CD8+ T Cell–Independent Tumor Regression Induced by Fc-OX40L and Therapeutic Vaccination in a Mouse Model of Glioma

Katherine A. Murphy, Jami R. Erickson, Charles S. Johnson, Charles E. Seiler, Jessica Bedi, Peisheng Hu, G. Elizabeth Pluhar, Alan L. Epstein and John R. Ohlfest
J Immunol January 1, 2014, 192 (1) 224-233; DOI: https://doi.org/10.4049/jimmunol.1301633

Abstract

Despite the growing number of preclinical and clinical trials focused on immunotherapy for the treatment of malignant gliomas, the prognosis for this disease remains grim. Although some promising advances have been made, the immune response stimulated as a result of immunotherapeutic protocols has been inefficient at complete tumor elimination, primarily due to our lack of understanding of the necessary effector functions of the immune system. We previously demonstrated that a tumor lysate vaccine/Fc-OX40L therapy is capable of inducing enhanced survival and tumor elimination in the GL261 mouse glioma model. The following experiments were performed to determine the mechanism(s) of action of this therapy that elicits a potent antitumor immune response. The evidence subsequently outlined indicates a CD8+ T cell–independent and CD4+ T cell–, NK cell–, and B cell–dependent means of prolonged survival. CD8+ T cell–independent tumor clearance is surprising considering the current focus of many cancer immunotherapy protocols. These results provide evidence for CD8+ T cell–independent means of antitumor response and should lead to additional examination of the potential manipulation of this mechanism for future treatment strategies.

This article is featured in In This Issue, p.1

Introduction

Cancer immunotherapy seeks to recruit an effective immune response to eliminate tumor cells. Unlike direct Ab therapies currently approved for clinical use, cancer vaccines have only shown limited effectiveness due to our incomplete understanding of the necessary effector cells and mechanisms that yield efficient tumor clearance. Despite these setbacks, cancer vaccines remain an enticing approach for the treatment of brain tumors, such as glioblastoma multiforme (GBM). Cytotoxic activity of CD8+ T cells has long been presumed to be the effector function necessary for tumor regression, and many studies have shown that CD8+ T cell responses can be elicited by a variety of immunotherapy approaches (1–4).

CD4+ T cells are more often regarded as helper cells to generate cytotoxic CD8+ T cells and as the source of regulatory T cells (Tregs) induced by the tumor. CD4+ T cells are necessary to prime and maintain a CD8+ T cell response and can alter the tumor microenvironment through secretion of cytokines and the recruitment of additional immune cells, such as eosinophils and macrophages (5). Emerging data suggest that CD4+ T cells may be more important for tumor clearance than previously assumed, and in some cases more efficient than CD8+ T cells (6–9). Although indirect means of tumor elimination have been more extensively studied, CD4+ T cells have been shown to act as efficient cytolytic cells in both tumor and viral infection models (10–12). Despite the growing evidence for CD4+ T cell–based mechanisms of cell killing that are independent of CD8+ T cells, the field of tumor immunotherapy has not fully embraced these important findings.

Many components of the immune system can contribute to effective immunotherapy. In addition to T cells, B cells can also induce antitumor responses. B cells can differentiate into Ab-secreting plasma cells or present Ags to T cells, contributing to T cell activation and memory (13). An intact blood brain barrier restricts the passive diffusion of large proteins, including Igs, into the brain, and thus the role of an Ab response in the efficacy of brain tumor immunotherapy has largely been ignored (14). Paradoxically, one study has shown a correlation between the presence of tumor-reactive Abs in GBM patients and survival (15).

Some important components of the innate immune system are often overlooked in tumor clearance. Characterization of an antitumor response induced by CD40 ligation through administration of an agonist CD40 Ab showed that tumor regression was dependent on macrophages and independent of CD4+ or CD8+ T cells (16). Additionally, eosinophils, mast cells, and NK cells have been shown to have antitumor capabilities (17–19).

The immune-specialized nature of the brain presents unique challenges in developing effective immunotherapies for brain tumors. The brain, even in the absence of tumor, has an immunological preference displayed by the secretion of immunosuppressive molecules, such as TGF-β. This biases the brain and the draining lymph nodes for induction of a Th2 response (20). In addition, even when a CD8+ T cell response is generated, the immunosuppressive nature of the brain can prevent effector function and limit tumor elimination (21).

The clinical success of cancer treatments such as CTLA-4 and PD-1 blockade has led to an increased interest in immune-modulatory agents (22, 23). Signaling through costimulatory molecules, including members of the TNFR superfamily, such as 41BB, OX40, and glucocorticoid-induced TNFR, can lead to T cell expansion and upregulation of effector cytokine production, and can break tolerance (24). These receptors can be targeted by agonist Abs or ligand fusion proteins, in which the corresponding ligands are fused to Ig proteins for systemic delivery in vivo (24).

Recently, we described an efficacious combination therapy involving tumor lysate and adjuvant vaccines with Fc-OX40L costimulation in a murine brain tumor model (25). The work described in this study aims to dissect the mechanisms at work in this potent antitumor therapy in a mouse GBM model. Our results indicate a CD4+ T cell–, B cell–, and NK cell–dependent means of tumor eradication, whereas CD8+ T cells appear to be unnecessary for enhanced tumor-free survival. The following work should aid in the understanding of mechanisms at play in an effective antitumor response and guide future therapeutic designs. Our previous research and current work suggest an alternative means of tumor eradication to the canonical CD8+ cytotoxic T cell mechanism, and may shed light on routes of immune modulation that result in effective tumor clearance in GBM.

Materials and Methods

Animal models and cell lines

GL261-Luc culture conditions have been described previously (26). Animals were maintained in a specific pathogen-free facility, according to the University of Minnesota Institutional Animal Care and Use Committee guidelines. Seven-week-old wild-type (WT) C57BL/6J, B6.129S2-Cd8atm1Mak/J (CD8a knockout [KO]), C57L/6-PrftmSz/J (perforin KO), and B6.129S2-Ighmtm1Cgn/J (μMT) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Five- to 7-wk-old B6.129P2-Fcer1gtm1Rav N12 (FcRγ KO) mice were purchased from Taconic. IgMi mice were previously developed by A. Waisman (University of Mainz, Mainz, Germany) and K. Rajewsky (Harvard Medical School, Boston, MA) (27, 28). Tumors were established by intracranial inoculation of 15,000 GL261-Luc glioma cells in 1 μl HBSS (Life Technologies) into animals anesthetized with a ketamine/xylazine mixture (54.7 mg/ml ketamine and 9.26 mg/ml xylazine). Cells were implanted into the right hemisphere at coordinates 2.5 mm lateral, 0.5 mm anterior from bregma, and 3 mm ventral to the surface of the brain and delivered at a rate of 0.2 μl/min over 5 min (26). Bioluminescence imaging 3 d following inoculation confirmed tumor implantation. Animals received 100 μl luciferin (Gold Biotechnology) by i.p. injection and were imaged with an IVIS50 system (Caliper Life Sciences). Living Image software (Caliper Life Sciences) was used to determine tumor burden in animals as a measure of photons/s (p/s); periodic bioluminescence imaging tracked tumor progression.

