Neuroblastomas and many other solid tumors produce high amounts of macrophage migration inhibitory factor (MIF), which appears to play a role in tumor progression. We found that MIF expression in neuroblastoma inhibits T cell proliferation in vitro, raising the possibility that MIF promotes tumorigenesis, in part, by suppressing antitumor immunity. To examine whether tumor-derived MIF leads to suppression of T cell immunity in vivo, we generated MIF-deficient neuroblastoma cell lines using short hairpin small interfering RNAs (siRNA). The MIF knockdown (MIFKD) AGN2a neuroblastoma cells were more effectively rejected in immune-competent mice than control siRNA-transduced or wild-type AGN2a. However, the increased rejection of MIFKD AGN2a was not observed in T cell-depleted mice. MIFKD tumors had increased infiltration of CD8+ and CD4+ T cells, as well as increased numbers of macrophages, dendritic cells, and B cells. Immunization with MIFKD AGN2a cells significantly increased protection against tumor challenge as compared with immunization with wild-type AGN2a, and the increased protection correlated with elevated frequencies of tumor-reactive CD8+ T cells in the lymphoid tissue of treated animals. Increased numbers of infiltrating tumor-reactive CD8+ T cells were also observed at the site of tumor vaccination. In vitro, treatment of AGN2a-derived culture supernatants with neutralizing MIF-specific Ab failed to reverse T cell suppressive activity, suggesting that MIF is not directly responsible for the immune suppression in vivo. This supports a model whereby MIF expression in neuroblastoma initiates a pathway that leads to the suppression of T cell immunity in vivo.

Macrophage migration inhibitory factor (MIF)3 was first described nearly four decades ago as a soluble factor produced by activated T lymphocytes (1, 2, 3). It plays a central role in the control of host inflammatory and immune responses (4). In addition to its potent effects on the immune system, it has also recently been recognized as a protumorigenic factor (5). MIF expression has been found to be elevated in patients with profound immune dysregulation or autoimmune disease, such as in Guillain-Barre syndrome and ulcerative colitis (6, 7), as well as in mice with experimental autoimmune encephalomyelitis (8). Several reports have linked MIF to fundamental processes controlling cell proliferation, cell survival, angiogenesis, and tumor progression (9, 10, 11, 12). Over-expression of MIF has been found in several tumor types examined (13, 14, 15, 16, 17, 18, 19). MIF has been shown to promote malignant cell transformation, inhibit tumor cell-specific immune cytolytic responses, and strongly enhance neovascularization (20).

Neuroblastoma is the second most common pediatric solid cancer and is responsible for ∼15% of all childhood cancer deaths (21). In analyzing the immune response to neuroblastoma, we found that MIF was over-expressed in both mouse- and patient-derived cell lines (22). MIF is able to regulate the expression of genes related to tumor cell proliferation, migration, and anti-apoptosis (23, 24). Inhibition of MIF expression by anti-sense DNA results in tumor growth inhibition (25). This data suggests that MIF may play an important role in the tumorgenesis of neuroblastoma. However, whether MIF is capable of influencing immunity to neuroblastoma is not known.

Our previous studies have shown that MIF produced by a murine neuroblastoma cell line (AGN2a) inhibits cytokine-, CD3-, and allo-induced T cell activation (26). In this study, transfection of a heterologous human cell line with plasmid-based gene expression vectors encoding cDNA for mouse MIF also resulted in the production of MIF-containing culture supernatants that inhibited T cell proliferation. Finally, we showed that the T cell inhibitory effects were reversed when MIF expression was transiently knocked down by dicer-generated small interfering RNA. Abe et al. (27) reported that splenocyte cultures treated with neutralizing anti-MIF mAb displayed a significant increase in CTL activity. The increased CTL activity was associated with enhanced expression of the common γc-chain of the IL-2R that promoted CD8+ T cell survival. These data suggest that MIF plays an important role in the regulation of antitumor T lymphocytes in vivo, and that it might exert protumorigenic effects by regulating T lymphocyte responses to tumors.

Tumor cell-based vaccines have been tested in several studies as an immunotherapeutic approach to treating cancer (28). These approaches have included engineering tumor cells to express immune costimulatory molecules or cytokines, and the fusion of tumor cells with dendritic cells (29). Our own work has demonstrated that when the murine neuroblastoma AGN2a (an aggressive sublcone of Neuro-2a) was engineered to express the immune costimulatory molecules CD80 and CD137L (AGN2a-CD80/137L), an effective cell-based vaccine was produced that induced strong T cell responses (30). Since MIF over-expressed by tumor cells was found to inhibit T cell expansion (25), we hypothesized that inhibiting production of MIF from AGN2a might enhance T cell immunity and aid in the generation of antitumor immunity.

In the present study, short hairpin RNA (shRNA) constructs were permanently transduced into the AGN2a cell line, effectively knocking down MIF expression. Effects of inhibiting MIF production on tumor cell growth in vivo were observed. MIFKD AGN2a and MIFKD CD80/CD137L-expressing AGN2a, and their MIF intact parental lines, were tested in vaccination protocols to determine the impact of MIFKD on host survival, tumor growth, and T cell immunity.

A/J mice were purchased from The Jackson Laboratory. The mice were housed in the Medical College of Wisconsin Biomedical Resource Center (Milwaukee, WI), and the Medical College of Wisconsin Animal Care and Use Committee approved experiments involving animals.

An aggressive variant of the mouse neuroblastoma cell line Neuro-2a (N2a), designated AGN2a, was derived through sequential in vitro and in vivo passage in our laboratory (31). The tumor cells express MHC class I Ags, but are MHC class II-negative.

MIFKD cell lines were generated by using the BLOCK-iT Lentiviral RNAi Expression System from Invitrogen. In breif, three pairs of complementary oligonucleotides encoding MIF shRNA target sequences (19–21 nucleotides in length) were designed and synthesized according to specified guidelines of the Invitrogen system. The three top and bottom strand oligo sequences are as follows (the MIF sense and anti-sense target sequences are underlined):

A1 – 5′-CACCAATAGTTGATGTAGACCCGGTCCGAAGACCGGG TCTACATCAACTA-3′,

A2 – 5′-AAAATAGTTGATGTAGACCCGGTCTTCGGACCGGGTC TACATCAACTATT-3′;

B1 – 5′-ACCGTAATAGTTGATGTAGACCCGGCGAACCGGGTCT ACATCAACTATTA-3′,

B2 – 5′-AAAATAATAGTTGATGTAGACCCGGTTCGCCGGGTCT ACATCAACTATTAC-3′;

C1 – 5′-CACCGGGTCTACATCAACTATTACGCGAACGTAATAG TTGATGTAGACCC-3′,

C2 – 5′-AAAAGGGTCTACATCAACTATTACGTTCGCGTAATAG TTGATGTAGACCC-3′.

A1 and A2, B1 and B2, and C1 and C2 were annealed to create double-stranded oligonucleotides A, B, and C, respectively. The double-stranded oligonucleotides were cloned into the pENTR/U6 entry vector. Competent Escherichia coli were transformed with the entry vector constructs and selected to obtain entry vector clones. The entry clones were screened for MIF inhibitory activity through transient transfection of AGN2a cells, and then the U6 RNA cassettes were cloned into the pLenti6/BLOCK-it-DEST expression vector. Lentiviral stocks of each MIF shRNA construct were generated in 293FT cells, concentrated by centrifugation in a 30,000 mw cut off filter unit (Millipore), and tittered on NIH3T3 cells (which gives ∼10-fold lower titer than the Invitrogen recommended HT1080 cell line). AGN2a cells were then transduced with MIF shRNA-A, -B (multiplicity of infection 0.07), and -C (multiplicity of infection 0.02) lentiviruses, stably transduced cells selected for by culturing in the presence of blasticidin, MIF levels determined, and then secondary transductions conducted with the alternate vectors. The transduced cells were screened in western blot, ELISA, and real-time PCR assays to determine the levels of MIF expression. Transduction with the shRNA-B lentivirus did not significantly affect MIF expression levels (see Fig. 1). Therefore, cells transduced with this lentiviral construct were used as a control in some experiments. Established lines that were dually transduced with shRNA-A and -C sequences showed profound MIF decreases and are referred to as MIFKD AGN2a.

