Novel TGF-β Antagonist Inhibits Tumor Growth and Angiogenesis by Inducing IL-2 Receptor-Driven STAT1 Activation

Claudia Penafuerte, Norma Bautista-Lopez, Manaf Bouchentouf, Elena Birman, Kathy Forner and Jacques Galipeau
J Immunol June 15, 2011, 186 (12) 6933-6944; DOI: https://doi.org/10.4049/jimmunol.1003816

Abstract

Carcinoma derived TGF-β acts as a potent pro-oncogenic factor and suppresses antitumor immunity. To antagonize TGF-β–mediated effects in tandem with a proinflammatory immune stimulus, we generated a chimeric protein borne of the fusion of IL-2 and the soluble extracellular domain of TGF-βR II (FIST). FIST acts as a decoy receptor trapping active TGF-β in solution and interacts with IL-2–responsive lymphoid cells, inducing a distinctive hyperactivation of STAT1 downstream of IL-2R, which in turn promotes SMAD7 overexpression. Consequently, FIST-stimulated lymphoid cells are resistant to TGF-β–mediated suppression and produce significant amounts of proinflammatory cytokines. STAT1 hyperactivation further induces significant secretion of angiostatic CXCL10. Moreover, FIST upregulates T-bet expression in NK cells promoting a potent Th1-mediated antitumor response. As a result, FIST stimulation completely inhibits pancreatic cancer (PANC02) and melanoma (B16) tumor growth in immunocompetent C57BL/6 mice. In addition, melanoma cells expressing FIST fail to form tumors in CD8−/−, CD4−/−, B cell-deficient (μMT), and beige mice, but not in NOD-SCID and Rag2/γc knockout mice, consistent with the pivotal role of FIST-responsive, cancer-killing NK cells in vivo. In summary, FIST constitutes a novel strategy of treating cancer that targets both the host’s angiogenic and innate immune response to malignant cells.

An extensive number of studies demonstrate that TGF-β acts as a potent tumor promoter in established carcinoma (1, 2). Cancer-derived TGF-β drives malignant progression by constitutively inducing epithelial to mesenchymal transition and tumor-associated angiogenesis and by suppressing tumor immunity, the sum of which effects are to promote tumor growth and metastasis (1, 3–6). There are three different TGF-β isoforms, TGF-β1, 2, 3. Among them, TGF-β1 is the most frequently overexpressed by carcinomas (7). In mammalian cells, TGF-β signals through heteromeric complexes of type I (activin receptor-like kinases, also known as TβRI) and type II (TβRII) transmembrane serine/threonine kinase receptors (8). Upon ligand binding, the constitutively active TβRII phosphorylates the type I receptor (ALK), which propagates signals into the cells by phosphorylating receptor-regulated SMADs (i.e., SMAD2 and SMAD3). Phosphorylated receptor-regulated SMADs form a heteromeric complex with the comediator SMAD4 (9) and translocate to the nucleus where this complex cooperates with transcription activators and repressors to regulate the expression of TGF-β target genes (10). Inhibitory SMADs (SMAD6 and SMAD7) act as negative regulators of TGF-β signaling pathway by preventing the interaction of TGF-βR complex with receptor-regulated SMADs and promoting the dephosphorylation and degradation of activated receptors (11, 12). As proangiogenic factors, TGF-β family members play an essential role in vasculogenesis and tumor angiogenesis. TGF-β regulates the expression of various extracellular matrix components that play a key role in both the initiation and resolution phase of angiogenesis (13–16). Consistent with this notion, we hypothesize that the blockade of TGF-β signaling combined with cytokine-driven immune activation will couple anti-angiogenesis to an effective immune antitumor response, resulting in potent anticancer properties. In particular, IL-2 is a cytokine that can promote an innate antitumor response due to its effectiveness at inducing locoregional tumor rejection, acting as autocrine factor for T cells and supporting the development of cytotoxic T cells, stimulating NK cells proliferation and cytolytic activity (17–19). In addition, IL-2 was approved by the U.S. Food and Drug Administration in 1998 for immunotherapy of melanoma (20). We have found in a preclinical model of B16 melanoma that IL-2–driven anticancer activity is severely curtailed by tumor-derived active TGF-β. Several studies have reported that TGF-β inhibits the phosphorylation and activation of components of the JAK/STAT cascade downstream of IL-2R (21). Moreover, TGF-β inhibitory activity occurs at the nuclear level on a subset of IL-2 target genes including c-Myc, cyclin D2, and cyclin (22). Therefore, the blockade of TGF-β activity markedly enhances immunotherapy (23). Several approaches have been used to neutralize tumor-derived active TGF-β in vivo (24–30). In particular, soluble type II TGF-βR (sTβRII) treatment in mice bearing tumors was shown to significantly decrease tumor metastasis (3, 26). We therefore speculated that IL-2 could serve as a rational cytokine partner of TβRII as part of a “sword and shield” pharmaceutical strategy. To test this concept, we created a novel recombinant protein fusion protein composed of IL-2 fused to the extracellular domain of TβRII and here demonstrate its biochemical effects on NK cells leading to potent antitumor activity in vivo.

Materials and Methods

Mice

B cell-deficient (μMT), NK, CD4, CD8 T cell, IFN-γ knockout (KO) mice, beige mice (C57BL/6J-bg), C57BL/6, BALB/c immunocompetent, CXCR3 KO, NOD-SCID mice (The Jackson Laboratory, Bar Harbor, ME), and Rag2/γc KO mice (Taconic, Hudson Valley, NY) were female, 6 to 8 wk old. STAT1 KO mice were generously donated by Joan Durbin (Lady Davis Institute for Medical Research) (31).

