A variety of cancers, including malignant gliomas, show aberrant activation of STAT3, which plays a pivotal role in negative regulation of antitumor immunity. We hypothesized that inhibition of STAT3 signals would improve the efficacy of T cell adoptive transfer therapy by reversal of STAT3-induced immunosuppression in a murine GL261 intracranial glioma model. In vitro treatment of GL261 cells with JSI-124, a STAT3 inhibitor, reversed highly phosphorylated status of STAT3. Systemic i.p. administration of JSI-124 in glioma-bearing immunocompetent mice, but not athymic mice, resulted in prolonged survival, suggesting a role of adaptive immunity in the antitumor effect. Furthermore, JSI-124 promoted maturation of tumor-infiltrating CD11c+ dendritic cells and activation of tumor-conditioned cytotoxic T cells, enhanced dendritic cells and GL261 production of CXCL-10, a critical chemokine for attraction of Tc1 cells. When i.p. JSI-124 administration was combined with i.v. transfer of Pmel-I mouse-derived type-1 CTLs (Tc1), glioma-bearing mice exhibited prolonged survival compared with i.p. JSI-124 or i.v. Tc1 therapy alone. Flow cytometric analyses of brain infiltrating lymphocytes revealed that JSI-124-treatment enhanced the tumor-homing of i.v. transferred Tc1 cells in a CXCL-10-dependent fashion. Systemic JSI-124 administration also up-regulated serum IL-15 levels, and promoted the persistence of transferred Tc1 in the host. These data suggest that systemic inhibition of STAT3 signaling can reverse the suppressive immunological environment of intracranial tumor bearing mice both systemically and locally, thereby promoting the efficacy of adoptive transfer therapy with Tc1.
Malignant gliomas, such as glioblastoma multiforme, represent the most common primary brain tumors and possess dismal prognosis. Over 12,000 new cases are annually diagnosed in the U.S. (1); the median survival rate is ∼15 mo with the current standard of care, including surgery, radiation therapy, and chemotherapy (2).
Development of novel, molecularly targeted, multimodal therapeutic approaches is critical to further improve the outcome of these deadly tumors. Immunotherapy for glioma is an attractive approach because activated, antitumor immune cells may have the potential to migrate into the CNS and selectively destroy malignant cells that have infiltrated normal CNS tissues (3, 4, 5, 6). Dedicated to development of novel immunotherapy approaches for glioma, we have addressed our major focus on identification and characterization of novel HLA-A2-restricted CTL epitopes derived from glioma-associated Ags (GAAs)3 such as IL-13Rα2 (7, 8) and a receptor-tyrosine kinase EphA2 (9). A novel vaccine trial using synthetic peptides encoding these GAA-derived CTL epitopes and a TLR-3 ligand poly-ICLC is presently underway (10).
An additional challenge in vaccine strategies for progressive malignant glioma is systemic suppression of immunity due to chemoradiotherapy (11) and tumor elaboration of immunosuppressive substances (reviewed in Ref. 12). Although active (i.e., direct) immunization with GAA-vaccines relies on intact host-immune reactivity, recent studies in melanoma patients demonstrated that passive immunization via i.v. adoptive transfer of tumor-reactive, ex vivo-activated T cells may instead take advantage of conditions induced by preceding nonmyeloablative but lympho-depleting chemotherapy regimens (13, 14, 15). This strategy may be particularly suitable for patients with malignant gliomas because the clinical use of chemotherapeutic agents has become a part of standard care in these patients (2).
We have previously demonstrated that intratumoral injections of DCs that are ex vivo transduced with IFN-α cDNA enhance the tumor-homing and therapeutic efficacy of i.v. transferred type-1 CTL (Tc1) in a CXCL-10-dependent manner (6). This study has suggested that reversal of immunosuppressive tumor microenvironment, especially at the level of local APCs, may be necessary to improve the therapeutic efficacy of adoptive CTL-transfer approaches.