Vaccine production and delivery

Vaccines were generated, as previously described (25). Tumor cells were washed three times with PBS, resuspended in PBS, and flash frozen with liquid nitrogen. Cells were subjected to five cycles of freezing in liquid nitrogen and thawing in a 37°C water bath, vortexing after each round, to induce cell lysis. Trypan blue dye exclusion was used to verify complete cell death. A Pierce bicinchoninic acid assay kit (Thermo Scientific) was used to determine protein concentration of the lysates. Purified, endotoxin-free CpG 1826, an unmethylated oligodeoxynucleotide sequence (5′-tccatgacgttcctgacgtt-3′) with a full phosphorothioate backbone (Integrated DNA Technologies, Coralville, IA), was resuspended in 1× Tris-EDTA buffer. Vaccines, consisting of 65 μg tumor lysate and 50 μg CpG 1826 brought to a final volume of 100 μl with saline, were delivered by intradermal injection above the shoulders.

Costimulatory fusion protein production and delivery

Fc-OX40L was developed and verified previously (29). Fc-OX40L was given at 50 μg/dose, brought to a final volume of 100 μl/dose with PBS, and delivered by i.p. injection. Animals received vaccine (intradermal) and Fc-OX40L (i.p.) on days 7, 10, and 13 postinoculation and Fc-OX40L (i.p.) days 15–19, unless otherwise stated.

Lymphocyte depletion

Specific lymphocyte populations were depleted by i.p. injection of depleting Abs. A total of 100 μg anti-NK1.1 (clone PK136; eBioscience) or anti-CD4 (clone GK1.5; eBioscience) or 200 μg anti-CD8 (clone 53-6.7; eBioscience) was delivered for 2 d before the first immunization, then 1 d before each additional vaccine, and 1 d prior to beginning Fc-OX40L treatment. Depletion was verified by euthanizing one animal from each group on day 3 before the first vaccination and analyzing splenocytes by flow cytometry.

Flow cytometry

Brain-infiltrating lymphocytes were harvested from animals on day 25 postinoculation. Mice were euthanized with a ketamine/xylazine mixture and perfused with PBS to flush the capillaries. The brains were removed, minced with a razor, dissociated with TrypLE (Invitrogen), and passed over a 70-μm filter. Leukocytes were collected from a two-layer Percoll gradient (70 and 30%). Cell counts were obtained using trypan blue dye exclusion and a hemocytometer. Cells were stained with the following Abs for phenotypic analysis: CD3 (clone 17A2; eBioscience), CD4 (clone GK1.5; eBioscience), CD8 (clone 5H10; Invitrogen), NK1.1 (clone PK136; eBioscience), MHC II (clone M5.114.15.2; eBioscience), CD11b (clone M1/70; eBioscience), CD11c (clone N418; eBioscience), Foxp3 (clone FJK-16s; eBioscience), and perforin (clone eBioOMAK-D; eBioscience). Intracellular staining of Foxp3 was achieved by utilizing the Foxp3/transcription factor intracellular staining kit, according to the manufacturer’s instructions (eBioscience). The BD Cytofix/Cytoperm plus kit (BD Biosciences) was used to gain intracellular staining of perforin. Flow cytometric analyses were performed on a BD Biosciences FACSCanto, and data were analyzed using Flowjo software (Tree Star). The percentage of stained cells was multiplied by the total number of viable cells, determined previously by trypan blue dye exclusion, to obtain the total number of stained cells, then divided by tumor burden (p/s) to obtain the number of cells relative to tumor size.

To detect tumor-reactive serum Abs, cultured GL261 cells were harvested and incubated with serum (1:100, by volume), washed thoroughly, and stained with a fluorescently labeled rat anti-mouse IgG (Jackson ImmunoResearch Laboratories) or IgM (clone II/41; eBioscience) Ab. Samples were analyzed on a BD Biosciences FACSCanto, and mean fluorescence intensities were determined using Flowjo software (Tree Star).

Histopathology

Tumor-bearing animals were euthanized and perfused with phosphate-buffered water and 4% paraformaldehyde. Formalin-fixed tumor-bearing brains were serially sectioned and processed into paraffin blocks using standard histology techniques, sectioned to 4-μm thickness, and stained with H&E. Slides were evaluated by a board-certified anatomic pathologist by light microscopy.

Suppression assay

T cells were isolated from a naive WT spleen using the mouse CD8a T cell isolation kit from Miltenyi Biotec and stimulated in vitro with plate-bound CD3 (1 μg/ml) (eBioscience; 2C11) in a round-bottom plate. CD11b+ cells isolated from spleen and brains of tumor-bearing animals with the CD11b isolation kit from Miltenyi Biotec were plated with 5 × 104 CD8+ T cells. Tritiated thymidine was added to cells after 48 h and allowed to incubate for an additional 24 h. Proliferation was determined by thymidine incorporation. Suppression was calculated by normalizing to CD3-stimulated T cells in the presence of CD11b+ cells isolated from naive animals, percent suppression = ([(T cells + CD3 + naive CD11b+) − (T cells + CD3 + tumor CD11b+)]/ [T cells + CD3 + naive CD11b+]) × 100.

Western blots

Tumor tissue was sonicated in radioimmunoprecipitation assay buffer (25 mM Tris-HCl, 0.1% SDS, 1% Triton X-100, 1% sodium deoxycholate, 0.15 M NaCl, 1 mM EDTA) with protease and phosphatase inhibitors (1:100; Calbiochem). Protein concentrations were determined using the Pierce bicinchoninic acid assay kit (Thermo Scientific). For SDS-PAGE, lysates were made 2 mg/ml with Laemmli reducing sample buffer. Protein standards (Bio-Rad) were loaded next to each 40 μg lysate and resolved on NuPAGE 4–12% Bis/Tris gels (Invitrogen). Gels were equilibrated for 30 min in Towbin’s transfer buffer, and proteins were then transferred to nitrocellulose membranes (Amersham) at 5 V constant voltage overnight using semidry transfer (Bio-Rad). The membranes were blocked in 5% nonfat dry milk/Tris–Tween 20–buffered saline (TTBS) at room temperature for 1 h. Membranes were then incubated at room temperature for 1 h in HRP-conjugated secondary Ab (Jackson ImmunoResearch) at 1:50,000 in 5% nonfat dry milk/TTBS and washed six times for 10 min each in TTBS. Immunoreactive bands were detected using ECL Plus Western blotting Detection System (Amersham) exposing membranes to HyBlot CL autoradiography film (Denville).