FIGURE 1.

Generation of MIFKD AGN2a cells using shRNA. MIFKD AGN2a cells (MIFKD-AGN2a) and AGN2a cells transduced with a nonfunctional shRNA construct (control) were analyzed for MIF mRNA expression by real-time PCR, and the results are expressed as (A) fold-decrease in expression vs expression in wild-type AGN2a cells (see Materials and Methods for details). Culture supernatants from wild-type (AGN2a), MIFKD (MIFKD-AGN2a), and the control-transduced AGN2a cells were analyzed for MIF protein content in western blot assays (B) and ELISA (C). The data are from one of three replicate experiments. ∗∗, p < 0.01.

FIGURE 1.

Generation of MIFKD AGN2a cells using shRNA. MIFKD AGN2a cells (MIFKD-AGN2a) and AGN2a cells transduced with a nonfunctional shRNA construct (control) were analyzed for MIF mRNA expression by real-time PCR, and the results are expressed as (A) fold-decrease in expression vs expression in wild-type AGN2a cells (see Materials and Methods for details). Culture supernatants from wild-type (AGN2a), MIFKD (MIFKD-AGN2a), and the control-transduced AGN2a cells were analyzed for MIF protein content in western blot assays (B) and ELISA (C). The data are from one of three replicate experiments. ∗∗, p < 0.01.

Close modal

AGN2a expressing the immune costimulatory molecules CD80 and CD137L (designated as AGN2a-80/137L) were generated by transfecting MIFKD or wild-type AGN2a cells with plasmid vectors encoding each molecule (30). Permanent transfectants were obtained by culturing the cells with geneticin (1000 μg/ml) and hygromycin (500 μg/ml). Cloned cell lines were then derived by limiting dilution.

The following mAbs, with or without a fluorescent label, were obtained from BD Biosciences: anti-CD3 (clone 145-2C11), anti-CD4 (clones GK1.5 and RM4-5), anti-CD8 (clone 53-6.7), anti-CD45.2 (clone 104), anti-4-1BBL (CD137L, clone TKS-1), anti-CD80 (clone 16-10A1), anti-Ly6G (clone RB6-8C5), anti-I-Ak (clone 10-3.6), anti-CD11c (clone HL3), anti-B220 (clone RA3-6B2), anti-FcγRII/III (clone 2.4G2), and anti-rat IgG2a (clone RG7/1.30). Anti-F4/80 was obtained from Serotec. Anti-CD8-conjugated microbeads used for immunomagnetic cell sorting were purchased from Miltenyi Biotec.

Ascites fluid containing anti-MIF mAb (clone NIHIII.D9; mouse IgG1) was kindly provided by Richard Bucala (Yale University, New Haven, CT). This Ab has been shown to neutralize MIF activity both in vitro and in vivo (32, 33). The Ab was purified using an ImmunoPure IgG (Protein A) Purification Kit from Pierce. Purity of the Ab was confirmed by PAGE.

Normal mice or T cell-depleted mice were injected s.c. with 104, 5 × 104, or 106 wild-type, control shRNA-transduced, or MIFKD AGN2a cells. T cell-depleted mice were generated by injecting the animals i.p. with anti-Thy1.2 mAb (500 μg) 2 days before tumor cell injection, and every 4 days thereafter until the mice died from tumor progression or until day 60 after tumor challenge. The peripheral blood of anti-Thy1.2-treated mice was analyzed by flow cytometry, and the results indicated that >95% of CD8+ T cells and >80% of CD4+ T cells were depleted. The length (L) and width (W) of tumors were measured with calipers every 2 days, and the tumor volume (TV) was calculated as: TV = L × W. Mice were considered moribund and euthanized when tumor volume exceeded 250 mm2.

A/J mice were vaccinated twice weekly, by injecting 2 × 106 irradiated (5000 rad) tumor cells s.c. Seven days after the second vaccination, the mice were challenged s.c. with 104, 5 × 104, or 106 viable AGN2a cells. Tumor size and survival was monitored every 2 days. In some experiments, the spleens and draining lymph nodes (dLNs) were harvested 5 days after secondary vaccination to isolate CD8+ T cells for testing in IFN-γ ELISPOT assays.

To examine immune cell infiltration at the site of tumor vaccination, tumor cells were inoculated s.c. in a mixed collagen matrix, GFR (growth factor reduced) Matrigel (BD Biosciences), as previously described (34). Matrigel is fluid at 4°C, and gels at body temperature. In brief, irradiated (5000 rad) wild-type or MIFKD AGN2a cells were suspended in ice-cold Matrigel (3.33 × 106 cells per ml) and kept on ice until in vivo inoculation. Recipient A/J mice were anesthetized with ketamine and injected s.c. in their hind flanks with 300 μl (106 tumor cells) of the Matrigel/tumor cell suspension. Five days after inoculation, the Matrigel plugs were removed by a wide excision of the flank wall. The Matrigel plugs were treated with collagenase (1 mg/ml) (Sigma-Aldrich) and DNase (0.3 mg/ml) (Sigma-Aldrich) at 37°C for 60 min to release the cells. The cells were counted and analyzed by flow cytometry with several different cell surface markers, or the cells were subjected to immunomagnetic sorting to obtain purified CD8+ T cells for testing in IFN-γ ELISPOT assays.

In some experiments, mice were first vaccinated s.c. with 2 × 106 irradiated (4000 rad) AGN2a-80/137L tumor cells and then, 7 days later, inoculated with 300 μl of Matrigel containing 105 viable MIF+ (wild-type) or MIFKD tumor cells (AGN2a or AGN2a-80/137L). Five days after inoculation, the Matrigel plugs were collected and digested with collagenase to obtain single-cell suspensions. Spleens were also collected from these same mice and processed separately. The cells were then stained with combinations of fluorescently labeled reagents including anti-CD3, anti-CD4, anti-CD8, anti-CD45, and annexin V (BD Biosciences). Propidium iodide (PI) was added to each sample, and the stained cells were analyzed by flow cytometry.

Cells were first incubated with 10% heat-inactivated normal rat serum in PBS and 100 μg/ml anti-FcγRII/III mAb (clone 2.4G2) to block the nonspecific binding of specific mAbs. The cells were washed, stained with specific fluorescently labeled mAbs, and PI was added to each sample to exclude dead cells from the analysis. The stained cells were run through a Becton Dickinson FACSCalibur flow cytometer, and the resulting data were analyzed using Flow-Jo software (Tree Star).

AGN2a and MIFKD AGN2a cells were cultured overnight in 4-well chamber slides. The adhered cells were washed in Tris buffer/Tween 20 (0.5 M Tris Base, 9% NaCl, 0.5% Tween 20 (pH 7.6)), fixed in 4% paraformaldehyde, permeabilized with 0.05% Triton X, and blocked with 10% normal goat serum. Cells were stained overnight at 4°C with rabbit polyclonal anti-MIF at 1 μg/ml (clone ab7207; Abcam), then incubated with goat-anti-rabbit Alexa fluor 568 (Invitrogen) diluted 1/1000 for 1 h at room temperature, followed by staining with 300 nM 4′,6-diamidino-2-phenylindole for 5 min. Cells were visualized on a Zeiss Axio Imager Z1 and imaged with a Nuance Multispectral Imaging System, model N-MSI-420-FL (Cambridge Research and Instrumentation).

To assess numbers of tumor-reactive IFN-γ-secreting CD8+ T cells, ELISPOT assays were done using mouse IFN-γ ELISPOT kits from BD Biosciences. Cells derived from Matrigel plugs, dLNs, and spleens were incubated with anti-CD8-conjugated microbeads (Miltenyi Biotec). CD8 T cells were positively selected using a Miltenyi automated immunomagnetic sorter (autoMACS). The level of T cell enrichment was determined by flow cytometric analysis, and the purity of CD8+ T cells from Matrigel plugs was >85%, from dLNs >98%, and from the spleen >90%. Various numbers of purified CD8+ T cells (105, 5 × 104, and 2.5 × 104 per well) were cocultured with wild-type AGN2a stimulator cells (5 × 104/well) for 18 h at 37°C. The plates were developed according to the manufacturer’s instructions, and the number of resulting spots counted with an ImmunoSpot Analyzer using included acquisition and analysis software (CTL Analyzers).