STAT3 has emerged as an important target for effective immunotherapy (reviewed in Ref. 16). A variety of cancers including gliomas exhibits aberrant activation of STAT3 (17, 18, 19, 20), which plays a pivotal role in the negative regulation of tumor immunity. Soluble factors derived from STAT3 active tumors, such as IL-6, IL-10, and vascular endothelial growth factor (VEGF), activate STAT3 in DC, thereby preventing their maturation (21). STAT3 appears to be constitutively activated in diverse tumor-infiltrating immune cells, and ablation of STAT3 in hemopoietic cells triggers an intrinsic immune surveillance system that inhibits tumor growth and metastasis via enhanced function of DCs, T cells, NK cells, and neutrophils in tumor-bearing mice with STAT3−/− hemopoietic cells (22). Activation of STAT3 in human glioma has been well documented (18, 19, 23, 24, 25). Gliomas produce high levels of IL-10 (reviewed in Ref. 12) and VEGF (reviewed in Ref. 26) and hence, it is postulated that glioma-infiltrating lymphocytes may display activated STAT3 status, which may constitute a key molecular mechanism underlying glioma-induced immunosuppression.
JSI-124 (cucurbitacin I) was identified as a highly selective and potent inhibitor of phosphotyrosine (p)STAT3 from the National Cancer Institute Diversity Set (27). JSI-124 not only directly inhibits the growth of STAT3-active tumors, but also promotes the differentiation of DCs (28, 29). JSI-124 significantly reduced the presence of immature myeloid cells in vivo and promoted accumulation of mature DCs, thereby enhancing the effect of cancer immunotherapy (29).
In the current study, we evaluated the effect of JSI-124 administration in mice bearing syngeneic GL261 glioma in the brain. Our data demonstrate that systemic JSI-124 administration promotes favorable functions of glioma-infiltrating immune cells, including production of a chemokine CXCL-10, and systemic (serum) IL-15 levels, thereby improving therapeutic effects of Tc1-cell adoptive transfer therapy.
Materials and Methods
Media, cytokines, and JSI-124
RPMI 1640, FBS, l
Abs and peptides
Purified mAbs against IL-4 (11B11) were obtained from BD Pharmingen. The following primary Abs for Western blot were purchased from Cell Signaling Technology: anti-phospho-Tyr 705 of STAT3 (D3A7), pan-STAT3 (7D1), anti-phospho-Tyr 701 of STAT1 (58D6), pan-STAT1 (polyclonal), and β-actin (13E5). The following fluorescent dye-conjugated Abs were obtained from BD Pharmingen: anti-CD11b (M1/70), anti-CD11c (HL3), anti-CD25 (PC61), anti-CD4 (L3T4), anti-CD40 (HM40-3), anti-CD8 (53-6.7), anti-CD80 (16-10A1), anti-CD86 (GL1), anti-Gr-1 (RB6-8C5), anti-I-Ab (AF6-120.1), anti-IFN-γ (XMG1.2), and anti-pSTAT3 (pSTAT3, C4). H-2Db human(h)gp10025-33 peptide (KVPRNQDWL) and H-2Kb OVA peptide (SIINFEKL) were synthesized in the University of Pittsburgh Peptide Synthesis Facility with >95% purity as indicated by analytical high-performance liquid chromatography and mass spectrometric analysis. Peptide was dissolved in DMSO at a concentration of 2 mg/ml and stored at −20°C until use. PE-H2-Db/hgp10025–33 tetramer was produced by National Institute of Allergy and Infectious Disease (NIAID) tetramer facility at the Emory University Vaccine Center (Altanta, GA).
The mouse (H-2b) GL261 glioma cell line was provided by Dr. Robert Prins (University of California, Los Angeles), and maintained in mouse complete medium (RPMI 1640 supplemented with 10% heat-inactivated FBS, 100 U/ml penicillin, 100 mg/ml streptomycin, and 10 μM l-glutamine) in a humidified incubator in CO2 at 37°C. These cells express gp10025–33 (30). In some experiments, GL261 cells were cultured in the presence of JSI-124 at indicated concentrations or DMSO vehicle control. Culture supernatants were harvested, filtered, and concentrated 10-fold using an Amicon Ultra Filter (Millipore). GL261-conditioned medium (GL261-CM) was prepared by mixing the concentrated medium and fresh medium at 1:9 ratio. The TAP2-deficient RMA-S mouse thymoma cell line was provided by Dr. Walter J. Storkus (University of Pittsburgh, Pittsburgh, PA). Cells were maintained in mouse complete medium in a humidified incubator in CO2 at 37°C.