FIST fusokine design, expression, and functionality

The mouse IL-2 cDNA (Invivogen, San Diego, CA) was modified by removing the 3′ nucleotide encoding the STOP codon, and subsequently cloned in frame with mouse TGFβ receptor II ectodomain cDNA (Invivogen) to generate the cDNA for mouse IL-2/sTβRII fusion protein (also known as fusion of IL-2 and the soluble extracellular domain of TGF-βR II; FIST). The stop codon and poly A were added at the C-terminal region. The vectors encoding FIST or control (IL-2 and sTβRII) were used to transfect 293T cells. The supernatants of transfected cells were collected after 48 h, concentrated with centricon plus-70 (Millipore, Billerica, MA), and the molar concentration of FIST and IL-2 were quantified by IL-2 ELISA, whereas sTβRII protein was quantified by Western blot using a concentration versus OD curve of recombinant mouse TβRII–mouse Fc as reference (R&D Systems, Minneapolis, MN). For all the experiments, the mock or RPMI 1640 media contain equal volume of concentrated media conditioned by non-transfected 293T cells. For the proliferation assay, 104 CTLL-2 cells (IL-2 dependent) per well were cultured in a 96-well plate and treated with 5 pmol FIST or controls for 72 h. Cell proliferation was assessed by MTT uptake. FIST and TGF-β complexes were precipitated using TGF-β1–specific Abs bound to protein G beads, and the precipitated FIST protein was detected by Western blotting with IL-2–specific Abs. Infectious retroparticles encoding FIST were generated with 293-GP2 packaging cells (Clontech, Mountain View, CA) and used to modify genetically C57BL/6-derived B16F0 melanoma and pancreatic cancer PANC02 (ductal adenocarcinoma of the pancreas) cell lines. CTLL-2 cells were maintained in RPMI 1640 supplemented with T-STIM with Con A (rat IL-2 culture supplement from Becton Dickinson) and 10% FBS; human NK cells (NK92) were maintained in α-MEM without ribonucleosides and deoxyribonucleosides but with 2 mM l-glutamine and 1.5 g/l sodium bicarbonate, supplemented with 0.2 mM inositol, 0.1 mM 2-mercaptoethanol, 0.02 mM folic acid, 100–200 U/ml recombinant IL-2 and 50 U/ml penicillin and streptomycin, adjust to a final concentration of 12.5% horse serum and 12.5% FBS (Wisent Technologies, Rocklin, CA). B16F0 and PANC02 cell lines were maintained in DMEM medium supplemented with 10% FBS and 50 U/ml penicillin and streptomycin (Wisent Technologies).

Immune cell isolation and cytokine production

Enriched T, B, and NK cell populations were obtained from splenocytes of immunocompetent or STAT1 KO mice by magnetic separation according to the manufacturer’s recommendation (Stemcell Technologies, Vancouver, Canada). The purity of each population was assessed by flow cytometry (for B cells 94%, NK cells 96%, and T cells 96%) using specific conjugated Abs for cell markers (BD Biosciences, San Diego, CA). Human blood was drawn from healthy donors, after informed consent had been obtained, into heparin- or citrate-coated tubes (BD Biosciences, San Jose, CA), and PBMCs were isolated by density centrifugation. Cytokine concentrations were quantified by ELISA kits (eBioscience, San Diego, CA; and R&D Systems).

Cell signaling and receptor expression

CTLL-2 cells (5 × 106) were stimulated with 5 pmol FIST or controls for 20 min before being lysed and probed by Western blot with rabbit phospho-specific and total SMAD2, SMAD3, STAT1, STAT3, and STAT5 (Cell Signaling Technology, Danvers, MA) Abs. CTLL-2 cells were also stimulated with FIST or controls for 2 h to detect SMAD7 expression with rabbit-specific Abs (Santa Cruz Biotechnology, Santa Cruz, CA). For T-bet expression, cytokine-stimulated NK cells were treated with brefeldin A for 4 h, fixed, permeabilized, and stained with T-bet–conjugated Ab or isotype control for 1 h. For CXCR3 and IL-2Rβ expression, stimulated NK cells were incubated with FcR III/II blocking Ab for 30 min and labeled with conjugated Abs specific for CXCR3 and IL-2Rβ for 1 h. The mean fluorescence intensity was measured by flow cytometry, and the data were analyzed using CellQuest software (Becton Dickinson).

Immune cell infiltration in the tumor site

To analyze the immune cell types infiltrated in the tumor site, 106 genetically modified B16 cells expressing FIST or controls were mixed with 500 μl Matrigel (BD Biosciences) at 4°C and injected s.c. in immunocompetent C57BL/6 mice. Implants were surgically removed 6 d after implantation and enzymatically dissociated as reported previously (32). Infiltrated cells were collected and incubated with anti-FcR III/II Ab for 30 min and with specific cell type or isotypic control Abs for 1 h at 4°C. The expression of surface markers was determined as previously described.

Endothelial cell isolation, proliferation, migration, and in vitro and in vivo angiogenesis assays

Livers of immunocompetent mice were digested with collagenase for 40 min at 37°C to prepare a cell suspension. Cells were labeled with PE-conjugated CD31-specific Abs to isolate endothelial cells (EC) by positive selection of PE-labeled cells (Stemcell Technologies). For the in vitro capillary-like structure formation assay, 5 × 103 EC were seeded per well in a 96-well plate previously coated with a type IV collagen matrix (Chemicon, Billerica, MA). EC were incubated at 37°C for 24 h in the presence of conditioned media (CM) by NK cells previously stimulated with 5 pmol FIST or control for 72 h (CM NK cells). A numerical score was assigned to each condition according to the degree of angiogenesis progression based on the number and size of polygons formed, capillary thickness, and cell alignment and fusion, as previously reported (33). For the proliferation assay, 2 × 103 cells were cultured in a 0.1% gelatin-coated 96-well plate for 6 d in a low-serum EC media supplemented with EC growth supplement and epidermal growth factor (BD Biosciences) in the presence of CM by stimulated NK cells. Cell number was determined using the ViaLight Plus kit (Lonza, Rockland, ME). For migration assay, 24-well cell invasion assay (Millipore) was used to examine the migration of EC according to the manufacturer’s protocol. Briefly, this modified Boyden chamber is composed of a well (lower compartment) containing an insert (upper compartment). EC (8 × 103) were suspended in 0.5 ml serum-free medium and loaded into upper chambers of the Transwell. Lower chambers were loaded with 1% RPMI 1640 medium in presence or absence of FIST, IL-2, sTβRII, or IL-2 plus sTβRII-stimulated NK cell conditioned medium (NK cells were stimulated with 5 pmol FIST or control). After a 24-h incubation period at 37°C, cells remaining on the upper surface of the insert (nonmigrated cells) were removed gently and placed into a sterile 24-well plate containing a cell detachment solution for 30 min at 37°C. Dislodged cells were treated with a lysis buffer and incubated with the CyQuant GR dye for 15 min at room temperature. The mixture was transferred to a 96-well plate, and fluorescence was read at 480 nm. Invaded cell number was determined by running a fluorescent versus cell dose curve using 8000, 4000, 2000 EC (n = 3). To determine FIST angiostatic properties in vivo, 106 genetically modified B16 cells expressing FIST or controls were mixed with 500 μl Matrigel (BD Biosciences) at 4°C and injected s.c. in immunocompetent C57BL/6 mice. Implants were surgically removed 20 d later, and histological sections of the tumors were labeled with red fluorescence CD31 Abs to detect the formation of tumor-derived blood vessels.