Total cell lysates or cell culture supernatants from wild-type, control shRNA-transduced, and MIFKD AGN2a cell lines were homogenized in reducing loading buffer (Invitrogen), the proteins resolved on SDS polyacrylamide gels, and the resolved proteins transferred to polyvinylidene difluoride membranes (Immuno-Blot polyvinylidene difluoride, 0.2 mm; Bio-Rad) using a NuPAGE Bis-Tris electrophoresis system (Invitrogen). The blots were probed with anti-mouse MIF polyclonal Ab (Abcam) at a 1/2500 dilution, followed by incubation with alkaline phosphatase-conjugated secondary Ab (Abcam) at a 1/1000 dilution. CDP star chemiluminescent reagent (PerkinElmer Life Science) was used for alkaline phosphatase detection.

Total RNA was extracted from wild-type, control shRNA-transduced, and MIFKD cells using RNAeasy kits (Qiagen) followed by DNase I treatment (Qiagen). cDNA was then synthesized using TaqMan reverse transcription reagents. MIF and β-actin mRNA expression were measured by real-time PCR in duplicate tubes using standard SYBR-green methods. Primers and probes for β-actin were purchased from Applied Biosystems. A comparative cycle threshold (CT) was used to determine MIF and β-actin mRNA expression relative to nontemplate controls. CT values were normalized for each sample using the formulas: ΔCT = CT(MIF) − CT(actin), and ΔΔCT = ΔCT(MIFKD or Control AGN2a) − ΔCT(wild-type AGN2a). Relative MIF expression levels in the MIFKD and control cells were calculated as 2−ΔΔCT, and results are expressed as fold-decrease in MIF mRNA expression vs expression in wild-type AGN2a cells.

Spleens were harvested from A/J mice and CD90+ T cells were isolated by positive selection using anti-Thy1.2-conjugated immunomagnetic beads (Miltenyi Biotec). T cells were plated at a density of 25,000 per well in 96-well plates. Culture supernatants, derived from AGN2a cells plated at a density of 3 × 106 cells in a 10-cm plate for 72 h, were serially diluted and preincubated with neutralizing anti-MIF Ab or mouse IgG1 (Serotec) control Ab (100 μg/ml) for 2 h at room temperature before adding them to the 96-well plates containing T cells. The T cells were activated by adding 75,000 anti-CD3/CD28-conjugated Dynal beads (Invitrogen) to each well. After 72 h of culture, 1 μCi of [3H]thymidine was added to each well and incubated for 4 h. The plates were then harvested to determine thymidine incorporation using a Skatron Micro 96-well harvester and a Trilux (PerkinElmer) 1450 microbeta scintillation counter.

The Student’s t test was used to compare MIF concentrations, cell numbers, and IFN-γ ELISPOT data. Survival curves were compared by log-rank analysis. Values of p < 0.05 were considered as significant.

To obtain MIF-deficient (or MIFKD) AGN2a cell lines, tumor cells were transduced with MIF shRNA lentiviral constructs. After drug selection and single-cell cloning, the resulting clones were screened for MIF production. One of the shRNA lentiviral constructs (construct B) did not significantly impact MIF expression, so cloned cells transduced with this construct were used in experiments as “control” cells. Cotransduction of AGN2a cells with constructs A and C (see Materials and Methods) produced cloned cells that were severely deficient in MIF production. Expression of MIF in wild-type, MIFKD, and transduced control AGN2a cells is shown in Fig. 1. Real-time PCR results showed that MIF mRNA levels in the MIFKD cells were decreased 16.7-fold as compared with parental AGN2a cells (Fig. 1,A), while MIF mRNA levels in AGN2a cells transduced with a control shRNA construct were nearly the same as wild-type AGN2a (Fig. 1,A; Control). When culture supernatants (from 105 cells cultured for 72 h) were analyzed for MIF protein content by western blot (Fig. 1,B) and ELISA (Fig. 1,C), MIF was nearly nondetectable in supernatants collected from MIFKD cells. The loss of MIF in the knockdown cells was also confirmed by immunofluorescent staining (Fig. 2).

FIGURE 2.

Immunofluorescent analysis of MIFKD AGN2a cells. AGN2a (A) and MIFKD AGN2a (B) were incubated with anti-MIF Ab followed by secondary Alexa fluor 568-conjugated Ab. C, AGN2a cells stained with IgG control Ab. The results are from one of two replicate experiments.

FIGURE 2.

Immunofluorescent analysis of MIFKD AGN2a cells. AGN2a (A) and MIFKD AGN2a (B) were incubated with anti-MIF Ab followed by secondary Alexa fluor 568-conjugated Ab. C, AGN2a cells stained with IgG control Ab. The results are from one of two replicate experiments.

Close modal

Since it has been suggested that tumor-derived MIF promotes tumor growth in vivo (19, 35, 36, 37, 38, 39), we examined whether decreased MIF production in the MIFKD cell lines altered their growth characteristics. First, cell growth rates were measured in vitro (Fig. 3). MIFKD AGN2a cells expanded at a slower rate than the parental wild-type or control shRNA-transduced AGN2a cells (Fig. 3).

FIGURE 3.

In vitro growth rate of MIFKD AGN2a cells. A total of 104 AGN2a cells (wild-type, control, or MIFKD) were seeded in culture. At the indicated times, the cultures were assessed for absolute numbers of viable cells. The data are from one of three replicate experiments. ∗∗, p < 0.01.

FIGURE 3.

In vitro growth rate of MIFKD AGN2a cells. A total of 104 AGN2a cells (wild-type, control, or MIFKD) were seeded in culture. At the indicated times, the cultures were assessed for absolute numbers of viable cells. The data are from one of three replicate experiments. ∗∗, p < 0.01.

Close modal

In the next series of experiments, we compared the in vivo growth patterns of MIFKD, wild-type, and control shRNA-transduced AGN2a cells in syngeneic A/J mice. Different doses of AGN2a cells (104, 5 × 104, and 106) were injected s.c., and the mice were followed for tumor growth and survival. Mice were considered moribund and euthanized when tumor volumes exceeded 250 mm2. As shown in Fig. 4, survival rates of the mice injected with MIFKD AGN2a cells were significantly higher at all cell inoculation doses than mice injected with wild-type or control shRNA AGN2a cells. Individual tumor growth rates were also monitored, and MIFKD AGN2a tumors appeared to grow slower than wild-type or control shRNA-transduced AGN2a tumors in mice inoculated with the high (106) and intermediate (5 × 104) cell doses, but tumor growth rates at the lower cell inoculation dose (104) were not noticeably different between the different cell types (data not shown).

FIGURE 4.

Survival of mice inoculated with wild-type, control-transduced, or MIFKD AGN2a cells. A/J mice were inoculated s.c. with 104 (A), 5 × 104 (B), or 106 (C) wild-type AGN2a (AGN2a), control small interfering RNA (control AGN2a), or MIFKD AGN2a cells and followed for survival. Mice were considered moribund and euthanized when tumor size exceeded 250 mm2. The data are the combined results of two to three experiments (n = 10–20 total mice per group).

FIGURE 4.

Survival of mice inoculated with wild-type, control-transduced, or MIFKD AGN2a cells. A/J mice were inoculated s.c. with 104 (A), 5 × 104 (B), or 106 (C) wild-type AGN2a (AGN2a), control small interfering RNA (control AGN2a), or MIFKD AGN2a cells and followed for survival. Mice were considered moribund and euthanized when tumor size exceeded 250 mm2. The data are the combined results of two to three experiments (n = 10–20 total mice per group).