Statistical analyses

A Mann–Whitney U test was used for the statistical comparison of lymphocyte counts and mean fluorescence intensities. Tumor burden between groups was compared using an unpaired Student t test. Animal survival was evaluated by log-rank test. All tests were performed with Prism 4 software (GraphPad Software). The p values <0.05 were considered statistically significant.

Results

Lymphocyte depletion reveals the importance of CD4+ T cells and NK cells in vaccine/Fc-OX40L treatment of GBM

The role of specific lymphocyte subsets contributing to tumor eradication by vaccine/Fc-OX40L treatment was assessed by depleting CD8+, CD4+, and NK cells. Depletion was verified by flow analysis of isolated splenocytes from one animal in each depletion group. Survival of the tumor-bearing depleted animals was compared with replete tumor-bearing animals. Interestingly, CD4 depletion completely abrogated the survival benefit conferred from treatment (Fig. 1A). NK depletion also had an effect on survival, albeit to a lesser magnitude than CD4 depletion. NK-depleted animals survived significantly longer than nondepleted untreated animals, but never as long as nondepleted animals that were treated with vaccine/Fc-OX40L. Surprisingly, CD8 depletion did not negatively affect overall survival (Fig. 1A). To confirm these unexpected findings suggesting the limited role of CD8+ T cells, CD8 KO mice were challenged with GL261 and treated with vaccine/Fc-OX40L. The CD8 KO mice had prolonged survival, confirming the previous results using CD8-depleted mice (Fig. 1B). These data suggest that the immune response responsible for increased survival and tumor clearance is CD8+ T cell independent or that CD8+ T cells somehow block the response to immunotherapy. Thus, contrary to many tumor models in which immunotherapy is employed, these data suggest our vaccine therapy elicits an immune response dominated by CD4+ T cells and NK cells and not CD8+ T cells.

FIGURE 1.
FIGURE 1.

In vivo depletion of lymphocytes. Necessity of lymphocyte populations for treatment efficacy was determined through in vivo depletion of CD4+, CD8+, and NK cells. Vaccine (tumor lysate and CpG) was administered on days 3, 7, and 10, and Fc-OX40L was delivered days 17–21 postinoculation. Depleting Abs were administered i.p. on days 1 and 2 postinoculation and subsequently 1 d before each vaccination (days 6 and 9) and the day before the first Fc-OX40L administration (day 16). (A) Glioma-bearing animals were treated, and survival of animals depleted of lymphocyte populations was compared with nondepleted saline and vaccine/Fc-OX40L–treated animals. (B) To verify the depletion results, the survival of glioma-bearing animals was assessed in WT animals compared with CD8 KO animals. Data represent one experiment; results have been reproduced one additional time, and statistical significance was determined by log-rank test.

B cell–dependent mechanism of vaccine/Fc-OX40L treatment

To determine the extent to which B cells were necessary for the Fc-OX40L treatment, WT and μMT (B cell–deficient) animals were inoculated with GL261 tumor cells and treated with vaccine and Fc-OX40L. Treatment benefit was lost in the μMT animals (Fig. 2A), suggesting that B cells are necessary for proper immune response and eradication of tumor.

FIGURE 2.
FIGURE 2.

B cell–dependent mechanism of treatment efficacy and Ab response. (A) To assess the role of B cells in treatment efficacy, the survival of WT animals was compared with μMT animals receiving treatment. Data represent one experiment; results were reproduced two additional times, and statistical significance was determined by log-rank test. (B) The presence of tumor-reactive Abs in the serum of animals was determined through flow cytometry analysis. Glioma-bearing animals were euthanized at noted time points, and sera collected. Tumor-reactive IgG Abs were plotted based on mean fluorescence intensities. The presence of Igs was detected in tumors harvested from brains of glioma-bearing animals at noted time points. n = 5–11/treatment group. Statistical significance is indicated as *p < 0.05, determined by Mann–Whitney U test. Tumors were run on SDS-PAGE and probed for (C) IgG and (D) IgM to detect the presence of these Igs. Tumors from μMT animals and normal (nontumor-bearing) brains) from WT animals were used as negative controls.

Ab response to vaccine/Fc-OX40L treatment

The B cell dependency noted for treatment efficacy may be mediated through different mechanisms of B cell action. B cells can act as APCs, although this role is less well described for B cells than for dendritic cells. B cells also differentiate into Ab-secreting cells (plasma cells) and exert effector function through Ab-mediated mechanisms. To test the extent to which Abs were playing a role in our tumor model, serum was collected from saline-, vaccine-, Fc-OX40L–, and vaccine/Fc-OX40L–treated animals at day 25 postinoculation. This time point was chosen because the treatment regimen had been completed, but the untreated animals were not yet moribund. Serum was also collected at days 35 and 45 for vaccine/Fc-OX40L–treated animals to assess the later time points when the tumors began to regress. There were no significant differences in the presence of IgM Ab binding to tumor cells among treatment groups as measured through flow cytometry (data not shown). All groups had elevated levels of tumor-reactive IgG Abs relative to serum from a normal (nontumor-bearing) animal (Fig. 2B). Tumor-reactive IgG levels at day 25 were similar among groups, but there was a trend to increasing IgG levels at days 35 and 45 in vaccine/Fc-OX40L–treated animals. The similarity in IgG levels at day 25 suggests an endogenous Ab response in tumor-bearing animals, which may indicate that Fc-OX40L is not involved in inducing Ab secretion. The trend toward increasing levels of IgG over time in the vaccine/Fc-OX40L–treated animals indicates that Fc-OX40L may aid in the maintenance of, or help promote, the Ab response over time.

