. It is well documented that CD4 T cells provide help to CD8 T cells by activating APC through CD40-CD40L interaction (30, 31, 32) and/or IL-2 production (33). In addition to providing help in the priming phase, CD4 T cells are also needed in the effector phase (34, 35), in which they are required for the maintenance of CTL in vivo and the infiltration of CD8 T cells at the tumor site (36, 37). More important, CD4 T cells have been found to participate in MHC class II-negative tumor destruction through the
MHC-independent pathway. CD4 T cells are implicated in the activation of innate arms of the immune response by recruiting macrophages and/or eosinophils (34). Adoptive transfer of CD4 T cells can control tumor growth (38, 39). Taken together, all of these observations underscore the importance of CD4 T cells in antitumor immunity and strongly argue for the design of tumor vaccine targeting activation of both CD4 and CD8 T cells.
How can MHC class I-expressing but class II-deficient DC-tumor fusion cells impair the induction of CTL and antitumor immunity, while MHC class II-expressing and class I-deficient DC-tumor fusion cells promote CTL induction and antitumor immunity against both MHC class I- and II-positive tumors? The conventional explanation is that the lower CTL induction and antitumor immunity by IIKO-FC are due to lack of help from MHC class II-restricted CD4 T cells in the priming phase, whereas the induction of CTL and antitumor immunity by IKO-FC is through cross-priming (40) by the host DC. One caveat of this interpretation is that cross-priming by host DC should be equally functional in the IIKO-FC vaccination. Theoretically, cross-priming by host DC should be more effective in activation of CD4 T cells through an exogenous pathway. The difficulty in reconciling these results with current knowledge suggests that a novel mechanism of antitumor immunity mediated by CD4 T cells may be involved.
Acknowledgments
We are indebted to Dr. Suzanne Ostrand-Rosenberg from University of Maryland for the gift of B16/Ia+ melanoma cells.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Cancer Institute Grant R01 CA87057; U.S. Department of Defense Breast and Ovarian Cancer Research programs, Grants DAMA170010220 and DAMA170010572; and the Leukemia & Lymphoma Society, Grant 667501.
2 Y.T. and S.K. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Jianlin Gong, Boston University School of Medicine, 650 Albany Street, Room 309, Boston, MA 02118. E-mail address: jgong@bu.edu
4 Abbreviations used in this paper: DC, dendritic cell; 2m, 2-microglobulin; FC, fusion cell; IKO, class I knockout; IIKO, class II knockout; LNC, lymph node cell; Tg, transgenic; WT, wild type.
Received for publication March 19, 2004. Accepted for publication November 12, 2004.
References
Ostrand-Rosenberg, S., A. Thakur, V. Clements. 1990. Rejection of mouse sarcoma cells after transfection of MHC class II genes. J. Immunol. 144:4068.
Townsend, S. E., J. P. Allison. 1993. Tumor rejection after direct costimulation of CD8+ T cells by B7-transfected melanoma cells. Science 259:368.
Baskar, S., L. Glimcher, N. Nabavi, R. T. Jones, S. Ostrand-Rosenberg. 1995. Major histocompatibility complex class II+B7-1+ tumor cells are potent vaccines for stimulating tumor rejection in tumor-bearing mice. J. Exp. Med. 181:619.
Steinman, R. M.. 1991. The dendritic cell system and its role in immunogenicity. Annu. Rev. Immunol. 9:271
Steinman, R. M.. 2001. Dendritic cells and the control of immunity: enhancing the efficiency of antigen presentation. Mt. Sinai J. Med. 68:106.
Young, J. W., K. Inaba. 1996. Dendritic cells as adjuvants for class I major histocompatibility complex-restricted antitumor immunity. J. Exp. Med. 183:7.
Berard, F., P. Blanco, J. Davoust, E. M. Neidhart-Berard, M. Nouri-Shirazi, N. Taquet, D. Rimoldi, J. C. Cerottini, J. Banchereau, A. K. Palucka. 2000. Cross-priming of naive CD8 T cells against melanoma antigens using dendritic cells loaded with killed allogeneic melanoma cells. J. Exp. Med. 192:1535.
Jenne, L., J. F. Arrighi, H. Jonuleit, J. H. Saurat, C. Hauser. 2000. Dendritic cells containing apoptotic melanoma cells prime human CD8+ T cells for efficient tumor cell lysis. Cancer Res. 60:4446.