Close modal

Since our laboratory had previously documented that that tumor-derived MIF was capable of inhibiting T cell activation in vitro (26) and we had shown that T cells were important for protective immunity to AGN2a in vivo (40), we hypothesized that the increased rejection of MIFKD AGN2a cells observed in Fig. 4 was due to increased T cell immunity. To address this hypothesis, mice were inoculated with MIFKD or wild-type AGN2a cells and the tumor cell-inoculated A/J mice were treated with a mAb (anti-Thy1.2) to deplete T cells in vivo. Treatment of mice with this Ab (500 μg i.p. 2 days before tumor cell injection, and every 4 days thereafter) resulted in the depletion of >95% of CD8+ and >80% of CD4+ splenic T cells (data not shown). The Ab-treated mice were injected with the same doses of tumor cells as in Fig. 3 (104, 5 × 105, and 106). Survival of T cell-depleted mice inoculated with MIFKD AGN2a tumor cells was not significantly different from mice inoculated with wild-type AGN2a tumor cells (Fig. 5), and the growth of MIFKD AGN2a tumors in T cell-depleted mice was also similar to the growth of wild-type AGN2a tumors (data not shown). These results confirmed our hypothesis that the increased survival of mice inoculated with MIFKD tumor cells in Fig. 3 was due to increased T cell immunity.

FIGURE 5.

Survival of T cell-depleted mice inoculated with wild-type or MIFKD AGN2a cells. The T cell-depleted mice were inoculated s.c. with 104 (A), 5 × 104 (B), or 106 (C) AGN2a or MIFKD AGN2a cells and followed for survival. Mice were considered moribund and euthanized when tumor size exceeded 250 mm2. The data are the combined results of two experiments in A (n = 10 mice/group), and from one experiment in B and C (n = 5 mice/group).

FIGURE 5.

Survival of T cell-depleted mice inoculated with wild-type or MIFKD AGN2a cells. The T cell-depleted mice were inoculated s.c. with 104 (A), 5 × 104 (B), or 106 (C) AGN2a or MIFKD AGN2a cells and followed for survival. Mice were considered moribund and euthanized when tumor size exceeded 250 mm2. The data are the combined results of two experiments in A (n = 10 mice/group), and from one experiment in B and C (n = 5 mice/group).

Close modal

To further investigate the impact of MIF-deficient neuroblastoma on antitumor immunity, mice were vaccinated with irradiated MIFKD or wild-type AGN2a cells twice weekly and then, 7 days later, the vaccinated mice were challenged with 5 × 104 or 105 live wild-type AGN2a cells (Fig. 6,A). Mice vaccinated with MIFKD AGN2a cells had significantly increased resistance to tumor challenge at both tumor cell doses as compared with mice vaccinated with wild-type AGN2a cells (Fig. 6, B and C). Our laboratory had previously documented that frequencies of IFN-γ-producing, tumor-reactive CD8+ T cells in the lymphoid tissues of vaccinated mice correlate with the magnitude of antitumor reactivity against AGN2a (41). When CD8+ cells were isolated from peripheral lymphoid tissues (spleens and dLNs) of vaccinated mice and tested in IFN-γ ELISPOT assays for tumor reactivity using wild-type AGN2a stimulators, significantly increased frequencies of AGN2a-reactive CD8+ cells were detected in mice vaccinated with MIFKD AGN2a cells (Fig. 6 D). Thus, the increased frequency of AGN2a-reactive CD8+ cells in MIFKD-vaccinated mice correlated with their ability to more effectively resist tumor challenge.

FIGURE 6.

Immunization of mice with MIFKD AGN2a cells increased the survival of tumor-challenged mice and increased CD8-tumor immunity. As depicted in the experimental design (A), A/J mice were given two weekly s.c. immunizations (vaccine) with irradiated wild-type or MIFKD AGN2a tumor cells. One week after the second immunization, the mice were challenged s.c. with 5 × 104 (B) or 105 (C) viable AGN2a cells. Mice were considered moribund and euthanized when tumor size exceeded 250 mm2. Five days after the second immunization, some mice (n = 4 mice/group) were euthanized and dLNs and spleens harvested, cells from each tissue pooled, and CD8+ cells isolated by immunomagnetic sorting. The CD8+ cells were tested for reactivity against AGN2a cells in IFN-γ ELISPOT assays (D). The survival curves represent are the combined results of two experiments (n = 10 mice/group), and the ELISPOT results are from one of three replicate experiments. ∗∗, p < 0.01.

FIGURE 6.

Immunization of mice with MIFKD AGN2a cells increased the survival of tumor-challenged mice and increased CD8-tumor immunity. As depicted in the experimental design (A), A/J mice were given two weekly s.c. immunizations (vaccine) with irradiated wild-type or MIFKD AGN2a tumor cells. One week after the second immunization, the mice were challenged s.c. with 5 × 104 (B) or 105 (C) viable AGN2a cells. Mice were considered moribund and euthanized when tumor size exceeded 250 mm2. Five days after the second immunization, some mice (n = 4 mice/group) were euthanized and dLNs and spleens harvested, cells from each tissue pooled, and CD8+ cells isolated by immunomagnetic sorting. The CD8+ cells were tested for reactivity against AGN2a cells in IFN-γ ELISPOT assays (D). The survival curves represent are the combined results of two experiments (n = 10 mice/group), and the ELISPOT results are from one of three replicate experiments. ∗∗, p < 0.01.

Close modal

As a way to monitor immune cell infiltration at the vaccine site, we used a mixed collagen matrix, Matrigel, which we had used in previous studies (34). Irradiated MIFKD or wild-type AGN2a cells were suspended in cold Matrigel and injected s.c. into normal A/J mice. The resulting solidified plugs were harvested 5 days after inoculation and infiltrating cells obtained by digesting the plugs with collagenase and DNase. The cells were counted and analyzed by flow cytometry, and infiltrating immune cells were distinguished from any residual tumor cells by excluding CD45-negative cells from the analysis. Significantly increased numbers of CD8+ and CD4+ T cells, as well as other immune cells including macrophages, granulocytes and dendritic cells, were recovered from the Matrigel plugs containing MIFKD AGN2a cells as compared with plugs containing wild-type AGN2a cells (Fig. 7,A). In a second set of experiments, CD8+ T cells were isolated from the Matrigel plugs by immunomagnetic sorting and the cells tested in IFN-γ ELISPOT assays for AGN2a reactivity. As shown in Fig. 7 B, the CD8-enriched cells from plugs containing the MIFKD tumor cells contained significantly higher frequencies of AGN2a-reactive cells as compared with CD8+ cells from wild-type AGN2a plugs.

FIGURE 7.

Immune cells captured at the site of tumor cell inoculation through the use of a collagen matrix (Matrigel). A/J mice were inoculated s.c. with irradiated wild-type, MIFKD, CD80+CD13L+ (AGN2a-80/137L), or MIFKD AGN2a-80/137L cells suspended in Matrigel. Five days after inoculation, the Matrigel plugs were harvested, processed, and the cells counted. A and C, AGN2a (A) and AGN2a-80/137L (C) immune cell numbers were assessed by flow cytometry using CD45 to distinguish infiltrating immune cells from residual tumor cells. B and D, AGN2a (B) and AGN2a-80/137L (D) CD8+ cells were isolated by immunomagnetic sorting and tested for reactivity against AGN2a cells in IFN-γ ELISPOT assays. The results in A and C are the combined data of two independent experiments, and the data in B and D are from one of two replicate experiments. ∗∗, p < 0.01.

FIGURE 7.

Immune cells captured at the site of tumor cell inoculation through the use of a collagen matrix (Matrigel). A/J mice were inoculated s.c. with irradiated wild-type, MIFKD, CD80+CD13L+ (AGN2a-80/137L), or MIFKD AGN2a-80/137L cells suspended in Matrigel. Five days after inoculation, the Matrigel plugs were harvested, processed, and the cells counted. A and C, AGN2a (A) and AGN2a-80/137L (C) immune cell numbers were assessed by flow cytometry using CD45 to distinguish infiltrating immune cells from residual tumor cells. B and D, AGN2a (B) and AGN2a-80/137L (D) CD8+ cells were isolated by immunomagnetic sorting and tested for reactivity against AGN2a cells in IFN-γ ELISPOT assays. The results in A and C are the combined data of two independent experiments, and the data in B and D are from one of two replicate experiments. ∗∗, p < 0.01.