Ab deposition in tumor

To determine whether Ab was being deposited in the tumors of the vaccine/Fc-OX40L–treated animals and to gain insight into the kinetics of this event, animals were inoculated with GL261 tumor, treated with vaccine/Fc-OX40L, and euthanized on days 25, 35, and 45. Saline- and vaccine-only controls were also collected on day 25. Western blot analysis revealed the presence of IgG in the tumors of animals treated with vaccine/Fc-OX40L (Fig. 2C). Bands consistent with IgG H and L chain (50 and 25 kDa, respectively) were detected in the tumors of animals treated with vaccine/Fc-OX40L. The 50-kDa band consistent with the H chain of IgG was not detected in the tumors of saline-treated animals. The presence of a 25-kDa band in these tumors is most likely due to the cross-reactivity of the Ab with the L chain of the other Igs. IgG was detected in the majority of the vaccine-only–treated tumors as well. Interestingly, whereas all of the tumors analyzed from vaccine/Fc-OX40L–treated mice at day 25 contain IgG, the number of animals with detectable IgG at the later time points decreases at day 35 (4 of 5) and day 45 (3 of 5). This may be due in part to the decreasing tumor burden at these later time points.

IgM was detected in the tumors of animals of all treatment groups, although with varying penetrance (Fig. 2D). The H chain of IgM was detectable and appears to be most penetrant at day 35 in vaccine/Fc-OX40L–treated animals. It is important to note that several gels were run to analyze these samples, so whereas the presence or absence of Ig can be noted, the relative amounts cannot be accurately compared. The presence of IgM and absence of IgG in the tumors of saline-treated mice suggest that infiltration of class-switched Ig may be impaired.

Analysis of brain-infiltrating cells

Flow cytometry analysis was performed on lymphocytes isolated from the brains of tumor-bearing animals. Glioma-bearing WT animals treated with saline, vaccine only, Fc-OX40L only, and vaccine/Fc-OX40L were euthanized on day 25. One day prior to euthanasia, tumor burden was assessed by bioluminescence imaging and revealed a significant difference in the tumor size among the groups. Animals receiving Fc-OX40L as part of the treatment showed significantly less tumor burden (Fig. 3A). Lymphocytes harvested from the brains were counted and plotted as absolute numbers. The number of brain-infiltrating cells observed in the WT animals receiving vaccine/Fc-OX40L treatment was not significantly greater than saline-treated animals (Fig. 3A). The large variation in tumor size may distort the true value of infiltrating cells present in the brain, as larger tumors may contain more infiltrating cells simply due to volume. To compensate for this difference, the number of brain-infiltrating cells was normalized to the tumor burden determined by bioluminescence imaging (p/s). This analysis revealed that, based on tumor burden, there were more lymphocytes infiltrating the tumors of Fc-OX40L–treated WT animals (Fig. 3A).

FIGURE 3.
FIGURE 3.

Analysis of brain-infiltrating cells. (A) Tumor burden in glioma-bearing animals was determined by bioluminescence imaging 1 d prior to euthanasia. The absolute number of brain-infiltrating lymphocytes for each treatment group was graphed, and the number of brain-infiltrating lymphocytes was normalized relative to tumor burden (p/s). The presence of T lymphocytes was assessed through flow cytometry analysis. (B) The CD4+ T cell population was determined by CD3+CD4+Foxp3-negative staining. (C) The CD8+ T cell population was determined by CD3+CD8+ staining. (D) A Treg population was determined by CD3+CD4+ Foxp3+ staining. The presence of a NK cell population was assessed through flow cytometry analysis. (E) NK cells were determined based on CD3-negative NK1.1+ staining. (F) The NK cell population was analyzed for expression of perforin. (G) The NKT cell population was determined by CD3+ NK1.1+ staining. (H) The NKT cell population was analyzed for expression of perforin. (I) The presence of a CD11b+ MHC II-negative granular population was determined. Each population is represented as percentage of total live cells, absolute number, and number of cells relative to tumor burden. Mean values are shown as ±SEM; n = 6–10/group. Statistical significance is indicated as *p < 0.05 compared with saline group. Data represent the combined results from two experiments, and statistical significance was determined by Mann–Whitney U test.

Phenotypic staining assessed the composition of the lymphocyte population harvested from the brains of glioma-bearing animals. The T cell population was determined by analyzing the CD3+CD4+Foxp3neg and CD3+CD8+ populations. An increase in the infiltration of CD4+ T cells was observed for the WT animals treated with Fc-OX40L (Fig. 3B), consistent with the earlier finding that CD4+ T cells are necessary for efficacy of treatment (Fig. 1A). Interestingly, an increase in infiltration of CD8+ T cells was observed in the vaccine/Fc-OX40L–treated animals (Fig. 3C), despite the fact that these cells are not necessary for tumor clearance (Fig. 1). Although these cells are present, this analysis does not indicate whether this population is active. Analysis for the presence of perforin, used in the cytolytic function of CD8+ cells, revealed that this population does not express perforin, which may suggest that they are not actively functioning as killer cells (data not shown). Importantly, CD4+ T cells also did not demonstrate expression of perforin.

Ligation of OX40 on the surface of Tregs can decrease the suppressive function of this population (30, 31). It is reasonable to assume that one mechanism of Fc-OX40L in tumor eradication is to suppress the regulatory function of these cells, thus allowing the effector cells to kill. Analysis of the CD3+ CD4+ Foxp3+ Treg population in the brains demonstrates an increase in the presence of Tregs in the tumors of animals treated with Fc-OX40L (Fig. 3D). This is contrary to what was expected; however, it is possible that the increased immune infiltration observed triggers an influx of Tregs to balance the inflammatory response. This analysis also does not indicate whether this population is actively suppressing effector cells; functional analysis of isolated cells would be required to fully understand the effect on Tregs.

The NK cell population was of particular interest as its depletion resulted in loss of treatment efficacy (Fig. 1A). The total CD3negNK1.1+ population, which excludes NKT cells, was assessed and observed to be increased in Fc-OX40L–treated animals only when normalized to tumor burden (Fig. 3E). Impressively, the CD3negNK1.1+perforin+ population was increased in the Fc-OX40L–treated groups (Fig. 3F), demonstrating that NK cells are present and presumably functional. An intriguing observation was the presence of a distinct CD3+NK1.1+ population, NKT cells, in the brains of Fc-OX40L–treated animals (Fig. 3G), and, surprisingly, a significant number of these cells expressed perforin (Fig. 3H).

The most noticeable difference among the treatment groups was a granular cell population in the vaccine/Fc-OX40L treatment groups that was absent in the saline and the vaccine treatment groups (Fig. 3I). Phenotypic analysis revealed that the majority of cells in this population are CD11b+ MHC IIneg, indicative of neutrophils, which like NK cells are able to kill Ab-coated cells through Ab-dependent cell-mediated cytotoxicity (ADCC) (32, 33).