Statistical analysis

We used the two-tailed unpaired Student t test to compare two experimental groups, Dunnett’s multiple comparisons to compare three or more test groups, and the log-rank test to compare two survival distributions.

Results

Generation and characterization of murine IL-2/sTβRII fusion protein: FIST

We generated a plasmid construct encoding for the fusion of IL-2 and the ectodomain of TβRII splicing variant B (sTBRIIB, from aa Leu 9 to Asp 184). The fusion transgene cDNA encodes for a single polypeptide chain of 328 aa (Fig. 1A) that migrates as an ∼55 kDa protein in SDS-PAGE under reducing conditions (Fig. 1B, 1C). In a coimmunoprecipitation assay, FIST precipitated active TGF-β1 (Fig. 1D), indicating that FIST acts as a decoy receptor trapping active TGF-β. As predicted, IL-2 as part of the fusion protein preserves the ability to recognize and bind IL-2R and induce the proliferation of IL-2–responsive cells (CTLL-2, Fig. 1E). The analysis of activation status of transcription factors downstream of TβRII (SMAD3 and SMAD2) indicate that FIST exerts a dominant negative effect on TGF-β canonical pathway not only by impairing the phosphorylation of SMAD2 and SMAD3 but also by inducing SMAD7 expression in CTLL-2 (Fig. 1F). The analysis of activation of transcription factors downstream of IL-2R signaling pathway indicates that FIST induces similar STAT5 phosphorylation levels compared with equimolar concentrations of controls but distinctive STAT3 and STAT1 activation levels. In particular, STAT1 is more greatly phosphorylated by FIST than by equimolar concentrations of controls (Fig. 1G). With regard to the activation levels of other signaling pathways downstream of IL-2R such as MAPK and PI3K/AKT pathways, no differences were observed between FIST and equimolar concentrations of controls (data not shown). FIST acts through the IL-2R, as blockade of IL-2R on CTLL-2 cells inhibits STAT5 phosphorylation (Supplemental Fig. 1A). Lymphocytes from STAT1 KO mice were used to investigate whether FIST induces SMAD7 expression via STAT1 activation. As a result, SMAD7 expression was significantly abrogated compared with lymphocytes from immunocompetent mice, yet we observed a preserved inhibition of SMAD2 phosphorylation compared with the controls (Fig. 1H). Based on the cross-reactivity between human and mouse IL-2 for its specific receptor, we tested the ability of mouse FIST to induce IFN-γ production by human immune cells, as well as its ability to antagonize TGF-β canonical pathway in human NK cells. FIST-stimulated human immune cells also produce significantly higher amounts of IFN-γ than that induced by equimolar concentrations of cytokine control. However, the levels of IFN-γ induced by FIST in human cells were lower than that produced by mouse immune cells (Supplemental Fig. 1B). Similar results were obtained with regard to Smad2 inhibition by mouse FIST-mediated Smad7 expression in human NK cells (Supplemental Fig. 1C). This discrepancy is possibly due to a reduced affinity of mouse IL-2 for human IL-2R (34).

FIGURE 1.
FIGURE 1.

Generation and characterization of FIST. A, Amino acid sequence of FIST. B, Western blotting of CM of 293T cells transfected with vectors encoding FIST or controls and protein detection with sTβRII-specific Ab. C, FIST protein detection by Western blotting with IL-2–specific Ab. D, FIST and active TGF-β1 complexes precipitated with TGF-β1–specific Abs bound to protein G beads and detected by Western blotting with IL-2–specific Abs. E, The bioactivity of IL-2 as part of FIST determined by in vitro CTLL-2 proliferation based on MTT uptake; cell number was quantified by an OD versus cell dose curve. The data are representative of three independent experiments performed in triplicate (n = 3 per group; data are shown as mean ± SD). F, Western blotting for phospho-specific and total SMAD2, SMAD3, and SMAD7 proteins from cell lysates of 5 × 106 CTLL-2 cells previously stimulated with 5 pmol FIST or controls for 20 min (for SMAD2 and SMAD3 phosphorylation) or 2 h (for SMAD7 protein expression). G, Western blotting for phospho-specific and total STAT1, STAT3, and STAT5 proteins from cell lysates of 5 × 106 CTLL-2 cells previously stimulated with 5 pmol FIST or controls for 20 min. H, Similar experiments as those of Fig. 1F were performed with unfractionated splenocytes from STAT1 KO or the companion strain BALB/c mice as control. The Western blot results are representative of at least three independent experiments. See also Supplemental Fig. 1.