Close modal

In Fig. 7, C and D, we tested the impact of MIF deficiency on Matrigel infiltration using MIFKD or parental AGN2a cells that had been further modified to express the immune costimulatory molecules CD80 and CD137L. This approach was taken to observe the impact of MIF deficiency on the immunogenicity of genetically engineered neuroblastoma cells that had previously been shown to induce potent antitumor immunity (30). Expression of CD80 and CD137L on the tumor cells (designated as AGN2a-80/137L) induced both higher numbers of Matrigel-infiltrating immune cells (Fig. 7,C) and higher frequencies of tumor-reactive CD8+ T cells (Fig. 7 D) than tumor cells without the costimulatory molecules. These data provide additional evidence that inhibiting production of MIF from AGN2a cells results in increased antitumor T cell immunity.

Based on our previous data showing that tumor-derived MIF induced T cell apoptosis in vitro (26), we hypothesized that T cells infiltrating MIFKD tumors would be of increased viability when compared with T cells responding to wild-type tumor. To address this hypothesis, mice were vaccinated with irradiated AGN2a-80/137L cells and then, 7 days later, the prevaccinated mice were inoculated with live tumor cells (wild-type MIF+ or MIFKD AGN2a cells) suspended in Matrigel. The prevaccination step was done to ensure robust T cell infiltration of the Matrigel/tumor cell plugs, and the rationale for inoculating mice with viable tumor cells in Matrigel was to allow for ongoing MIF production. The use of Matrigel also allowed us to easily harvest early tumors and the immune cells responding to them. Five days after inoculation of the Matrigel/tumor cell suspensions into prevaccinated mice, the solidified Matrigel plugs and spleens were harvested and processed into single-cell suspensions. The tumor-infiltrating cells and spleen cells were stained with FITC-labeled annexin V, PI, and phenotypic cell surface markers, and the stained cells analyzed by flow cytometry. As shown in Fig. 8, PI-negative CD3+ (A), CD8+ (B), and CD4+ cells (C) from the MIFKD tumors bound less annexin V than cells isolated from MIF+ tumors. In contrast, splenic T cells isolated from these same mice bound comparable amounts of annexin V (see dotted histograms in Fig. 8). These results support our hypothesis that MIF produced in the local tumor microenvironment promotes the death of infiltrating T cells in vivo, directly correlating to our results seen with in vitro activated T cells.

FIGURE 8.

Influence of tumor-derived MIF on annexin V binding to tumor-infiltrating T cells. A/J mice were s.c. vaccinated with irradiated CD80+CD13L+ AGN2a (AGN2a-80/137L) cells. Seven days later, the mice were s.c. inoculated with 300 μl of Matrigel containing viable AGN2a or MIFKD AGN2a cells (top histograms in A–C), or viable AGN2a-80/137L or MIFKD AGN2a-80/137L cells (bottom histograms in A–C). Five days after tumor cell/Matrigel inoculation, the Matrigel plugs (i.e., tumors) and spleens were harvested and processed to obtain single-cell suspensions. The tumor-infiltrating cells and splenocytes were stained with combinations of cells surface markers, annexin V and PI, and the cells were analyzed by flow cytometry. Histograms depicting annexin V staining of PI-negative cells are shown for gated CD3+ (A), CD8+ (B), and CD4+ (C) cells. Each histogram is from the pooled cells of three mice, and the results are representative of three independent experiments.

FIGURE 8.

Influence of tumor-derived MIF on annexin V binding to tumor-infiltrating T cells. A/J mice were s.c. vaccinated with irradiated CD80+CD13L+ AGN2a (AGN2a-80/137L) cells. Seven days later, the mice were s.c. inoculated with 300 μl of Matrigel containing viable AGN2a or MIFKD AGN2a cells (top histograms in A–C), or viable AGN2a-80/137L or MIFKD AGN2a-80/137L cells (bottom histograms in A–C). Five days after tumor cell/Matrigel inoculation, the Matrigel plugs (i.e., tumors) and spleens were harvested and processed to obtain single-cell suspensions. The tumor-infiltrating cells and splenocytes were stained with combinations of cells surface markers, annexin V and PI, and the cells were analyzed by flow cytometry. Histograms depicting annexin V staining of PI-negative cells are shown for gated CD3+ (A), CD8+ (B), and CD4+ (C) cells. Each histogram is from the pooled cells of three mice, and the results are representative of three independent experiments.

Close modal

Although our data indicated that MIF played an important role in suppression of T cell immunity by AGN2a cells in vivo, it was unclear whether MIF was directly or indirectly responsible for the suppression. To address this, we turned to an in vitro assay where we had shown that MIF-containing tumor cell culture supernatants suppressed T cell proliferation (26). A neutralizing MIF-specific mAb (kindly provided by Richard Bucala, Yale University, New Haven, CT) was purified and tested in T cell activation assays to determine whether the Ab could reverse suppression of T cell proliferation mediated by MIF-containing AGN2a culture supernatants. The culture supernatants potently suppressed thymidine incorporation of T cells activated with anti-CD3/CD28-coated beads, and nearly complete suppression was observed even at a 107 final dilution of culture supernatant (Fig. 9). Anti-MIF (100 μg/ml) had no effect on the culture supernatant-mediated suppression, indicating that MIF was not directly responsible for the T cell suppression. Similar results were obtained using culture supernatants from U2OS cells engineered to over-express mouse MIF (data not shown).

FIGURE 9.

Neutralizing MIF-specific Ab failed to reverse suppressed T cell proliferation induced by AGN2a culture supernatants. Purified A/J T cells were plated at a density of 25,000 cells per well and activated by addition of anti-CD3/28-conjugated beads (75,000/well). AGN2a cell culture supernatants were diluted with media and added to each well at the indicated final dilutions. For supernatants treated with MIF-specific Ab or mouse IgG1 control Ab, supernatants were preincubated with Ab at 100 μg/ml for 2 h at room temperature and then added to the T cells. After 72 h, the cultures were pulsed for 4 h with [3H]thymidine. The data is shown as mean cpm ± SD of triplicate wells, and the results are representative of three separate experiments.

FIGURE 9.

Neutralizing MIF-specific Ab failed to reverse suppressed T cell proliferation induced by AGN2a culture supernatants. Purified A/J T cells were plated at a density of 25,000 cells per well and activated by addition of anti-CD3/28-conjugated beads (75,000/well). AGN2a cell culture supernatants were diluted with media and added to each well at the indicated final dilutions. For supernatants treated with MIF-specific Ab or mouse IgG1 control Ab, supernatants were preincubated with Ab at 100 μg/ml for 2 h at room temperature and then added to the T cells. After 72 h, the cultures were pulsed for 4 h with [3H]thymidine. The data is shown as mean cpm ± SD of triplicate wells, and the results are representative of three separate experiments.

Close modal

AGN2a expresses low levels of class I MHC that can be increased by treatment with IFN-γ. Nevertheless, despite the presence of this key immune recognition protein, an immune response capable of controlling AGN2a tumor growth in vivo does not occur. The generation of tumor-specific CTL requires not only the appropriate processing and presentation of tumor Ags by MHC molecules, but also T lymphocytes expressing TCRs of appropriate specificity for tumor Ags and activation of the appropriate immune costimulatory molecules. Moreover, these T cells must encounter Ag in a microenvironment where they are free from tolerogenic signals. Once the CTL response is initiated, there also must be vigorous and sustained activation so as to achieve successful tumor regression (42). Secreted cytokines, chemokines, and growth factors all play pivotal roles in regulating tumor pathology by either promoting or inhibiting T cell activation (43). These secreted molecules can directly, or indirectly, play a central role in uncontrolled tumor cell division, proangiogenic stimulation, or suppression of tumor cell immune surveillance (44). The focus of our studies, macrophage MIF, is that this cytokine regulates the immune response at numerous levels. It appears that MIF represents a unique cytokine superfamily (45). Over-expression of MIF has been associated with a wide variety of tumor cell types including breast (14), prostate (17), lung (15), colon (46), liver (13), glioblastoma (47), and also neuroblastoma (22). MIF has also been shown to promote malignant cell transformation, inhibit tumor cell-specific immune cytolytic responses, NK responses, and strongly enhance neovascularization (48, 49, 50).