Involvement of the FcR and Ab

The indication that Ab production may play a significant role in tumor elimination combined with the loss of treatment efficacy with NK cell depletion (Fig. 1A) led to the investigation of the role that ADCC may play in treatment-induced tumor regression. ADCC requires the Fc portion of Igs bound to target Ag to bind to FcR on the surface of effector cells, most commonly NK cells. This binding induces lysis of the target cell by the effector cell, generally through the production and release of perforin and granzymes. To test whether this mechanism was the link between the NK cell and the Ab results, treatment efficacy was tested in mice deficient for the γ-chain of the FcR (FcRγ KO). These mice lack the FcRs capable of binding IgG and IgE, and therefore ADCC action mediated through these Igs would not be possible. Whereas there was a significant difference in the overall survival of saline-treated WT and FcRγ KO animals (p = 0.0146), suggestive of an endogenous response involving the FcR, the overall survival of WT and FcRγ KO animals that received vaccine/Fc-OX40L treatment did not differ significantly (Fig. 4A). However, there is clearly a trend toward an intermediate phenotype that is best illustrated in the tumor burden of these animals. Bioluminescence imaging was performed to track changes in tumor burden. These studies showed that the rate of tumor growth was rapid in saline-treated mice, but was slowed by vaccine/Fc-OX40L treatment especially in the WT mice compared with the FcRγ KO mice (Fig. 4B).

FIGURE 4.
FIGURE 4.

Involvement of the FcR and Ab. To determine the necessity of Ab-dependent mechanisms in tumor clearance, the tumor growth of FcRγ KO and IgMi mice was compared with WT mice. (A) Overall survival of glioma-bearing WT and FcRγ KO animals; data represent the combined results from two experiments, and statistical significance was determined by log-rank test. (B) Bioluminescence imaging plots tumor burden (p/s) over time; each line represents the average signal for each group. Mean values are shown as ±SEM, n = 5–8/group. Statistical significance is indicated as *p < 0.05 compared with FcRγ KO vaccine/Fc-OX40L group by unpaired Student t test. (C) Survival of IgMi mice compared with WT animals. Data represent the combined results from two experiments, and statistical significance was determined by log-rank test. (D) Average tumor burden (p/s) for each group over time. Mean values are shown as ±SEM, n = 5–7/group. Statistical significance is indicated as *p < 0.05 compared with IgMi vaccine/Fc-OX40L group by unpaired Student t test.

To verify whether indeed Abs were necessary for the efficacy of our treatment, we tested our therapy in an animal model deficient for plasma cells (IgMi mice) (27, 28). We observed an intermediate phenotype in response to treatment in these mice. The IgMi vaccine/Fc-OX40L–treated animals survived significantly longer than the saline-treated animals, but did not reach the full survival benefit of the WT animals (Fig. 4C). Once again, the intermediate phenotype in response to treatment is reflected in the tumor burden of these animals, as measured by bioluminescence imaging (Fig. 4D).

Overall, survival after vaccine/Fc-OX40L treatment of the plasma cell-deficient mice paralleled that of FcRγ KO, suggesting that the Ab dependence is likely through a FcR-based mechanism. Whereas Ab-mediated mechanisms are not required for complete tumor clearance, Ab does play a role in controlling tumor growth early in the course of treatment based on the greater tumor burden seen in the IgMi and FcRγ KO mice.

A mechanism of NK killing through ADCC is through the release of perforin and granzymes, and we observed an increase in the perforin-expressing NK cells after treatment; however, when the efficacy of the vaccine/Fc-OX40L treatment was tested in perforin-deficient (perforin KO) animals, we found that treatment was not perforin dependent (Fig. 5).

FIGURE 5.
FIGURE 5.

Perforin-independent survival. Perforin-deficient and WT mice were treated with vaccine/Fc-OX40L and analyzed for survival. Data represent one experiment, and statistical significance was determined by log-rank test.

Tumor infiltration of granular population with Fc-OX40L treatment

The granular population observed by flow cytometry (Figs. 3I, 6A) was further analyzed to determine its role in the therapeutic response. Histologic analysis revealed that tumors from multiple vaccine/Fc-OX40L–treated animals were infiltrated by a moderate to large number of neutrophils, which in some areas of the tumor formed dense aggregates and replaced neoplastic cells. Neutrophils were characterized by a multilobed nucleus and a homogenous eosinophilic cytoplasm. Tumors in saline-treated mice were infiltrated by a moderate number of mononuclear cells, but neutrophils were not a prominent component of the inflammatory cell infiltrate as determined by a pathologist (Fig. 6B). It is important to note that this population is also consistent with the myeloid-derived suppressor cell population, which is often induced by the tumor itself and generally only distinguishable from neutrophils by functional assays. Thus, a suppression assay was performed, revealing that the CD11b+ cells isolated from the brains of vaccine/Fc-OX40L–treated animals were not functionally suppressive (Fig. 6C). Furthermore, when nontumor-bearing animals were treated with Fc-OX40L, there was a dramatic increase in the presence of CD11b+ cells in the blood, suggesting that the treatment, rather than the tumor, is responsible for neutrophil recruitment (Fig. 6D).

FIGURE 6.
FIGURE 6.

Observation of a distinct granular population. (A) Flow cytometry scatter plots of brain-infiltrating cells revealed a distinct granular population. Representative plot, data from three independent experiments reproduce this result. (B) Histologic analysis demonstrates a dense aggregate of neutrophils (arrows) within the tumor of a vaccine/Fc-OX40L–treated animal that is absent in a saline-treated animal. H&E stain. (C) The suppressive capacity of CD11b+ cells was determined by incubating these cells with CD8+ T cells and plate-bound CD3 at a ratio of 1:1. T cell activation was measured by [3H]-thymidine uptake and graphed as percent suppression, n = 3 for vaccine/Fc-OX40L and saline brain-infiltrating lymphocytes, n = 1 for saline spleen; all samples were performed in triplicate. This experiment was performed once. (D) Analysis of blood from tumor- and nontumor-bearing animals at day 14. Mean values are shown as ±SEM; n = 3–5/group. Statistical significance, determined by unpaired Student t test, is indicated as *p < 0.05.

Discussion

Despite the increasing interest in utilizing immunotherapeutic approaches in cancer treatment, the prognosis for patients diagnosed with glioma remains poor. Advances in the field of cancer immunotherapy have been incremental, and an inherent hurdle to progress is the lack of a basic understanding of the mechanisms needed for an effective immune-based antitumor response. One route of gaining insight into these mechanisms is to examine the immune players at work during an effective antitumor response. Because prior studies by our laboratory have identified a potent antitumor treatment in a murine model of glioma (25), we were in the position to define the mechanisms responsible for tumor clearance.