C57BL/6 mice (H-2b), C57BL/6-background athymic mice, and OVA-specific TCR transgenic OT-1 mice (C57BL/6-background) were purchased from Taconic Farms. Pmel-1 mice (The Jackson Laboratory) are C57BL/6-background mice transgenic for a hgp10025–33-specific TCR, which cross-reacts with mouse (m) gp10025–33 (31). Animals were handled in the Animal Facility at the University of Pittsburgh per an Institutional Care and Use Committee-approved protocol.
Generation of bone marrow-derived DCs
The protocol used to generate DCs has been previously described (31, 32). In brief, bone marrow cells were cultured in complete medium supplemented with 10 ng/ml rmGM-CSF. Medium and supplements were replaced every 3 days. Cells were collected on day 6, followed by isolation of CD11c-positive cells using CD11c-microbeads (Miltenyi Biotec). Purity of CD11c-positive populations was >95% as determined by flow cytometry. These cells were then resuspended in control or GL261-conditioned media (CM) containing GM-CSF with/without 0.5 ng/ml JSI-124. After 24 h, supernatants were collected and used in subsequent experiments.
Generation of Tc1
Tc1 cells recognizing hgp10025–33 and mgp10025–33 were generated by the methods we described previously with slight modifications (6). In brief, CD8+ cells were enriched from Pmel-1 mouse-derived splenocytes (SPC) using CD8-microbeads (Miltenyi Biotec), and stimulated with the hgp25–33 peptide (5 μg/ml) in the presence of irradiated (3,000 rad) C57BL/6 SPC as feeder cells, 2 ng/ml rmIL-12, 1 ng/ml anti-IL-4 mAb, and 100 U/ml rhIL-2. Cells were restimulated under the same conditions at 48 h after the initial stimulation, and were harvested between days 9–12.
Brain tumor-bearing mice
Preparation of intracranial (i.c.) tumor-bearing mice was performed as previously described (10, 33). Using a Hamilton syringe (Hamilton Company), 1 × 105 GL261 cells in 2 μl PBS were stereotactically injected through an entry site at the bregma, 3 mm to the right of sagittal suture and 4 mm below the surface of the skull of anesthetized mice using a stereotactic frame (Kopf). JSI-124 was administered i.p. daily on days 14–17 (1 ng/g body weight/day). In some experiments, mice received an i.v. injection with 5 × 106 to 8 × 106 Tc1 cells on day 18. Animals were monitored daily after treatment for the manifestation of any pathologic signs. In some experiments, symptom-free survival was monitored as the primary endpoint, and in other experiments, treated mice were sacrificed on indicated days to evaluate immunological endpoints, such as brain infiltrating lymphocytes (BILs).
Isolation of BILs
BILs were isolated using the methods described previously (5, 6, 10). In brief, mice were sacrificed by CO2 asphyxia, then perfused through the left cardiac ventricle with PBS. Brains were mechanically minced and cells from each brain were resuspended in 70% Percoll (Sigma-Aldrich), overlayed with 37 and 30% Percoll, then centrifuged for 20 min at 500 × g. Enriched BIL populations were recovered at the 70–37% Percoll interface (34). For in vivo inhibition of CXCL-10, mice received i.p. injections with 100 μg/mouse anti-CXCL-10 mAb (1F11; provided from Dr. Andrew D. Luster, Harvard Medical School, Charlestown, MA; Ref. 35) at the same time T cells were transferred.
Western blot was performed as previously described (36). In brief, cells were lysed in lysis buffer (20 mM Tris-HCL, 150 mM NaCl, 1 mM Na2EDTA, 1 mM EGTA, 1% Triton, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM NA3VO4, 1 μm/ml leupeptin, and 1 mM PMSF). The protein content of cell lysate was separated on 7.5% polyacrylamide gels using SDS-PAGE and transferred to nitrocellulose membranes. The blots were blocked in 5% milk in TBS/Tween 20 (TBS/T) at room temperature for 1 h and proved overnight at 4°C with the primary Abs. Membranes were then washed with TBS/T, incubating with corresponding secondary Abs (Santa Cruz Biotechnology), and then washed again with TBS/T. Specific bands were visualized using the chemiluminescence reagent kit (ECL and ECL plus; Amersham Biosciences).