FIST desensitizes immune cells to TGF-β–mediated suppression

To test whether FIST can override TGF-β−mediated suppression of lymphocyte activation, lymphocytes were stimulated with 5 pmol FIST or control. FIST on its own primes 5 × 106 unfractionated splenocytes to produce a significant increase (>35%) of IFN-γ compared with equimolar concentration of IL-2 in 96 h. In contrast to IL-2 alone, this effect was not suppressed in the presence of 0.5 ng/ml of active TGF-β1 (Fig. 2A). In addition, we observed that FIST induces c-Myc upregulation, another gene suppressed by active TGF-β (Supplemental Fig. 2A, 2B). In an attempt to define the immune cell type responsive to FIST, 1 × 106 NK cells, T cells, or B cells were cultured with 5 pmol FIST or equimolar concentration of controls for 72 h, and it was determined by ELISA that FIST-stimulated Τ, B, and NK cells produce significantly higher amounts (>90% increase) of IFN-γ, TNF-α, and GM-CSF than those of controls (Fig. 2B–F).

FIGURE 2.
FIGURE 2.

FIST induces significant Th1 cytokine production by activated immune cells. A, Quantification by ELISA of IFN-γ produced by 5 × 106 unfractionated splenocytes from immunocompetent C57BL/6 mice cultured with 5 pmol FIST or controls for 96 h (n = 4; data are shown as mean ± SD). T, B, and NK cells were cultured with 5 pmol FIST or controls for 72 h, and the supernatants were collected and used to measure the concentration of IFN-γ, TNF-α, and GM-CSF by ELISA. B, IFN-γ produced by 1 × 106 FIST- or control-stimulated T cells (n = 3; data are shown as mean ± SD). C and D, TNF-α and IFN-γ produced by 1 × 106 FIST- or control-stimulated B cells (n = 3; data are shown as mean ± SD). E and F, GM-CSF and IFN-γ produced by 1 × 106 FIST- or control-stimulated NK cells (n = 3; data are shown as mean ± SD). For each image, the results are representative of three independent experiments performed in triplicate (n = 3) or quadruplicate (n = 4). See also Supplemental Fig. 2. *p < 0.05, **p < 0.005.

FIST induces a robust immune bystander effect and inhibits tumor growth in vivo

We have previously demonstrated that B16 melanoma secrete large amounts of active TGF-β that results in significant resistance to melanoma-targeted immunotherapy (23). We used this TGF-β biased tumor model and the metastatic model of pancreatic cancer PANC02 cells to test the activity of FIST in vivo. Therefore, we generated polyclonal populations of genetically modified B16 cells expressing FIST or equimolar concentration of controls (10 pmol per million cells in 24 h), and PANC02 cells expressing FIST or equimolar concentration of control (0.002 pmol per million cells in 24 h). Genetically modified as well as nonmodified B16 and PANC02 cells display similar proliferation rates in vitro (data not shown). In the mouse model of melanoma, FIST and IL-2 secreted by a polyclonal population of genetically modified B16 cells are equally effective in inducing a potent antitumor response. Mice injected s.c. with 5 × 105 B16 cells expressing FIST (B16 FIST) or IL-2 (B16 IL-2) did not develop tumors. However, a cohort of mice injected with 5 × 105 B16 IL-2 cells mixed with 5 × 105 null B16 cells developed tumors at a rate comparable with that of the control group implanted with 5 × 105 null B16 cells. In contrast, FIST-mediated bystander effect protected 60% of the injected mice with mixed cells (Fig. 3A). Although mice implanted with B16 cells secreting sTβRII (B16 sTβRII) show a significant delay of tumor growth, B16 sTβRII tumors grew exponentially over time, and after 80–100 d, mice were sacrificed because of large tumors (data not shown). After 80 d, mice previously implanted with B16 FIST or B16 IL-2 cells were challenged with 1 × 106 null B16 cells in the contralateral flank, and the percentage of survival was monitored over time. Fifty percent of mice previously treated with FIST rejected null B16 cells in contrast to the failure of IL-2 in conferring protective immunity (Fig. 3B). Similarly, immunocompetent mice implanted orthotopically with PANC02 cells in their pancreas expressing low concentrations of FIST do not develop tumors, whereas mice implanted with PANC02 cells expressing low concentrations of controls developed large tumors and were sacrificed (Fig. 3C). FIST-treated mice showed normal body weight and behavior and had normal blood counts (Supplemental Table I).

FIGURE 3.
FIGURE 3.

FIST induces a potent antitumor response and immune bystander effect. A, 5 × 105 null or genetically modified B16 cells expressing FIST (B16 FIST cells) or controls were injected s.c. in C57BL/6 immunocompetent mice, and the percentage of survival was monitored over time. B, Mice previously implanted with B16 FIST or B16 IL-2 cells were challenged with 1 × 106 null B16 cells, and the percentage of survival was monitored over time. C, 5 × 105 null or genetically modified PANC02 cells expressing FIST (PANC02 FIST cells) or controls were orthotopically implanted in C57BL/6 immunocompetent mice, and the percentage of survival was monitored over time. The results displayed in A–C are representative of two independent experiments (n = 10 per group; data are shown as mean ± SD). See also Supplemental Table I. *p < 0.05, **p < 0.005.

To assess the ability of FIST to promote tumor regression, mice with pre-established B16 tumors were s.c. injected with 1 × 106 irradiated (50 Gy) B16 FIST cells in the contralateral flank, and tumor volume was measured over time. As result, FIST-treated mice displayed a significantly lower rate of tumor growth compared with that of the control (nontreated group) and transient tumor regression. However, tumor regression was not complete, and the percentage of survival between the groups was not significant (data not shown).

FIST induces significant recruitment of lymphocytes to the tumor site

FIST induces significantly greater recruitment (>50% increase) of host-derived lymphocytes to Matrigel plugs (tumor site) than equimolar concentration of controls after 7 d of tumor implantation. Specifically, we observed a significant increase of NK (>56%), NKT (>70%), B (>80%), and CD8+ T (>85%) cells recruited in the tumor site (Fig. 4A, 4B).

FIGURE 4.
FIGURE 4.