In the present report, our data demonstrate that this response is dependent on CD4+ T cell and NK cell responses, as efficacy is lost upon depletion of these cells. In contrast to many other studies, our data indicate that the CD4+ T cell response in this model was critical for tumor regression. In this immunotherapeutic model, however, further work remains to be done to identify the exact mechanism of CD4+ T cell action. CD4+ T cells could potentially be orchestrating the activation and infiltration of other immune cells that carry out the cytotoxic functions. In contrast, CD4+ T cells are capable of cytolytic function through their direct and indirect killing in viral and tumor models (6, 12, 34). It is possible that our model may be reproducing the results observed in these previous publications. Whereas the exact mechanisms remain to be elucidated, the perforin-independent tumor regression would suggest that if CD4+ T cells are mediating a cytotoxic response, pathways such as TRAIL and FasL should be examined.

Many cancer vaccines have been designed to induce a robust CD8+ T cell response. Surprisingly, we did not observe a dependency of CD8+ T cells, as CD8 depletion did not alter survival outcome. An increase in the median survival of animals depleted of CD8+ T cells was observed (Fig. 1A), initially suggesting that CD8+ T cells may have a deleterious effect on survival. However, the overall survival did not achieve statistical significance and further analysis of survival in CD8 KO animals did not reveal any difference in median survival. Although tumor-infiltrating CD8+ T cells were observed in vaccine/Fc-OX40L–treated animals (Fig. 3C), the results of the depletion study suggest these cells may be functionally inactive. It is also possible that, in an attempt to prevent tissue damage from an inflammatory response, the CD8+ T cells may be suppressive. Previous laboratories have shown that CD8+ T cells express the suppressive cytokine IL-10 in response to viral infections in the lung and brain, where prolonged activation may cause deleterious bystander effects (35, 36). These remain attractive theories that need to be tested, and the optimized strategy of vaccine/Fc-OX40L in the GL261 glioma model presents as a robust model to pursue these questions in the future.

Additionally, B cells are required for the efficacy of treatment. B cells may be necessary to act as APCs for CD4+ T cell activity or they may be differentiating into plasma cells and secreting tumor-reactive Abs or Abs required for ADCC. The observation that tumor-bearing animals generate tumor-reactive Abs and the deposition of Ab in the tumors of vaccine/Fc-OX40L–treated animals points in this direction. The loss of control of tumor growth in FcRγ KO and IgMi mice suggests that ADCC mechanisms may be a contributing factor in the therapeutic response; however, due to the incomplete loss of efficacy, there are obviously other mechanisms contributing to tumor clearance.

The infiltration of tumor by immune cells is key in understanding the mechanism of tumor clearance. The most notable observation was the infiltration of NK cells, NKT cells, and the large granular population. Recruitment and activation of NK cells, NKT cells, and neutrophils may be mediated by cytokine secretion by CD4+ T cells activated through OX40 ligation. Another explanation could involve direct activation by OX40L. Although the majority of research conducted on OX40:OX40L interactions has focused on activated T cells, there have been some reports indicating that NK cells (37), NKT cells (38), and neutrophils (39) may express the OX40 receptor. It is possible that Fc-OX40L administration is acting directly on the innate immune cells to induce a productive antitumor response, although this remains controversial.

The observed infiltration of innate immune cells is particularly interesting in light of our Ab data (Figs. 2, 4), due to the known link between the NK cells, neutrophils, and ADCC. It is possible that these cells are actively killing Ab-coated tumor cells and inhibiting growth in a FcR-dependent mechanism. The presence of a large granulocyte population remains an interesting potential mechanism of tumor clearance. These cells were found to infiltrate into the tumors of animals treated with Fc-OX40L, and these mice also demonstrated the greatest response to treatment.

As shown by our data, multiple mechanisms of tumor clearance may be occurring in this tumor model. Too often in the design of immunotherapeutic studies, particular immune subsets are targeted while ignoring the remainder of the immune system. This is particularly noted for CD8+ T cell–biased therapies. Many vaccines have focused on CD8+ T cell–restricted Ags or adoptive transfer exclusively (2). Even in the case of CD8+ T cell–mediated tumor clearance, recruitment of additional immune cells is necessary, particularly CD4+ Th1 cells that are needed for maintenance of the CD8+ T cell response. The observation of an increase in the influx of innate cells to the tumor site upon treatment and the presence of a tumor-reactive Ab response may be linked through a FcR-dependent mechanism. A deeper understanding of immune-mediated mechanism of tumor clearance will certainly drive the design of immunotherapy regimens in the clinic with the hope of fully harnessing the power of the immune system to mediate lasting tumor clearance with memory. Considering that the CNS may provide unique challenges for tumor immunology, it is entirely possible that different cell populations may be involved in effective tumor clearance than that found in solid tumors of peripheral tissues.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We acknowledge the University of Minnesota Masonic Cancer Center Comparative Pathology Shared Resource for histology preparations. We are grateful to Deb Todhunter for assistance with the thymidine incorporation assay. The IgMi mice were a gift of Dr. Ari Waisman.

Footnotes

  • This work was supported in part by National Institutes of Health Grants R01 CA154345 and R01 CA160782, American Cancer Society Grant RSG-09-189-01-LIB, the Minnesota Medical Foundation, the Hedberg Family Foundation, and the Children’s Cancer Research Fund (to J.R.O.).

  • Abbreviations used in this article:

    ADCC
    Ab-dependent cell-mediated cytotoxicity
    GBM
    glioblastoma multiforme
    KO
    knockout
    p/s
    photons/s
    μMT
    B6.129S2-Ighmtm1Cgn/J
    Treg
    regulatory T cell
    TTBS
    Tris–Tween 20–buffered saline
    WT
    wild-type.

  • Received June 20, 2013.
  • Accepted November 2, 2013.