Cytokine and chemokine release assay
In vitro chemotaxis assays
Cell migration was evaluated using a 96-well microchemotaxis chamber assay (ChemoTx; Neuroprobe). In brief, Tc1s were washed in PBS and resuspended in RPMI 1640 medium (1 × 106 cells/ml). The lower well of each chamber was filled with 30 μl CM. A 3-μm pore filter was layered over the wells, and the T cell suspension (5 × 104 in assay medium) was seeded on the top surface of the filter. The chemotaxis chamber was incubated at 37°C, 100% humidity, 5% CO2 for 2 h. The nonmigrated cells were washed from the top surface with PBS and the chamber was centrifuged for 5 min at 1,500 rpm. Cell numbers in bottom wells were microscopically measured.
Survival data were analyzed by log rank test. Student’s t test was used for comparison of two samples with unequal variances. One-way ANOVA with Holm’s post hoc test was used for comparing means of three or more variables.
Therapeutic effects of systemic JSI-124 administration in tumor-bearing mice depend on host adaptive immunity
Cultured GL261 glioma cells expressed high levels of pSTAT3, which was rapidly and effectively dephosphorylated by treatment with JSI-124 for 2 h in a concentration-dependent manner, as demonstrated by Western blot analyses (Fig. 1⇓a). Consistent with a previous study with human cancer cells (37), inhibition of pSTAT3 in cultured GL261 cells with JSI-124 suppressed their growth (data not shown). We thus sought to determine whether JSI-124 could also inhibit the growth of i.c. GL261 glioma in vivo. With regard to the dose of JSI-124 for in vivo treatment, we chose 1 ng/g body weight/day JSI-124 on days 14–17 following the tumor inoculation because 0.5 ng/ml JSI-124 efficiently inhibited pSTAT3 in vitro (Fig. 1⇓a), and a similar dose has been shown to induce effective antitumor effects in vivo (27). C57BL/6 mice or C57BL/6-background athymic mice bearing i.c. GL261 tumor received daily i.p. injections of JSI-124 or control vehicle. As depicted in Fig. 1⇓b, all wild-type mice receiving control vehicle died by day 50, whereas the treatment with JSI-124 significantly improved the survival of mice compared with the control group. In contrast, treatment with JSI-124 in athymic nude did not demonstrate any therapeutic benefit, suggesting that the in vivo antitumor effect by JSI-124 requires a host adaptive immune response, such as antitumor T cells. These data also suggest that the therapeutic effect of JSI-124 at the used dose-level and schedule may not induce direct growth-inhibitory or cytotoxic effects on GL261 cells in vivo.
Soluble factors from GL261 glioma cells induce phosphorylation of STAT3 in immune cells, which is efficiently reversed by JSI-124
To evaluate the effects of JSI-124 on the pSTAT3 status of immune cells that are exposed to glioma-derived soluble factors, SPC freshly harvested from C57BL/6 mice were cultured for 24 h in the presence or absence of GL261-CM and then treated for 2 h with 0.5 ng/ml JSI-124. As shown in Fig. 2⇓a, GL261-CM induced phosphorylation of STAT3 in SPC, which was inhibited by JSI-124. SPCs derived from C57BL/6-background athymic mice exhibited similar pSTAT3 responses as regular C57BL/6-derived SPC following treatment with GL261-CM and/or JSI-124 (data not shown). As it has been demonstrated by others that inhibition of STAT3 may lead to reciprocal up-regulation of pSTAT1 (22, 38, 39), we evaluated the status of pSTAT1 in C57BL/6 mouse-derived SPC, and found JSI-124 induced pSTAT1 in both GL261-CM-treated and nontreated SPC (Fig. 2⇓b). Flow cytometric analyses of CD4-, CD8-, or CD11b-gated populations with intracellular staining for pSTAT3 revealed rapid reversal of pSTAT3 by JSI-124 in each population (Fig. 2⇓c). These data thus indicate that GL261 glioma cells induce phosphorylation of STAT3 in immune cells via soluble factors and that JSI-124 efficiently dephosphorylates pSTAT3 in these cells, while inducing phosphorylation of STAT1.