FIST induces significant recruitment of lymphocytes to the tumor site. A, Analysis of immune cells recruited in the Matrigel plugs (tumor site). The total lymphocyte number was determined based on forward and side scattergram (FSC/SSC) pattern for lymphocytes. B, The total number of each cell type was determined by labeling the infiltrated cells with cell surface-specific Abs (n = 3 per group; data are shown as the mean ± SD). The results presented in A and B are representative of three independent experiments. *p < 0.05, **p < 0.005.

FIST antitumor effect in immunodeficient mice

To determine the main immune cell type implicated in FIST-dependent antitumor response in vivo, 5 × 105 null or B16 FIST cells were injected s.c. into an array of immune defective mice, including CD8 KO, CD4 KO, B cell-deficient (μMT), NK-defective beige mice, NOD-SCID mice, and Rag2/γc KO mice, and survival was monitored over time. Surprisingly, B16 FIST cells do not form tumors in CD4 KO, CD8 KO, B cell-deficient (μMT), and NK-defective beige mice (Fig. 5A–D). However, 80% of NOD-SCID mice implanted with B16 FIST cells developed tumors and were sacrificed with a significant delay of 2 mo compared with the control, whereas Rag2/γc KO mice developed tumors at a rate indistinguishable from that of the control group (Fig. 5E, 5F). Based on the fact that NK cells from beige mice display a severe defect of natural killing capability but preserve an intact NK cytokine production (35), whereas NOD-SCID mice display defects in both NK cell number and function (36), it is suggested that FIST antitumor effect can be mediated by an NK cell-derived soluble factor. Notably, tumors expressing FIST implanted in NOD-SCID mice display very necrotic appearance, suggesting a defect in angiogenesis as well (data not shown). We investigated the contribution of IFN-γ in vivo because FIST-stimulated NK cells secrete substantial amounts of IFN-γ. However, B16 FIST cells also fail to form tumors in IFN-γ KO mice (Fig. 5G) as well as in IFN-γR KO mice (data not shown). The analysis in vitro of the expression profile of angiogenesis-related proteins secreted by FIST-stimulated NK cells reveals the presence of high levels of the angiostatic chemokines CXCL10 (IFN-γ inducible protein, also known as IP-10) and CXCL9 (monokine inducible by IFN-γ, also known as MIG), and absence of VEGF expression or other genes required for VEGF response such as angiopoietin-2, tissue inhibitor of matrix metalloproteinase-1, and basic fibroblast growth factor (data not shown). We measured by ELISA the concentration of CXCL10 in the secretome of NK cells stimulated by FIST or controls from immunocompetent, Beige, IFN-γ KO, and NOD-SCID mice. FIST-stimulated NK cells from Beige mice secrete similar CXCL10 amounts as immunocompetent mice, whereas FIST-stimulated NK cells from IFN-γ KO, although displaying a dramatic decrease of CXCL10 production, still have significantly higher CXCL10 production than that of NOD-SCID mice (Fig. 5H). In addition, FIST-stimulated NK cells from immunocompetent mice secrete significantly higher amounts of CXCL9 than those of control (Supplemental Fig. 3).

FIGURE 5.
FIGURE 5.

NK cells as essential mediator of FIST antitumor effects. Null or B16 FIST cells (5 × 105) were injected s.c. into CD8+ T (A), CD4+ T (B), μMT (C), Beige (D), NOD-SCID (E), common γ-chain (γc)/RAG-2 double KO (F), and IFN-γ KO (G) mice, and percentage of survival was monitored over time (n = 5 per group; data are shown as the mean ± SD). H, CXCL10 production by FIST- or controls-stimulated NK cells from C57BL/6 immunocompetent, Beige, IFN-γ KO, and NOD-SCID mice. The results are representative of two independent experiments performed in triplicate (data are shown as the mean ± SD). See also Supplemental Fig. 3. *p < 0.05, **p < 0.005.

FIST inhibits tumor-derived angiogenesis

In contrast to controls, Matrigel plugs containing B16 FIST cells were devoid of any detectable B16-associated blood vessel formation (Fig. 6A). Although we determined that the secretome of FIST-stimulated NK cells did not decrease the proliferation rate of EC in vitro (Fig. 6B), it significantly suppressed their migration (Fig. 6C) and formation of capillary-like structures in a three-dimensional collagen matrix as indicative of progressive angiogenesis (Fig. 6D). To assess the contribution in vitro of CXCL10 as an angiostatic factor, we determined the ability of CXCR3 (CXCL10 receptor) KO EC to form capillary-like structures in the presence of the secretome of FIST-stimulated NK cells from immunocompetent or STAT1 KO mice. CXCR3 KO EC form perfect capillary-like structures in all the conditions tested. In contrast, wild-type EC display the lowest score of progressive angiogenesis in the presence of the secretome of FIST-stimulated NK cells derived from immunocompetent mice, but not from STAT1 KO mice, which show a significant increase of angiogenesis score (Fig. 6E). To assess the contribution of CXCL10 angiostatic effect in vivo, we injected CXCR3 KO mice with 5 × 105 null or B16 FIST cells s.c., and the percentage of survival was determined over time. B16 FIST cells also fail to form tumors in the CXCR3 KO mice (Fig. 6F).

FIGURE 6.
FIGURE 6.

FIST inhibits angiogenesis in vivo and in vitro. A, Matrigel plugs sections stained with DAPI and red fluorescence conjugated CD31-specific Ab to detect blood vessel formation in Matrigel plugs removed from C57BL/6 immunocompetent mice previously injected with 5 × 105 B16 FIST or control cells embedded in Matrigel (original magnification ×40). B, EC proliferation in the presence of CM by FIST- or controls-stimulated NK cells from immunocompetent mice. Cell number was quantified by running a luminescence versus cell dose curve (n = 3; data are shown as the mean ± SD). C, In vitro EC migration in response to the CM of FIST- or controls-stimulated NK cells from immunocompetent mice. Migrated cell number was quantified by running a fluorescence versus cell dose curve (n = 3; data are shown as the mean ± SD). D, In vitro angiogenesis assay of 5 × 103 EC cultured in the presence of the CM of FIST- or controls-stimulated NK cells from immunocompetent mice (n = 3; data are shown as the mean ± SD). E, Comparison of angiogenesis score between 5 × 103 wild-type and CXCR3 KO EC cultured in the presence of the CM of FIST- or control-stimulated NK cells from immunocompetent and STAT1 KO mice (n = 3; data are shown as the mean ± SD). F, Percentage of survival over time of CXCR3 KO mice s.c. injected with 5 × 105 B16 FIST or null cells (n = 5 per group; data are shown as the mean ± SD). *p < 0.05, **p < 0.005.