Hoffmann, T. K., N. Meidenbauer, G. Dworacki, H. Kanaya, T. L. Whiteside. 2000. Generation of tumor-specific T-lymphocytes by cross-priming with human dendritic cells ingesting apoptotic tumor cells. Cancer Res. 60:3542.
Hoffmann, T. K., N. Meidenbauer, J. Muller-Berghaus, W. J. Storkus, T. L. Whiteside. 2001. Proinflammatory cytokines and CD40 ligand enhance cross-presentation and cross-priming capability of human dendritic cells internalizing apoptotic cancer cells. J. Immunother. 24:162.
Gong, J., D. Chen, M. Kashiwaba, D. Kufe. 1997. Induction of antitumor activity by immunization with fusions of dendritic and carcinoma cells. Nat. Med. 3:558.
Phan, V., F. Errington, S. C. Cheong, T. Kottke, M. Gough, S. Altmann, A. Brandenburger, S. Emery, S. Strome, A. Bateman, et al 2003. A new genetic method to generate and isolate small, short-lived but highly potent dendritic cell-tumor cell hybrid vaccines. Nat. Med. 9:1215.
Koido, S., Y. Tanaka, D. Chen, D. Kufe, J. Gong. 2002. The kinetics of in vivo priming of CD4 and CD8 T cells by dendritic/tumor fusion cells in MUC1-transgenic mice. J. Immunol. 168:2111.
Gong, J., D. Avigan, D. Chen, Z. Wu, S. Koido, M. Kashiwaba, D. Kufe. 2000. Activation of antitumor cytotoxic T lymphocytes by fusions of human dendritic cells and breast carcinoma cells. Proc. Natl. Acad. Sci. USA 97:2715.
Gong, J., N. Nikrui, D. Chen, S. Koido, Z. Wu, Y. Tanaka, S. Cannistra, D. Avigan, D. Kufe. 2000. Fusions of human ovarian carcinoma cells with autologous or allogeneic dendritic cells induce antitumor immunity. J. Immunol. 165:1705.
Zijlstra, M., M. Bix, N. E. Simister, J. M. Loring, D. H. Raulet, R. Jaenisch. 1990. 2-microglobulin deficient mice lack CD4–8+ cytolytic T cells. Nature 344:742.
Grusby, M. J., R. S. Johnson, V. E. Papaioannou, L. H. Glimcher. 1991. Depletion of CD4+ T cells in major histocompatibility complex class II-deficient mice. Science 253:1417.
Grusby, M. J., H. Auchincloss, Jr, R. Lee, R. S. Johnson, J. P. Spencer, M. Zijlstra, R. Jaenisch, V. E. Papaioannou, L. H. Glimcher. 1993. Mice lacking major histocompatibility complex class I and class II molecules. Proc. Natl. Acad. Sci. USA 90:3913.
Inaba, K., M. Inaba, N. Romani, H. Aya, M. Deguchi, S. Ikehara, S. Muramatsu, R. M. Steinman. 1992. Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony-stimulating factor. J. Exp. Med. 176:1693.
Siddiqui, J., M. Abe, D. Hayes, E. Shani, E. Yunis, D. Kufe. 1988. Isolation and sequencing of a cDNA coding for the human DF3 breast carcinoma-associated antigen. Proc. Natl. Acad. Sci. USA 85:2320.
Akagi, J., J. W. Hodge, J. P. McLaughlin, L. Gritz, G. Mazzara, D. Kufe, J. Schlom, J. A. Kantor. 1997. Therapeutic antitumor response after immunization with an admixture of recombinant vaccinia viruses expressing a modified MUC1 gene and the murine T-cell costimulatory molecule B7. J. Immunother. 20:38.
Mukherjee, P., A. R. Ginardi, T. L. Tinder, C. J. Sterner, S. J. Gendler. 2001. MUC1-specific cytotoxic T lymphocytes eradicate tumors when adoptively transferred in vivo. Clin. Cancer Res. 7:848s.
Reiner, S. L., S. Zheng, D. B. Corry, R. M. Locksley. 1993. Constructing polycompetitor cDNAs for quantitative PCR. J. Immunol. Methods 165:37.
Rowse, G. J., R. M. Tempero, M. L. VanLith, M. A. Hollingsworth, S. J. Gendler. 1998. Tolerance and immunity to MUC1 in a human MUC1 transgenic murine model. Cancer Res. 58:315.
Wexler, H.. 1966. Accurate identification of experimental pulmonary metastases. J. Natl. Cancer Inst. 36:641.(Yasuhiro Tanaka, Shigeo K)