JSI-124 promotes maturation of tumor-conditioned CD11c+ DCs and activation of tumor-conditioned cytotoxic T cells
Our data in Fig. 1⇑ suggested a pivotal role of adaptive immunity in the therapeutic effects induced by JSI-124. As our previous studies have demonstrated critical roles of DCs in CNS tumor-microenvironment (6, 32, 33), we then addressed the effects of JSI-124 on phenotypes and functions of DCs that were exposed to GL261-CM. Day 6 CD11c+ bone marrow-derived DCs were cultured with GL261-CM alone or GL261-CM and JSI-124 for 24 h. As demonstrated in Fig. 3⇓a, treatment with JSI-124 up-regulated surface expression of MHC class II, CD40, CD80, and CD86 on CD11c+ DCs in the presence of GL261-CM. Furthermore, DC production of IL-12p70, which was markedly down-regulated by GL261-CM, was significantly improved by JSI-124 treatment (Fig. 3⇓b).
Next, we addressed whether JSI-124 could restore effector functions of cytotoxic T cells that might be affected by GL261-CM. To this end, we purified CD8+ T cells from Pmel-I mice and cultured them for 3 days with GL261-CM and/or JSI-124 in the presence of 100 U/ml rhIL-2 and hgp10025–33-pulsed, irradidated feeder cells. Then, we evaluated the production of IFN-γ in culture supernatants, an essential cytokine of type-1 immune response. As shown in Fig. 3⇑c, production of IFN-γ, which was significantly diminished when T cells were cultured with GL261-CM, was recovered to levels comparable to the control nontumor conditioned group when these cells were treated with JSI-124. In parallel to IFN-γ production, JSI-124 recovered Ag-specific killing ability of T cells that was suppressed by GL261-CM (Fig. 3⇑d). These results collectively indicate that JSI-124 reverts suppressed functions of glioma-conditioned APCs (i.e., DCs) and effector CTLs.
JSI-124 enhances CXCL-10 production by DCs and GL261 cells in vitro, promoting the attraction of Tc1
We sought whether JSI-124 would promote immunological surveillance against GL261 glioma particularly in the context of T cell trafficking to the tumor. We have previously demonstrated the pivotal role of a type-1 chemokine CXCL-10/IP-10 in efficient CNS tumor-homing of Ag-specific Tc1 (6). JSI-124 significantly enhanced CXCL-10 secretion by in vitro-cultured GL261 cells (Fig. 4⇓a), and restored CXCL-10 production from DCs conditioned with GL261-CM (Fig. 4⇓b). We then evaluated whether culture supernatants from JSI-124-treated GL261 cells or DCs could indeed attract Tc1 cells that express CXCR3, a receptor for CXCL-10 (data not shown). Consistent with Fig. 4⇓a, Tc1 exhibited enhanced chemotaxis toward JSI-124-treated GL261-CM compared with untreated GL261-CM (Fig. 4⇓c). Moreover, chemotaxis assays with supernatants from DCs demonstrated a mirror-image profile with CXCL-10 (Fig. 4⇓d). These data indicate that the effects of JSI-124 on glioma cells and DCs include up-regulation of CXCL-10 production, which promotes attraction of effector Tc1.
Systemic administration of JSI-124 inhibits phosphorylation of STAT3 in BILs and reverts their suppressive immunological phenotype
We next sought to address the effects of in vivo treatment with JSI-124 on immunological environment of i.c. gliomas. We isolated BILs from mice bearing i.c. GL261 after treatment with or without JSI-124 (Fig. 5⇓a). BILs isolated from JSI-124-treated mice exhibited reduced pSTAT3 levels in each of CD4-, CD8-, and CD11b-gated populations when compared with BILs derived from control mice. JSI-124 treatment also enhanced the expression of surface maturation markers, MHC class II, CD40, CD80, and CD86 on tumor-infiltrating CD11c+ BILs (Fig. 5⇓b). Moreover, in vivo JSI-124 treatment reduced CD11b+/Gr1+ myeloid suppressor and CD4+/CD25+ regulatory T cells in BILs (Fig. 5⇓c). When BILs were stimulated in vitro with LPS for 24 h, BILs obtained from JSI-124-treated mice produced higher levels of IL-12p70 (Fig. 5⇓d) and CXCL-10 (Fig. 5⇓e) compared with BILs derived from control mice. Taken together, these data indicate that in vivo administration of JSI-124 inhibits phosphorylation of STAT3 in BILs and reverts their suppressive profiles to type-1 cytokine/chemokine production profiles.