FIST mediates upregulation of STAT1 target genes

FIST angiostatic properties arise in part from CXCL10 production by FIST-stimulated NK cells (Fig. 7A). Similarly, we observed significant upregulation of T-bet (Fig. 7C) and T-bet target genes such as IFN-γ, CXCR3, and IL-2Rβ (Fig. 7E, 7G, 7I). The expression of CXCL10, T-bet, IFN-γ, and CXCR3 were significantly inhibited in FIST-stimulated NK cells from STAT1 KO mice (Fig. 7B, 7D, 7F, 7H), whereas IL-2Rβ expression was not affected by STAT1 deletion and T-bet downregulation (Fig. 7J), which suggest that other transcription factors such as Eomesodermin may complement their functions (37).

FIGURE 7.
FIGURE 7.

FIST mediates upregulation of STAT1 target genes. A and B, CXCL10 production over time by 1 × 106 FIST- or control-stimulated NK cells from immunocompetent and STAT1 KO mice, respectively. C and D, T-bet expression over time by 1 × 106 FIST- or controls-stimulated NK cells from immunocompetent or STAT1 KO mice, respectively. T-bet expression over time was determined by intracellular staining, and the mean fluorescence intensity (MFI) values were measured by flow cytometry. E and F, IFN-γ production over time by 1 × 106 FIST- or controls-stimulated NK cells from immunocompetent and STAT1 KO mice, respectively. G and H, Cell surface CXCR3 expression over time by 1 × 106 FIST- or controls-stimulated NK cells from immunocompetent and STAT1 KO mice, respectively. I and J, Cell surface IL-2Rβ expression over time by 1 × 106 FIST- or controls-stimulated NK cells from immunocompetent and STAT1 KO mice, respectively. For all the experiments, n = 3 per group; data are shown as the mean ± SD.

FIST working model

Upon binding to IL-2R, FIST not only induces the activation of STAT5 and STAT3 but also promotes a potent activation of STAT1 and consequently upregulates the expression of STAT1 target genes essential for an effective Th1 cell-mediated cell immunity such as T-bet, SMAD7, and CXCL10. In turn, T-bet upregulates the expression of its target genes: CXCR3 and IL-2Rβ. Through SMAD7 expression, FIST inhibits active TGF-β, which exerts dramatic suppression of proinflammatory cytokine-mediated IFN-γ production indirectly by downregulating T-bet expression via SMAD-dependent mechanism and directly by T-bet–independent negative regulatory effect on the IFNG promoter (43). IFN-γ may amplify FIST-dependent activation by binding to IFN-γR and creating a positive autocrine loop that enhances the expression of STAT1 target genes (Fig. 8).

FIGURE 8.
FIGURE 8.

FIST mechanism of action. Upon binding to IL-2R, FIST not only induces the activation of STAT5 and STAT3 but also promotes a potent activation of STAT1 and consequently upregulates the expression of STAT1 target genes essential for an effective Th1 cell-mediated immunity such as IFN-γ, T-bet, SMAD7, CXCR3, IL-2Rβ, and CXCL10. T-bet upregulates the expression of its target genes IFN-γ, CXCR3, and IL-2Rβ, and via SMAD7, FIST inhibits the suppressive effects of active TGF-β on T-bet gene expression and IFN-γ production. In turn, IFN-γ may amplify FIST-dependent activation by binding to IFN-γR and creating a positive autocrine loop that enhance the expression of STAT1 target genes.

Discussion

Upon IL-2 binding, the IL-2R activates two tyrosine kinases, JAK1 and JAK3, which interact with the cytoplasmic domains of the IL-2Rβ and γc subunits (38). Activated IL-2R becomes phosphorylated on specific tyrosine residues that serve as docking sites for proteins containing Src-homology 2 or phosphotyrosine binding domains including Shc adapter and STAT proteins (STAT1, STAT3, and STAT5). IL-2R recruits STAT proteins to different subdomains of the IL-2Rβ-chain. STAT5 associates with phosphorylated Tyr-510 of the IL-2Rβ C-terminal region. In contrast, STAT1 and STAT3 interaction with the IL-2Rβ-chain occurs through its acidic subdomain and may not require either phosphorylation of the receptor or even the presence of tyrosine residues of IL-2Rβ (39, 38). In addition, IL-2–mediated STATs activation is concentration sensitive, as low IL-2 concentration is sufficient to induce STAT5 activation, whereas STAT1 activation requires high IL-2 concentrations (39, 40).

Our new bifunctional pharmaceutical (FIST) is endowed with the ability to signal through the IL-2R inducing similar STAT5 activation and distinctive STAT3 and STAT1 activation compared with equimolar concentrations of IL-2 or IL-2 combined with sTβRII. Specifically, FIST induces a partial hyperagonist response in IL-2R–expressing cells characterized by a potent activation of STAT1 combined with contemporaneous inhibition of TGF-β signaling pathway through transcriptional induction of the inhibitory SMAD7. Based on previous studies, JAK1/STAT1 activation acts as a positive regulator of SMAD7 expression (41). STAT1 deletion completely abrogated FIST-dependent SMAD7 overexpression, but not the inhibition of SMAD2 phosphorylation, which indicates that the TβRII ectodomain moiety of FIST effectively functions as a decoy receptor trapping active TGF-β in solution. SMAD7 directly inhibits TGF-β signaling in the nucleus by interacting with transcriptional repressors or disrupting the formation of the TGF-β–induced functional Smad–DNA complexes (42). SMAD7 also recruits HECT type of E3 ubiquitin ligases, Smurf1 and Smurf2, to the activated TβRI leading to the degradation of the receptor through the proteasomal pathway (11).