The molecular and cellular mechanisms involved in MIF bioactivity remain an active area of study. With regard to cancer biology, effects on both the tumor and the immune system must be considered. In macrophages, MIF protects against apoptosis by inactivating p53-dependent pathways (51). These studies were based on biological screens conducted by the Beach laboratory wherein MIF was isolated as a factor able to overcome p53 activity in biological assays and suppress its activity as a transcriptional activator (52). These studies indicate that MIF may provide a direct link between inflammatory and protumorigenic environments. This was directly demonstrated by Fingerle-Rowson et al. (9), who showed that carcinogen-induced fibrosarcomas are smaller in size and have a lower mitotic index in MIFKD mice. MIF also has been described to modulate other cell-cycle control pathways, modulating JNK and p27Kip1 expression levels through association with Jab1 and direct activation of MAPK, ERK, and phospholipase A2 (11, 53).

The immunobiology of MIF is equally complex. Although MIF was discovered as a cytokine secreted by activated T lymphocytes, the precise role of MIF in T cell responses has remained undefined (35, 54). In addition to MIF-dependent modulation of macrophage transcription, activation, and viability, T lymphocytes have also been shown to be targets of MIF regulation. Bacher and Bucula (55) reported that MIF plays an important regulatory role in the activation of T cells. Both mitogen- and Ag-induced TH2 lymphocyte activation appear to depend on autocrine production of MIF. Mitogen- or Ag-activated T cells express significant quantities of MIF mRNA and protein, and neutralization of MIF inhibits IL-2 production and T cell proliferation in vitro and decreases the TH cell response to soluble Ag in vivo (55). It has also been shown that neutralizing anti-MIF Abs inhibit T cell proliferation and IL-2 production in vitro, and suppress Ag-driven T cell activation and Ab production in vivo (56). Interestingly, Abe et al. (27) described an inhibitory function for MIF in TH1-dependent CTL responses. They found that splenocyte cultures treated with neutralizing anti-MIF mAb showed a significant increase in the CTL response. This effect was accompanied by elevated production of IFN-γ. Histological examination of EG.7 tumors from anti-MIF Ab-treated animals showed a prominent increase in both CD4+ and CD8+ T cells as well as apoptotic tumor cells, consistent with the observed augmentation of CTL activity in vivo by anti-MIF. This increased CTL activity was associated with enhanced expression of the common γ-chain of the IL-2R that mediates CD8+ T cell survival. Because antitumor immunity is thought to rely heavily on TH1-associated cytolytic activity (57), over-expression of MIF in a tumor microenvironment may provide a selective growth and protective advantage for developing malignancies.

Previously described roles for MIF in T cell activation suggest that tumor-derived MIF might exert protumorigenic effects by regulating antitumor T lymphocyte responses. In the present study, we examined the effect of MIF on immune responses in vivo. Knockdown of MIF expression in AGN2a-inoculated mice enhanced their survival, unless the mice were depleted of T cells. The reduction of MIF may enhance CD8-mediated recognition and, thus, enhance the rejection of AGN2a. We also found that the tumor growth rate in vitro was reduced in MIFKD AGN2a, consistent with observations made by Ren et al. (25). They found that reduction of MIF inhibited neuroblastoma cell growth, and that this inhibitory effect correlated with decreased N-Myc, IL-8, c-Met, and TrkB expression, and increased expression of tumor suppressor genes (BLU, VISNL-1) and other neuroblastoma transcripts (EPHB6) (25). These data indicate that in tumor-bearing animals, MIF plays a dual role by promoting tumor cell growth and inhibiting T cell responses.

Our laboratory previously showed that culture supernatants from the MIF-producing murine neuroblastoma cell line AGN2a or from a human ostosarcoma (U2OS) cell line engineered to over-express mouse MIF inhibited cytokine-, CD3-, and allo-induced T cell proliferation (26). Based on the inhibitory effect of the MIF-containing culture supernatants on T cell immunity in vitro, we hypothesized that MIFKD in AGN2a should increase the CD8-mediated response to this tumor in vivo. This was demonstrated by the increased survival rates in MIFKD AGN2a-inoculated mice (Fig. 4) and MIFKD AGN2a-vaccinated/challenged mice (Fig. 6, A and B). Furthermore, the number of IFN-γ-secreting CD8+ T cells was also significantly increased in the lymphoid tissues of mice vaccinated with MIFKD cells (Fig. 7,D). One might argue that modification of AGN2a cells with the MIF shRNA lentiviral construct somehow increased tumor immunogenicity, perhaps through induction of a neoantigen, and that increased immunogenicity rather than MIF deficiency was responsible for increased rejection of the modified cells (Fig. 4). However, the fact that vaccination with the MIFKD AGN2a cells resulted in increased immunity to the unmodified wild-type tumor cells (Fig. 6), which would not contain the putative neoantigen, argues against this. We contend that the body of evidence presented in this study makes a strong case that expression of MIF by AGN2a leads to the suppression of T cell immunity.

We recently described the ability to measure immune responses to cell-based vaccines at the s.c. site of immunization (34). Based on the results shown in Fig. 6, we hypothesized that MIFKD of AGN2a cells would increase the CD8-mediated antitumor response at the vaccine site, and that MIFKD of a more immunogenic AGN2a (modified to express CD80 and CD137L) would further increase this response. When AGN2a embedded in Matrigel was used to capture immune-reactive cells in vivo, IFN-γ ELISPOT assays of infiltrating CD8+ T cells confirmed our hypothesis (Fig. 7, B and D). Both CD4+ and CD8+ T cell numbers have been previously observed in the MIF-deficient microenvironment, and both cell types have been described as being involved in mediating tumor destruction (58). Since tumor Ag-specific IFN-γ-producing cells have been shown to be highly effective in mediating antitumor immunity, even when Ag density is low on the target cells (59), the enhancement of IFN-γ-producing CD8+ cells through the inhibition of MIF may be an attractive strategy for increasing the efficacy of immunotherapy for neuroblastoma and other MIF-producing tumors. Finally, the results in Fig. 7 showing that inhibition of MIF increased the accumulation of monocytes/macrophages, granulocytes, and dendritic cells at the vaccine site indicate that MIF expression also plays a role in modulating the trafficking of innate immune cells as well as T cells. We do not yet know why increased numbers of these innate immune cells infiltrate the vaccine site containing MIFKD AGN2a or what impact MIF has on the viability or function of these cells. However, it is interesting to speculate that tumor-derived MIF impacts both innate and adaptive tumor immunity. This will be examined in future studies.

The mechanism by which MIF regulates T cell immunity is still unclear. Our laboratory previously showed that culture supernatants from the MIF-producing murine neuroblastoma cell line AGN2a or from a human ostosarcoma (U2OS) cell line engineered to over-express mouse MIF inhibited cytokine-, CD3-, and allo-induced T cell activation (26). Although the data clearly showed that expression of MIF was required for the suppressive activity in the culture supernatant, it was unknown whether the suppression was directly due to MIF itself or another factor secreted by the tumor cells in response to MIF. In the current study, using neutralizing MIF-specific mAb we show that MIF is not directly responsible for suppressive activity in the tumor cell culture supernatants (Fig. 9). These in vitro data suggest that MIF acts in vivo by programming the tumor cells to secrete a factor, or factors, that mediate the potent immune suppressive activities associated with MIF expression. We are currently investigating the nature of the suppressive factor(s).

The authors have no financial conflict of interest.

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 the Midwest Athletes Against Childhood Cancer Fund and U.S. Public Health Service Grant CA100030.

3

Abbreviations used in this paper: MIF, migration inhibitory factor; shRNA, short hairpin RNA; MIFKD, MIF knockdown; dLN, draining lymph node; PI, propidium iodide; CT, cycle threshold.