JSI-124 administration promotes the efficacy of adoptive transfer therapy with GL261-specific Tc1
Promotion of type-1 cytokine chemokine/profiles in the i.c. tumor environment by JSI-124 lead us to hypothesize that this strategy could be efficiently combined with i.v. Tc1 transfer therapy. C57BL/6 mice bearing i.c. GL261 glioma received daily i.p. injections of JSI-124 or control vehicle on days 14–17 following the tumor inoculation. On day 18, Pmel-I-derived Tc1 (gp10025–33 specific) or control Tc1 recognizing an irrelevant Ag (OVA257–264 specific) were adoptively transferred via tail vein. As depicted in Fig. 6⇓a, all mice receiving control vehicle and irrelevant Tc1 died by day 37. Treatment with JSI-124 improved the survival of mice even when they received irrelevant Tc1, consistent with data in Fig. 1⇑b demonstrating a role of JSI-124 alone. Nevertheless, all these mice eventually died by day 54. The i.v. transfer of Pmel-I-derived Tc1 without JSI-124 resulted in prolonged survival of mice compared with the transfer of irrelevant Tc1 without JSI-124 (p = 0.0323). Moreover, when JSI-124 pretreatment was followed by i.v. transfer of Pmel-I-derived Tc1, mice exhibited the most prolonged survival among all groups, with three of ten mice surviving for longer than 100 days.
BIL analyses following i.v. transfer of Pmel-I-derived Tc1 revealed that administration of JSI-124 enhanced the tumor-homing of gp100-reactive/CD8+ cells (Fig. 6⇑b). In contrast, i.v. infusion of Ag-irrelevant Tc1 cells did not increase gp100-reactive/CD8+ BILs even with JSI-124-treatment (Fig. 6⇑b), suggesting that the increased gp100-reactive/CD8+ BILs mostly represented i.v. transferred Pmel-I-derived Tc1. The number of IFN-γ-producing/CD8+ BILs also increased by JSI-124 pretreatment (Fig. 6⇑c), suggesting improved Tc1 functions in situ.
As shown in Fig. 4⇑, enhanced in vitro T cell chemotaxic activity by JSI-124 was associated with elevated CXCL-10 levels. To address whether Tc1 trafficking toward the brain tumor site was mediated by CXCL-10, we performed in vivo CXCL-10-blocking experiments using a neutralizing anti-CXCL-10 mAb in GL261-bearing mice receiving JSI-124 and i.v. transfer of Pmel-I Tc1. As shown in Fig. 6⇑, d and e, CXCL-10-blockade almost completely abrogated CNS tumor trafficking of gp100-reactive/CD8+ BILs (Fig. 6⇑d), and prolongation of survival (Fig. 6⇑e), suggesting that CXCL-10 plays a critical role in the tumor homing of gp100-specific Tc1. Taken together, these data suggested that systemic inhibition of STAT3 signaling by JSI-124 enhanced CNS tumor homing and therapeutic efficacy of i.v. transferred Ag-specific Tc1 in a CXCL-10-dependent manner.
JSI-124 administration prolongs systemic IL-15 production to promote the persistence of Tc1
It has been demonstrated that increased access to homeostatic cytokines, such as IL-7 and IL-15, is crucial for effective antitumor response following T cell transfer to tumor-bearing hosts (40, 41). We thus hypothesized that the improved antitumor effect by JSI-124-treatment in the i.v. Tc1 transfer therapy might be associated with systemic up-regulation of IL-7 and/or IL-15, thereby improving the persistence of i.v. transferred Tc1 in the hosts. C57BL/6 mice with or without i.c. GL261 glioma received daily i.p. injections of JSI-124 or control vehicle daily on days 14–17, followed by i.v. transfer of Pmel-I-derived Tc1 on day 18. On day 35, significantly elevated serum IL-15 levels were detected in glioma-bearing mice treated with JSI-124 compared with corresponding mice without JSI-124 treatment (Fig. 7⇓a), although presence of i.c. glioma appears to reduce the IL-15 levels compared with nontumor bearing mice treated with JSI-124 (p = 0.0397). The i.p. administration of JSI-124 in athymic mice also induced elevated serum IL-15 levels (data not shown), suggesting that non-T cell populations in the host are the major source of IL-15 (42). IL-7 production, in contrast, was not affected by JSI-124 administration (data not shown). Flow cytometric analyses of SPCs revealed elevated proportions of gp100-reactive/CD8+ cells in JSI-124-treated mice compared with control mice (Fig. 7⇓b), suggesting that the treatment with JSI-124 promoted the persistence of i.v. transferred Tc1.