TGF-β is also produced by immune cells including activated T and B cells, macrophages, dendritic cells, neutrophils, and NK cells, and its expression regulates the differentiation, proliferation, and activation state of these immune cells. In particular, resting NK cells have a constitutively active TGF-β–TGF-βR autocrine loop in the absence of proinflammatory immune stimulation (43–46).

In a STAT1-dependent mechanism, FIST acts intracellularly to inhibit TGF-β signaling pathway via SMAD7 expression. This property characterizes FIST’s mechanism of action and is responsible for the superiority of FIST as proinflammatory compound versus IL-2 or IL-2 combined with sTβRII. In contrast, total lymphocyte or purified immune cells stimulated by IL-2 plus sTβRII display similar activation as IL-2–stimulated cells indicating that although sTβRII traps active TGF-β secreted by immune cells, it does not operate intracellularly and does not enhance the activation status of STAT proteins downstream of IL-2R. Consequently, FIST-stimulated lymphocytes in vitro become resistant to TGF-β–mediated suppression and produce significantly greater amounts of proinflammatory cytokines. However, IL-2–dependent cells (CTLL-2) cultured with FIST display a similar proliferation rate as CTLL-2 stimulated with equimolar concentrations of IL-2 or IL-2 combined with sTβRII. This confirms the observation that FIST-mediated STAT5 and PI3K activation levels are similar to the one induced by equimolar concentration of IL-2, as these transcription factors are implicated in T cell proliferation and survival (47, 48). FIST-dependent SMAD7 overexpression was only detected in the presence of active TGF-β, which can be explained by the reported SMAD binding elements in the SMAD7 promoter required for TGF-β–dependent transcriptional activation (49).

FIST-mediated hyperagonist signal transduction downstream of the IL-2R coupled to TGF-β blockade leads to resistance to TGF-β–mediated suppression and to a potent antitumor effect in vivo. Consequently, melanoma B16 FIST and PANC02 FIST cells fail to form tumors in immunocompetent mice. However, FIST secreted by irradiated B16 cells failed to induce complete tumor progression in mice with pre-established B16 tumors. The ineffectiveness of this whole-cell vaccine strategy may be due to rapid immune clearance of irradiated B16 FIST cells before they could secrete systemically effective amounts of FIST.

Because it is not possible to modify all preexisting tumor cells with suicide or proinflammatory cytokine genes in situ by any contemporary gene transfer technology, an important feature to consider for cancer immunotherapy is the bystander effect (50). The secretion of FIST by genetically modified B16 cells promotes a bystander anticancer effect that protected 60% of mice bearing non-genetically modified cancer cells from tumor development. Unfortunately, we could not quantify tumor angiogenesis based on CD31 expression as readout of FIST-mediated bystander effect because of the very necrotic consistency of the tumors containing B16 FIST cells. Although FIST promoted significant recruitment of IL-2R–expressing immune cells such as NK, NKT, CD8+ T cells and B cells to the tumor site, FIST still prevents tumor formation in CD4 KO, CD8 KO, and B cell-deficient mice, which indicates that these immune cell types, although they may contribute, are not essential mediators of FIST antitumor activity. In contrast, B16 FIST cells were tumorigenic in NOD-SCID and Rag2/γc KO mice, which implicates by exclusion NK cells as the major immune mediator of FIST effects. One way to precisely evaluate the role of NK cells in this model would be to perform the NK cell depletion with anti-asialo GM1 Ab.

The necrotic appearance of tumors secreting FIST from NOD-SCID mice also suggests a profound defect of tumor-driven angiogenesis. FIST not only antagonizes TGF-β, an essential regulator of both phases of angiogenesis (16), but also stimulates virtually all lymphoid subsets, including NK cells, to secrete substantial amounts of IFN-γ, which may act as an angiostatic by suppressing the expression of genes required for VEGF response via STAT1 activation (51). However, B16 cells expressing FIST fail to form tumors in IFN-γ and IFN-γR KO mice, suggesting that additional effector molecules are at play. FIST-stimulated NK cells produced significant amounts of CXCL10 and CXCL9. CXCL10 and CXCL9 as well as CXCL11 are chemokines induced by IFN-γ via STAT1 activation and are CXCR3 ligands that share potent angiostatic and chemoattractant properties (52–54). In particular, high level of CXCL10 in the tumor is correlated with high amounts of tumor-infiltrating NK cells and better survival (55). FIST-stimulated NK cells from IFN-γ KO mice display a significant decrease of CXCL10 production, which supports the role of IFN-γ as the main inducer of CXCL10 expression. Notably, FIST-stimulated NK cells from IFN-γ KO mice still produce elevated levels of CXCL10 indicating that upon binding to IL-2R, FIST directly induces CXCL10 production via STAT1 hyperactivation. As expected, CXCL10 secretion was significantly reduced in FIST-stimulated NK cells from NOD-SCID mice, which display severe defect in NK cell functions, and was completely abrogated in FIST-stimulated NK cells from STAT1 KO mice. The specific angiostatic effect of FIST on tumor-derived angiogenesis is likely due to the transient CXCR3 expression restricted to newly formed blood vessels in vivo and in vitro (56). In concordance with previous studies, we found that CXCL10 behaves as an inhibitor of EC migration and differentiation into branching networks of tubular structures in vitro, a process that requires the interaction of EC with extracellular matrix components (53, 57, 58). However, the ability of CXCL10 to inhibit EC proliferation is controversial (53, 59). Despite this potent angiostatic effect observed in vitro and in vivo in immunocompetent mice, B16 FIST cells fail to form tumors in CXCR3 KO mice, which indicates that FIST-dependent CXCL10 induction is not essentially required for the FIST-mediated antitumor effect observed in these mice. Therefore, we speculate that other FIST-dependent mechanisms that prevent tumor formation are at play in CXCR3 KO mice such as the induction of a potent type I Th cell-mediated antitumor immunity triggered by STAT1 target genes or the upregulation of other NK cell effector functions such as NK cell cytotoxicity against tumor cells. However, we cannot discard the possibility that several FIST-dependent mechanisms that rely on STAT1 activation may simultaneously operate to prevent cancer progression in immunocompetent mice.