References

  1. ↵
    1. Nelson D. J.,
    2. S. Mukherjee,
    3. C. Bundell,
    4. S. Fisher,
    5. D. van Hagen,
    6. B. Robinson
    . 2001. Tumor progression despite efficient tumor antigen cross-presentation and effective “arming” of tumor antigen-specific CTL. J. Immunol. 166: 5557–5566.
    OpenUrlAbstract/FREE Full Text
  2. ↵
    1. Wang R. F.
    2001. The role of MHC class II-restricted tumor antigens and CD4+ T cells in antitumor immunity. Trends Immunol. 22: 269–276.
    OpenUrlCrossRefPubMed
    1. Heimberger A. B.,
    2. J. H. Sampson
    . 2011. Immunotherapy coming of age: what will it take to make it standard of care for glioblastoma? Neuro-oncol. 13: 3–13.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Lee K. H.,
    2. E. Wang,
    3. M. B. Nielsen,
    4. J. Wunderlich,
    5. S. Migueles,
    6. M. Connors,
    7. S. M. Steinberg,
    8. S. A. Rosenberg,
    9. F. M. Marincola
    . 1999. Increased vaccine-specific T cell frequency after peptide-based vaccination correlates with increased susceptibility to in vitro stimulation but does not lead to tumor regression. J. Immunol. 163: 6292–6300.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Hung K.,
    2. R. Hayashi,
    3. A. Lafond-Walker,
    4. C. Lowenstein,
    5. D. Pardoll,
    6. H. Levitsky
    . 1998. The central role of CD4(+) T cells in the antitumor immune response. J. Exp. Med. 188: 2357–2368.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Perez-Diez A.,
    2. N. T. Joncker,
    3. K. Choi,
    4. W. F. Chan,
    5. C. C. Anderson,
    6. O. Lantz,
    7. P. Matzinger
    . 2007. CD4 cells can be more efficient at tumor rejection than CD8 cells. Blood 109: 5346–5354.
    OpenUrlAbstract/FREE Full Text
    1. Hirschhorn-Cymerman D.,
    2. S. Budhu,
    3. S. Kitano,
    4. C. Liu,
    5. F. Zhao,
    6. H. Zhong,
    7. A. M. Lesokhin,
    8. F. Avogadri-Connors,
    9. J. Yuan,
    10. Y. Li,
    11. et al
    . 2012. Induction of tumoricidal function in CD4+ T cells is associated with concomitant memory and terminally differentiated phenotype. J. Exp. Med. 209: 2113–2126.
    OpenUrlAbstract/FREE Full Text
    1. Muranski P.,
    2. Z. A. Borman,
    3. S. P. Kerkar,
    4. C. A. Klebanoff,
    5. Y. Ji,
    6. L. Sanchez-Perez,
    7. M. Sukumar,
    8. R. N. Reger,
    9. Z. Yu,
    10. S. J. Kern,
    11. et al
    . 2011. Th17 cells are long lived and retain a stem cell-like molecular signature. Immunity 35: 972–985.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Braumüller H.,
    2. T. Wieder,
    3. E. Brenner,
    4. S. Aßmann,
    5. M. Hahn,
    6. M. Alkhaled,
    7. K. Schilbach,
    8. F. Essmann,
    9. M. Kneilling,
    10. C. Griessinger,
    11. et al
    . 2013. T-helper-1-cell cytokines drive cancer into senescence. Nature 494: 361–365.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Mucida D.,
    2. M. M. Husain,
    3. S. Muroi,
    4. F. van Wijk,
    5. R. Shinnakasu,
    6. Y. Naoe,
    7. B. S. Reis,
    8. Y. Huang,
    9. F. Lambolez,
    10. M. Docherty,
    11. et al
    . 2013. Transcriptional reprogramming of mature CD4+ helper T cells generates distinct MHC class II-restricted cytotoxic T lymphocytes. Nat. Immunol. 14: 281–289.
    OpenUrlCrossRefPubMed
    1. Curran M. A.,
    2. T. L. Geiger,
    3. W. Montalvo,
    4. M. Kim,
    5. S. L. Reiner,
    6. A. Al-Shamkhani,
    7. J. C. Sun,
    8. J. P. Allison
    . 2013. Systemic 4-1BB activation induces a novel T cell phenotype driven by high expression of Eomesodermin. J. Exp. Med. 210: 743–755.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Soghoian D. Z.,
    2. H. Streeck
    . 2010. Cytolytic CD4(+) T cells in viral immunity. Expert Rev. Vaccines 9: 1453–1463.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Crawford A.,
    2. M. Macleod,
    3. T. Schumacher,
    4. L. Corlett,
    5. D. Gray
    . 2006. Primary T cell expansion and differentiation in vivo requires antigen presentation by B cells. J. Immunol. 176: 3498–3506.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Deeken J. F.,
    2. W. Löscher
    . 2007. The blood-brain barrier and cancer: transporters, treatment, and Trojan horses. Clin. Cancer Res. 13: 1663–1674.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Pallasch C. P.,
    2. A. K. Struss,
    3. A. Munnia,
    4. J. König,
    5. W. I. Steudel,
    6. U. Fischer,
    7. E. Meese
    . 2005. Autoantibodies against GLEA2 and PHF3 in glioblastoma: tumor-associated autoantibodies correlated with prolonged survival. Int. J. Cancer 117: 456–459.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Beatty G. L.,
    2. E. G. Chiorean,
    3. M. P. Fishman,
    4. B. Saboury,
    5. U. R. Teitelbaum,
    6. W. Sun,
    7. R. D. Huhn,
    8. W. Song,
    9. D. Li,
    10. L. L. Sharp,
    11. et al
    . 2011. CD40 agonists alter tumor stroma and show efficacy against pancreatic carcinoma in mice and humans. Science 331: 1612–1616.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Tepper R. I.,
    2. R. L. Coffman,
    3. P. Leder
    . 1992. An eosinophil-dependent mechanism for the antitumor effect of interleukin-4. Science 257: 548–551.
    OpenUrlAbstract/FREE Full Text
  18. Purwar, R., C. Schlapbach, S. Xiao, H. S. Kang, W. Elyaman, X. Jiang, A. M. Jetten, S. J. Khoury, R. C. Fuhlbrigge, V. K. Kuchroo, et al. 2012. Robust tumor immunity to melanoma mediated by interleukin-9-producing T cells. Nat. Med. 18: 1248-1253.
  19. ↵
    1. Liu R. B.,
    2. B. Engels,
    3. A. Arina,
    4. K. Schreiber,
    5. E. Hyjek,
    6. A. Schietinger,
    7. D. C. Binder,
    8. E. Butz,
    9. T. Krausz,
    10. D. A. Rowley,
    11. et al
    . 2012. Densely granulated murine NK cells eradicate large solid tumors. Cancer Res. 72: 1964–1974.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Harling-Berg C. J.,
    2. T. J. Park,
    3. P. M. Knopf
    . 1999. Role of the cervical lymphatics in the Th2-type hierarchy of CNS immune regulation. J. Neuroimmunol. 101: 111–127.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Gordon L. B.,
    2. S. C. Nolan,
    3. H. F. Cserr,
    4. P. M. Knopf,
    5. C. J. Harling-Berg