1
George, M., J. H. Vaughn.
1962
. In vitro cell migration as a model for delayed sensitivity.
Proc. Soc. Exp. Biol. Med.
111
:
514
-521.
2
Bloom, B. R., B. Bennett.
1966
. Mechanism of a reaction in vitro associated with delayed-type hypersensitivity.
Science
153
:
80
-82.
3
David, J. R..
1966
. Delayed hypersensitivity in vitro: its mediation by cell-free substances formed by lymphoid cell-antigen interaction.
Proc. Natl. Acad. Sci. USA
56
:
72
-77.
4
Lue, H., R. Kleemann, T. Calandra, T. Roger, J. Bernhagen.
2002
. Macrophage migration inhibitory factor (MIF): mechanisms of action and role in disease.
Microbes Infect.
4
:
449
-460.
5
Mitchell, R. A., R. Bucala.
2000
. Tumor growth-promoting properties of macrophage migration inhibitory factor (MIF).
Semin. Cancer Biol.
10
:
359
-366.
6
De Jong, Y. P., A. C. Abadia-Molina, A. R. Satoskar, K. Clarke, S. T. Rietdijk, W. A. Faubion, E. Mizoguchi, C. N. Metz, M. Alsahli, T. ten Hove, et al
2001
. Development of chronic colitis is dependent on the cytokine MIF.
Nat. Immunol.
2
:
1061
-1066.
7
Nicoletti, F., A. Creange, D. Orlikowski, F. Bolgert, K. Mangano, C. Metz, R. Di Marco, Y. Al Abed.
2005
. Macrophage migration inhibitory factor (MIF) seems crucially involved in Guillain-Barre syndrome and experimental allergic neuritis.
J. Neuroimmunol.
168
:
168
-174.
8
Powell, N. D., T. L. Papenfuss, M. A. McClain, I. E. Gienapp, T. M. Shawler, A. R. Satoskar, C. C. Whitacre.
2005
. Cutting edge: macrophage migration inhibitory factor is necessary for progression of experimental autoimmune encephalomyelitis.
J. Immunol.
175
:
5611
-5614.
9
Fingerle-Rowson, G., O. Petrenko, C. N. Metz, T. G. Forsthuber, R. Mitchell, R. Huss, U. Moll, W. Muller, R. Bucala.
2003
. The p53-dependent effects of macrophage migration inhibitory factor revealed by gene targeting.
Proc. Natl. Acad. Sci. USA
100
:
9354
-9359.
10
Liao, H., R. Bucala, R. A. Mitchell.
2003
. Adhesion-dependent signaling by macrophage migration inhibitory factor (MIF).
J. Biol. Chem.
278
:
76
-81.
11
Mitchell, R. A., C. N. Metz, T. Peng, R. Bucala.
1999
. Sustained mitogen-activated protein kinase (MAPK) and cytoplasmic phospholipase A2 activation by macrophage migration inhibitory factor (MIF): regulatory role in cell proliferation and glucocorticoid action.
J. Biol. Chem.
274
:
18100
-18106.
12
Petrenko, O., G. Fingerle-Rowson, T. Peng, R. A. Mitchell, C. N. Metz.
2003
. Macrophage migration inhibitory factor deficiency is associated with altered cell growth and reduced susceptibility to Ras-mediated transformation.
J. Biol. Chem.
278
:
11078
-11085.
13
Akbar, S. M., M. Abe, H. Murakami, K. Tanimoto, T. Kumagi, Y. Yamashita, K. Michitaka, N. Horiike, M. Onji.
2001
. Macrophage migration inhibitory factor in hepatocellular carcinoma and liver cirrhosis; relevance to pathogenesis.
Cancer Lett.
171
:
125
-132.
14
Bini, L., B. Magi, B. Marzocchi, F. Arcuri, S. Tripodi, M. Cintorino, J. C. Sanchez, S. Frutiger, G. Hughes, V. Pallini, et al
1997
. Protein expression profiles in human breast ductal carcinoma and histologically normal tissue.
Electrophoresis
18
:
2832
-2841.
15
Kamimura, A., M. Kamachi, J. Nishihira, S. Ogura, H. Isobe, H. Dosaka-Akuta, A. Ogata, M. Shindoh, T. Ohbuchi, Y. Kawakami.
2000
. Intracellular distribution of macrophage migration inhibitory factor predicts the prognosis of patients with adenocarcinoma of the lung.
Cancer
89
:
334
-341.
16
Li, Z., Y. Ren, Q. C. Wu, S. X. Lin, Y. J. Liang, H. Z. Liang.
2004
. Macrophage migration inhibitory factor enhances neoplastic cell invasion by inducing the expression of matrix metalloproteinase 9 and interleukin-8 in nasopharyngeal carcinoma cell lines.
Chin. Med. J.
117
:
107
-114.
17
Meyer-Siegler, K., P. B. Hudson.
1996
. Enhanced expression of macrophage migration inhibitory factor in prostatic adenocarcinoma metastases.
Urology
48
:
448
-452.
18
Meyer-Siegler, K. L., E. C. Leifheit, P. L. Vera.
2004
. Inhibition of macrophage migration inhibitory factor decreases proliferation and cytokine expression in bladder cancer cells.
BMC Cancer
4
:
34
19
Shimizu, T., R. Abe, H. Nakamura, A. Ohkawara, M. Suzuki, J. Nishihira.
1999
. High expression of macrophage migration inhibitory factor in human melanoma cells and its role in tumor cell growth and angiogenesis.
264
:
751
-758.
20
Bucala, R., S. C. Donnelly.
2007
. Macrophage migration inhibitory factor: a probable link between inflammation and cancer.
Immunity
26
:
281
-285.
21
Maris, J. M., K. K. Matthay.
1999
. Molecular biology of neuroblastoma.
J. Clin. Oncol.
17
:
2264
-2279.
22
Bin, Q., B. D. Johnson, D. W. Schauer, J. T. Casper, R. J. Orentas.
2002
. Production of macrophage migration inhibitory factor by human and murine neuroblastoma.
Tumour Biol.
23
:
123
-129.
23
Fan, J., Y. Chen, H. M. Chan, P. K. Tam, Y. Ren.
2005
. Removing intensity effects and identifying significant genes for Affymetrix arrays in macrophage migration inhibitory factor-suppressed neuroblastoma cells.
Proc. Natl. Acad. Sci. USA
102
:
17751
-17756.
24
Ren, Y., H. M. Chan, Z. Li, C. Lin, J. Nicholls, C. F. Chen, P. Y. Lee, V. Lui, M. Bacher, P. K. Tam.
2004
. Upregulation of macrophage migration inhibitory factor contributes to induced N-Myc expression by the activation of ERK signaling pathway and increased expression of interleukin-8 and VEGF in neuroblastoma.
Oncogene
23
:
4146
-4154.
25
Ren, Y., H. M. Chan, J. Fan, Y. Xie, Y. X. Chen, W. Li, G. P. Jiang, Q. Liu, A. Meinhardt, P. K. Tam.
2006
. Inhibition of tumor growth and metastasis in vitro and in vivo by targeting macrophage migration inhibitory factor in human neuroblastoma.
Oncogene
25
:
3501
-3508.
26
Yan, X., R. J. Orentas, B. D. Johnson.
2006
. Tumor-derived macrophage migration inhibitory factor (MIF) inhibits T lymphocyte activation.
Cytokine
33
:
188
-198.
27
Abe, R., T. Peng, J. Sailors, R. Bucala, C. N. Metz.
2001
. Regulation of the CTL response by macrophage migration inhibitory factor.
J. Immunol.
166
:
747
-753.
28
Copier, J., A. Dalgleish.
2006
. Overview of tumor cell-based vaccines.
Int. Rev. Immunol.
25
:
297
-319.
29
Emens, L. A..
2006
. Roadmap to a better therapeutic tumor vaccine.
Int. Rev. Immunol.
25
:
415
-443.
30
Yan, X., B. D. Johnson, R. J. Orentas.
2004
. Murine CD8 lymphocyte expansion in vitro by artificial antigen-presenting cells expressing CD137L (4-1BBL) is superior to CD28, and CD137L expressed on neuroblastoma expands CD8 tumour-reactive effector cells in vivo.