The current study indicates that inhibition of STAT3 by systemic JSI-124 administration promotes type-1 phenotype and function of glioma-infiltrating immune cells, thereby enhancing the efficacy of adoptive transfer therapy using Tc1 cells in a CXCL10-dependent manner. JSI-124 also induced phosphorylation of STAT1 in SPC and elevation of systemic (serum) IL-15 levels. These events may have contributed to the prolonged persistence of adoptively transferred Tc1 in mice bearing i.c. GL261 gliomas.
STAT3 was constitutively phosphorylated in cultured GL261 glioma cells, which secreted soluble factors capable of inducing phosphorylation of STAT3 in immune cells. Although we have not identified these factors, we postulate that similar mechanisms that have been reported for human cancers, including gliomas, operate in the GL261 murine glioma model. With regard to phosphorylation of STAT3 in human cancer cells, genetic alterations, such as amplification of epidermal growth factor receptor, lead to deregulation of multiple intracellular signal transduction pathways, including STAT3 (43, 44). Recent studies have demonstrated that protein inhibitor of activated STAT3 and suppressors of cytokine signaling 3, both of which are intrinsic inhibitors of STAT3, are often impaired in cancer cells (45, 46, 47). Cancer cells with activated STAT3 produce STAT3-activating cytokines, such as VEGF and IL-10 (16), and indeed, glioma cells produce these cytokines at high levels (12, 26). These cytokines activate STAT3 in not only tumor-infiltrating immune cells, but likely glioma cells themselves via an autocrine fashion. Our data from both in vitro and in vivo experiments indicate JSI-124 effectively reverses pSTAT3 in the GL261 glioma and its immunological environment. Although penetration of therapeutic agents is often limited to the CNS and CNS tumors (1, 48), our data suggest that systemically administered JSI-124 can efficiently modulate the immunological environment of CNS tumors. Further studies with varying doses and/or timing of JSI-124 administration are warranted to determine whether JSI-124 can directly inhibit the growth of glioma cells in the CNS while inducing efficient type-1 antitumor immune response.
Our data characterizing cytokine/chemokine production from BILs suggest that JSI-124 promotes type-1 immunity, represented by increased IL-12 p70 and CXCL-10 production. Although activation of DCs (28, 29) and up-regulation of CXCL-10 (21) by STAT3 inhibition have already been reported in other models, our data demonstrate a dynamic paradigm change of immunological microenvironment in CNS tumors.
Our previous study has indicated that direct intratumoral delivery of DC ex vivo engineered to express IFN-α induces similar effects in terms of CXCL-10-dependent CNS-tumor homing of Tc1 cells (6). As IFN-α signaling promotes STAT1 pathway to exert its biological properties, such as induction of CXCL-10 (49), data from our current and previous studies collectively suggest a possibility that in vivo inhibition of STAT3 may lead to reciprocal up-regulation of STAT1, as shown in previous studies by others (22, 38, 39). With this regard, a recent study in patients with melanoma has demonstrated that high-dose IFN-α therapy up-regulates pSTAT1, whereas it down-regulates pSTAT3 in melanoma cells and lymphocytes (50). In this study, higher pSTAT1/pSTAT3 ratios in tumor cells following the neoadjuvant treatment with IFN-α were associated with longer overall survival, suggesting that the pSTAT1/pSTAT3 ratio in tumor cells may serve as a useful predictor of therapeutic effect. We are currently conducting a vaccine trial in patients with malignant glioma using GAA-loaded DCs in combination with a TLR-3 ligand, poly-ICLC, which is a potent IFN-inducer (10). Hence, a combinatorial strategy of poly-ICLC and STAT-3 inhibition may be warranted in our future studies to augment the pSTAT1/pSTAT3 ratio in the CNS tumor microenvironment.