    . 1997. Growth of P511 mastocytoma cells in BALB/c mouse brain elicits CTL response without tumor elimination: a new tumor model for regional central nervous system immunity. J. Immunol. 159: 2399–2408.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Hodi F. S.,
    2. S. J. O’Day,
    3. D. F. McDermott,
    4. R. W. Weber,
    5. J. A. Sosman,
    6. J. B. Haanen,
    7. R. Gonzalez,
    8. C. Robert,
    9. D. Schadendorf,
    10. J. C. Hassel,
    11. et al
    . 2010. Improved survival with ipilimumab in patients with metastatic melanoma. N. Engl. J. Med. 363: 711–723.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Topalian S. L.,
    2. F. S. Hodi,
    3. J. R. Brahmer,
    4. S. N. Gettinger,
    5. D. C. Smith,
    6. D. F. McDermott,
    7. J. D. Powderly,
    8. R. D. Carvajal,
    9. J. A. Sosman,
    10. M. B. Atkins,
    11. et al
    . 2012. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N. Engl. J. Med. 366: 2443–2454.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Pardee A. D.,
    2. A. K. Wesa,
    3. W. J. Storkus
    . 2009. Integrating costimulatory agonists to optimize immune-based cancer therapies. Immunotherapy 1: 249–264.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Murphy K. A.,
    2. M. G. Lechner,
    3. F. E. Popescu,
    4. J. Bedi,
    5. S. A. Decker,
    6. P. Hu,
    7. J. R. Erickson,
    8. M. G. O’Sullivan,
    9. L. Swier,
    10. A. M. Salazar,
    11. et al
    . 2012. An in vivo immunotherapy screen of costimulatory molecules identifies Fc-OX40L as a potent reagent for the treatment of established murine gliomas. Clin. Cancer Res. 18: 4657–4668.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Wu A.,
    2. S. Oh,
    3. S. Gharagozlou,
    4. R. N. Vedi,
    5. K. Ericson,
    6. W. C. Low,
    7. W. Chen,
    8. J. R. Ohlfest
    . 2007. In vivo vaccination with tumor cell lysate plus CpG oligodeoxynucleotides eradicates murine glioblastoma. J. Immunother. 30: 789–797.
    OpenUrlCrossRef
  27. ↵
    1. Waisman A.,
    2. M. Kraus,
    3. J. Seagal,
    4. S. Ghosh,
    5. D. Melamed,
    6. J. Song,
    7. Y. Sasaki,
    8. S. Classen,
    9. C. Lutz,
    10. F. Brombacher,
    11. et al
    . 2007. IgG1 B cell receptor signaling is inhibited by CD22 and promotes the development of B cells whose survival is less dependent on Ig alpha/beta. J. Exp. Med. 204: 747–758.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Waisman A.,
    2. A. L. Croxford,
    3. F. Demircik
    . 2008. New tools to study the role of B cells in cytomegalovirus infections. Med. Microbiol. Immunol. 197: 145–149.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Sadun R. E.,
    2. W. E. Hsu,
    3. N. Zhang,
    4. Y. C. Nien,
    5. S. A. Bergfeld,
    6. H. Sabzevari,
    7. M. E. Lutsiak,
    8. L. Khawli,
    9. P. Hu,
    10. A. L. Epstein
    . 2008. Fc-mOX40L fusion protein produces complete remission and enhanced survival in 2 murine tumor models. J. Immunother. 31: 235–245.
    OpenUrlCrossRef
  30. ↵
    1. Ruby C. E.,
    2. M. A. Yates,
    3. D. Hirschhorn-Cymerman,
    4. P. Chlebeck,
    5. J. D. Wolchok,
    6. A. N. Houghton,
    7. H. Offner,
    8. A. D. Weinberg
    . 2009. Cutting edge: OX40 agonists can drive regulatory T cell expansion if the cytokine milieu is right. J. Immunol. 183: 4853–4857.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Burocchi A.,
    2. P. Pittoni,
    3. A. Gorzanelli,
    4. M. P. Colombo,
    5. S. Piconese
    . 2011. Intratumor OX40 stimulation inhibits IRF1 expression and IL-10 production by Treg cells while enhancing CD40L expression by effector memory T cells. Eur. J. Immunol. 41: 3615–3626.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Challacombe J. M.,
    2. A. Suhrbier,
    3. P. G. Parsons,
    4. B. Jones,
    5. P. Hampson,
    6. D. Kavanagh,
    7. G. E. Rainger,
    8. M. Morris,
    9. J. M. Lord,
    10. T. T. Le,
    11. et al
    . 2006. Neutrophils are a key component of the antitumor efficacy of topical chemotherapy with ingenol-3-angelate. J. Immunol. 177: 8123–8132.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Delves P. J.,
    2. I. M. Roitt
    . 2000. The immune system: second of two parts. N. Engl. J. Med. 343: 108–117.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Brady M. S.,
    2. F. Lee,
    3. H. Petrie,
    4. D. D. Eckels,
    5. J. S. Lee
    . 2000. CD4(+) T cells kill HLA-class-II-antigen-positive melanoma cells presenting peptide in vitro. Cancer Immunol. Immunother. 48: 621–626.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Trandem K.,
    2. J. Zhao,
    3. E. Fleming,
    4. S. Perlman
    . 2011. Highly activated cytotoxic CD8 T cells express protective IL-10 at the peak of coronavirus-induced encephalitis. J. Immunol. 186: 3642–3652.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Palmer E. M.,
    2. B. C. Holbrook,
    3. S. Arimilli,
    4. G. D. Parks,
    5. M. A. Alexander-Miller
    . 2010. IFNgamma-producing, virus-specific CD8+ effector cells acquire the ability to produce IL-10 as a result of entry into the infected lung environment. Virology 404: 225–230.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Liu C.,
    2. Y. Lou,
    3. G. Lizée,
    4. H. Qin,
    5. S. Liu,
    6. B. Rabinovich,
    7. G. J. Kim,
    8. Y. H. Wang,
    9. Y. Ye,
    10. A. G. Sikora,
    11. et al
    . 2008. Plasmacytoid dendritic cells induce NK cell-dependent, tumor antigen-specific T cell cross-priming and tumor regression in mice. J. Clin. Invest. 118: 1165–1175.
    OpenUrlPubMed
  38. ↵
    1. Zaini J.,
    2. S. Andarini,
    3. M. Tahara,
    4. Y. Saijo,
    5. N. Ishii,
    6. K. Kawakami,
    7. M. Taniguchi,
    8. K. Sugamura,
    9. T. Nukiwa,
    10. T. Kikuchi
    . 2007. OX40 ligand expressed by DCs costimulates NKT and CD4+ Th cell antitumor immunity in mice. J. Clin. Invest. 117: 3330–3338.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Baumann R.,
    2. S. Yousefi,
    3. D. Simon,
    4. S. Russmann,
    5. C. Mueller,
    6. H. U. Simon
    . 2004. Functional expression of CD134 by neutrophils. Eur. J. Immunol. 34: 2268–2275.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section