Immunology
112
:
105
-116.
31
Johnson, B. D., X. Yan, D. W. Schauer, R. J. Orentas.
2003
. Dual expression of CD80 and CD86 produces a tumor vaccine superior to single expression of either molecule.
Cell Immunol.
222
:
15
-26.
32
Leech, M., C. Metz, P. Hall, P. Hutchinson, K. Gianis, M. Smith, H. Weedon, S. R. Holdsworth, R. Bucala, E. F. Morand.
1999
. Macrophage migration inhibitory factor in rheumatoid arthritis: evidence of proinflammatory function and regulation by glucocorticoids.
Arthritis Rheum.
42
:
1601
-1608.
33
Leech, M., C. Metz, L. Santos, T. Peng, S. R. Holdsworth, R. Bucala, E. F. Morand.
1998
. Involvement of macrophage migration inhibitory factor in the evolution of rat adjuvant arthritis.
Arthritis Rheum.
41
:
910
-917.
34
Zhou, Q., B. D. Johnson, R. J. Orentas.
2007
. Cellular immune response to an engineered cell-based tumor vaccine at the vaccination site.
Cell Immunol.
245
:
91
-102.
35
Chesney, J., C. Metz, M. Bacher, T. Peng, A. Meinhardt, R. Bucala.
1999
. An essential role for macrophage migration inhibitory factor (MIF) in angiogenesis and the growth of a murine lymphoma.
Mol. Med.
5
:
181
-191.
36
Nishihira, J., T. Ishibashi, T. Fukushima, B. Sun, Y. Sato, S. Todo.
2003
. Macrophage migration inhibitory factor (MIF): its potential role in tumor growth and tumor-associated angiogenesis.
Ann. NY Acad. Sci.
995
:
171
-182.
37
Ogawa, H., J. Nishihara, Y. Sato, M. Kondo, N. Takahashi, T. Oshima, S. Todo.
2000
. An antibody for macrophage migration inhibitory factor suppresses tumor growth and inhibits tumor-associated angiogenesis.
Cytokine
12
:
309
-314.
38
Sasaki, Y., K. Kasuya, J. Nishihira, Y. Magami, A. Tsuchida, T. Aoki, Y. Koyanagi.
2002
. Suppression of tumor growth through introduction of an antisense plasmid of macrophage migration inhibitory factor.
Int. J. Mol. Med.
10
:
579
-583.
39
Takahashi, N., J. Nishihira, Y. Sato, M. Kondo, H. Ogawa, T. Ohshima, Y. Une, S. Todo.
1998
. Involvement of macrophage migration inhibitory factor (MIF) in the mechanism of tumor cell growth.
Mol. Med.
4
:
707
-714.
40
Johnson, B. D., W. Jing, R. J. Orentas.
2007
. CD25+ regulatory T cell inhibition enhances vaccine-induced immunity to neuroblastoma.
J. Immunother.
30
:
203
-214.
41
Johnson, B. D., J. A. Gershan, N. Natalia, H. Zujewski, J. J. Weber, X. Yan, R. J. Orentas.
2005
. Neuroblastoma cells transiently transfected to simultaneously express the costimulatory molecules CD54, CD80, CD86, and CD137L generate antitumor immunity in mice.
J. Immunother.
28
:
449
-460.
42
Kiessling, R., K. Wasserman, S. Horiguchi, K. Kono, J. Sjoberg, P. Pisa, M. Petersson.
1999
. Tumor-induced immune dysfunction.
Cancer Immunol Immunother.
48
:
353
-362.
43
Lazar-Molnar, E., H. Hegyesi, S. Toth, A. Falus.
2000
. Autocrine and paracrine regulation by cytokines and growth factors in melanoma.
Cytokine
12
:
547
-554.
44
Slingluff, C. L..
1999
. Targeting unique tumor antigens and modulating the cytokine environment may improve immunotherapy for tumors with immune escape mechanisms.
Cancer Immunol Immunother.
48
:
371
-373.
45
Sun, H. W., J. Bernhagen, R. Bucala, E. Lolis.
1996
. Crystal structure at 2.6-A resolution of human macrophage migration inhibitory factor.
Proc. Natl. Acad. Sci. USA
93
:
5191
-5196.
46
Wilson, J. M., P. L. Coletta, R. J. Cuthbert, N. Scott, K. MacLennan, G. Hawcroft, L. Leng, J. B. Lubetsky, K. K. Jin, E. Lolis, et al
2005
. Macrophage migration inhibitory factor promotes intestinal tumorigenesis.
Gastroenterology
129
:
1485
-1503.
47
Bacher, M., J. Schrader, N. Thompson, K. Kuschela, D. Gemsa, G. Waeber, J. Schlegel.
2003
. Up-regulation of macrophage migration inhibitory factor gene and protein expression in glial tumor cells during hypoxic and hypoglycemic stress indicates a critical role for angiogenesis in glioblastoma multiforme.
Am. J. Pathol.
162
:
11
-17.
48
Lolis, E., R. Bucala.
2003
. Macrophage migration inhibitory factor.
Expert Opin. Ther. Targets
7
:
153
-164.
49
Apte, R. S., D. Sinha, E. Mayhew, G. J. Wistow, J. Y. Niederkorn.
1998
. Role of macrophage migration inhibitory factor in inhibiting NK cell activity and preserving immune privilege.
J. Immunol.
160
:
5693
-5696.
50
Repp, A. C., E. S. Mayhew, S. Apte, J. Y. Niederkorn.
2000
. Human uveal melanoma cells produce macrophage migration-inhibitory factor to prevent lysis by NK cells.
J. Immunol.
165
:
710
-715.
51
Mitchell, R. A., H. Liao, J. Chesney, G. Fingerle-Rowson, J. Baugh, J. David, R. Bucala.
2002
. Macrophage migration inhibitory factor (MIF) sustains macrophage proinflammatory function by inhibiting p53: regulatory role in the innate immune response.
Proc. Natl. Acad. Sci. USA
99
:
345
-350.
52
Hudson, J. D., M. A. Shoaibi, R. Maestro, A. Carnero, G. J. Hannon, D. H. Beach.
1999
. A proinflammatory cytokine inhibits p53 tumor suppressor activity.
J. of Experimental Medicine
190
:
1375
-1382.
53
Kleemann, R., A. Hausser, G. Geiger, R. Mischke, A. Burger-Kentischer, O. Flieger, F.-J. Johannes, T. Roger, T. Calandra, A. Kapurniotu, et al
2000
. Intracellular action of the cytokine MIF to modulate AP-1 activity and the cell cycle through Jab1.
Nature
408
:
211
-216.
54
White, E. S., S. R. Strom, N. L. Wys, D. A. Arenberg.
2001
. Non-small cell lung cancer cells induce monocytes to increase expression of angiogenic activity.
J. Immunol.
166
:
7549
-7555.
55
Bacher, M., C. N. Metz, T. Calandra, K. Mayer, J. Chesney, M. Lohoff, D. Gemsa, T. Donnelly, R. Bucala.
1996
. An essential regulatory role for macrophage migration inhibitory factor in T-cell activation.
Proc. Natl. Acad. Sci. USA
93
:
7849
-7854.
56
Calandra, T., L. A. Spiegel, C. N. Metz, R. Bucala.
1998
. Macrophage migration inhibitory factor is a critical mediator of the activation of immune cells by exotoxins of Gram-positive bacteria.
Proc. Natl. Acad. Sci. USA
95
:
11383
-11388.
57
Zhou, Y., M. L. Bosch, M. L. Salgaller.
2002
. Current methods for loading dendritic cells with tumor antigen for the induction of antitumor immunity.
J. Immunother.
25
:
289
-303.
58
Shiku, H..
2003
. Importance of CD4+ helper T-cells in antitumor immunity.
Int. J. Hematol.
77
:
435
-438.
59
Matsumura, M., D. H. Fremont, P. A. Peterson, I. A. Wilson.
1992
. Emerging principles for the recognition of peptide antigens by MHC class I molecules.
Science
257
:
927
-934.