Based on our data with in vitro CTL assays using Tc1 conditioned with GL261-CM, JSI-124 promoted the recovery of their CTL activity. Although it has been demonstrated that CD8+ T cells in mice with a Stat3-ablated hemopoietic system can produce elevated levels of IFN-γ upon exposure to tumor cells (22), it is intriguing to determine how JSI-124 promoted the cytotoxic activity of Tc1 cells in our experiments. The granule exocytosis pathway, including the membrane-disruptive protein perforin, of CTL is crucial for the delivery of granule proteases (granzymes) into the target cell and its destruction through apoptosis (51). Phosphorylation of STAT3 in CTLs may down-regulate this pathway, and hence, it is postulated that inhibition of pSTAT3 may restore function of these effector molecules. Experiments are underway to evaluate this hypothesis.
With regard to key factors that correlate with clinical efficacy of T cell adoptive transfer therapy, in a pilot clinical trial in patients with metastatic melanoma receiving nonmyeloablative chemotherapy followed by adoptive transfer of activated tumor-reactive T cells (13), a significant correlation was found between tumor regression and the degree of persistence in peripheral blood of adoptively transferred T cell clones (52). Our data evaluating numbers of transferred Pmel-I-derived cells in the spleen of treated mice have demonstrated that systemic JSI-124 administration enhances serum IL-15 levels and the persistence of adoptively transferred T cells. IL-15 promotes the life span of naive as well as memory CD8+ T cells (53, 53, 54), and protects type-1 T cells from T regulatory cell-mediated inhibition (42, 55). Administration of low-dose IL-15 has been shown to promote the persistence of adoptively transferred tumor-specific T cells in murine tumor models (56, 57). A question remains as to how inhibition of STAT3 led to the systemic up-regulation of IL-15. Based on our in vitro data indicating JSI-124 induced phosphorylation of STAT1 in SPC, it is postulated that activated STAT1 may be responsible for production of IL-15 in vivo, as demonstrated by others in different disease models (58, 59). In addition, blockade of IL-15 in our current model (e.g., by use of IL-15 specific blocking mAb) would directly determine the significance of IL-15 induction in the therapeutic effect. As a logical extension of these studies, when clinical trials are designed based on the current study, serum IL-15 may be used as an immunological biomarker.
The current study has significant scientific implications for both basic and translational aspects. Although the CNS and CNS tumors have been often described as immunologically privileged sites, inhibition of pSTAT3 in the CNS by systemic administration of JSI-124 may represent a key to overcome suppressive immunological status of the CNS tumors. Although JSI-124 improved the CNS tumor homing and function of tumor-specific CTLs, it is noteworthy that we have not observed any sign of autoimmune encephalitis upon necropsy of treated mice (data not shown), suggesting safety of this approach.
Although the direct biochemical target of JSI-124 remains unknown, the effect of this compound is highly selective for STAT3 (27). Promising immunomodulatory and/or antitumor effects have also been demonstrated by other STAT3 inhibitors. CPA-7, which is a platinum compound that specifically disrupts STAT3 signaling, has exhibited antitumor effects associated with activation of immune effector cells (60). More recently, a novel synthetic STAT3 inhibitor, WP1066, has been shown to efficiently induce apoptotic death in malignant glioma cells (61) and reverse tolerance in immune cells derived from patients with malignant glioma (62). Implementation of early phase clinical trials for patients with gliomas awaits availability of clinical-grade small-molecule STAT3 inhibitors.
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 grants from National Institutes of Health/National Institute of Neurological Disorders and Stroke (P01 NS40923) (to H.O. and I.F.P.), from National Institutes of Health/National Cancer Institute (1P01 CA100327) (to H.O.), and a grant from the James S. McDonnell Foundation (to H.O.).
↵2 Address correspondence and reprint requests to Dr. Hideho Okada, G12a Research Pavilion of the Hillman Cancer Center, University of Pittsburgh Cancer Institute, 5117 Centre Avenue, Pittsburgh, PA 15213. E-mail address:
↵3 Abbreviations used in this paper: GAA, glioma-associated Ag; Tc1, type 1 CTL; VEGF, vascular endothelial growth factor; rm, mouse recombinant; rh, human recombinant; CM, conditioned media; SPC, splenocyte; i.c., intracranial; BIL, brain infiltrating lymphocyte; poly-ICLC, polyinosinic-polycytidylic acid stabilized with polylysine and carboxymethyl cellulose.
- Received August 20, 2007.
- Accepted December 4, 2007.
- Copyright © 2008 by The American Association of Immunologists