Consequently, FIST-treated mice develop a robust anticancer adaptive immunity that protects 50% of mice from high-burden tumor challenge. Notably, FIST-stimulated NK cells from STAT1 KO mice still produce significant higher amounts of IFN-γ than equimolar concentrations of controls. This suggests that the contemporaneous blockade of active TGF-β combined with an immune stimulation significantly enhances the activation level of NK cells, which constitutively produce active TGF-β1 as an autocrine/negative regulator of IFN-γ (43).

By targeting several immune cell types expressing the IL-2R, inhibiting active TGF-β, and upregulating the expression of STAT1 target genes, FIST conveys the host’s angiogenic and immune response against cancer cells. Therefore, cancer cells expressing FIST only form tumors in NOD-SCID and Rag2/γc double-KO mice, which lack most of these immune system components. Although other TGF-β antagonists like Fc:TβRII may neutralize large amounts of active TGF-β (60), Fc:TβRII treatment does not alter primary tumor growth of transplantable models of breast cancer metastases, which suggests that the antimetastatic effects of Fc:TβRII in vivo are independent of tumor cell proliferation. In addition, inhibition of tumor angiogenesis in vivo by Fc:TβRII treatment is controversial. Muraoka et al. (3) observed no reduction of vascular density in endogenously arising tumors.

FIST treatment could be delivered in humans systemically as recombinant protein, as low doses of the fusion (0.002 pmol) are effective in the treatment of carcinoma.

Alternatively, FIST could be delivered by vehicle cells such as genetically modified stem cells or autologous immune cells.

In conclusion, FIST is a novel biopharmaceutical characterized by inhibiting TGF-β canonical pathway simultaneously with a distinctive STAT1 hyperactivation via IL-2R on immune cells, which conveys a robust upregulation of STAT1 target genes including key factors essential for an effective antitumor response. This first-in-class biological agent may represent a new paradigm in cancer immunotherapy contemporaneously modulating pro- and anti-inflammatory immune checkpoints operative on immune effector cells.

Disclosures

C.P. and J.G. are named coinventors on a provisional U.S. patent for FIST technology.

Acknowledgments

We thank Naciba Benlimame at the George and Olga Minarik Research Pathology Facility and Nicoletta Eliopoulos, Catherine Lemarie, and Joan Durbin for technical advice and materials.

Footnotes

  • This work was supported by a Canadian Institute for Health Research operating grant (MOP-15017). C.P. is a recipient of a U.S. Army predoctoral training award, and J.G. is a Georgia Cancer Coalition Distinguished Cancer Scholar.

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    CM
    conditioned media
    EC
    endothelial cell
    FIST
    fusion of IL-2 and the soluble extracellular domain of TGF-βR II
    KO
    knockout
    sTβRII
    soluble type II TGF-βR
    TβRI
    TGF-βR type I
    TβRII
    TGF-βR type II.

  • Received November 22, 2010.
  • Accepted April 13, 2011.

References

  1. ↵
    1. Reiss M.
    1999. TGF-beta and cancer. Microbes Infect. 1: 1327–1347.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Elliott R. L.,
    2. G. C. Blobe
    . 2005. Role of transforming growth factor beta in human cancer. J. Clin. Oncol. 23: 2078–2093.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Muraoka R. S.,
    2. N. Dumont,
    3. C. A. Ritter,
    4. T. C. Dugger,
    5. D. M. Brantley,
    6. J. Chen,
    7. E. Easterly,
    8. L. R. Roebuck,
    9. S. Ryan,
    10. P. J. Gotwals,
    11. et al
    . 2002. Blockade of TGF-beta inhibits mammary tumor cell viability, migration, and metastases. J. Clin. Invest. 109: 1551–1559.
    OpenUrlCrossRefPubMed
    1. Peng S. B.,
    2. L. Yan,
    3. X. Xia,
    4. S. A. Watkins,
    5. H. B. Brooks,
    6. D. Beight,
    7. D. K. Herron,
    8. M. L. Jones,
    9. J. W. Lampe,
    10. W. T. McMillen,
    11. et al
    . 2005. Kinetic characterization of novel pyrazole TGF-beta receptor I kinase inhibitors and their blockade of the epithelial-mesenchymal transition. Biochemistry 44: 2293–2304.
    OpenUrlCrossRefPubMed
    1. Bandyopadhyay A.,
    2. J. K. Agyin,
    3. L. Wang,
    4. Y. Tang,
    5. X. Lei,
    6. B. M. Story,
    7. J. E. Cornell,
    8. B. H. Pollock,
    9. G. R. Mundy,
    10. L. Z. Sun
    . 2006. Inhibition of pulmonary and skeletal metastasis by a transforming growth factor-beta type I receptor kinase inhibitor. Cancer Res. 66: 6714–6721.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Wick W.,
    2. U. Naumann,
    3. M. Weller
    . 2006. Transforming growth factor-beta: a molecular target for the future therapy of glioblastoma. Curr. Pharm. Des. 12: 341–349.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Derynck R.,
    2. D. V. Goeddel,
    3. A. Ullrich,
    4. J. U. Gutterman,
    5. R. D. Williams,
    6. T. S. Bringman,
    7. W. H. Berger
    . 1987. Synthesis of messenger RNAs for transforming growth factors alpha and beta and the epidermal growth factor receptor by human tumors. Cancer Res. 47: 707–712.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. ten Dijke P.,
    2. K. Miyazono,
    3. C. H. Heldin
    . 2000. Signaling inputs converge on nuclear effectors in TGF-beta signaling. Trends Biochem. Sci. 25: 64–70.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Lagna G.,
    2. A. Hata,
    3. A. Hemmati-Brivanlou,
    4. J. Massagué
    . 1996. Partnership between DPC4 and SMAD proteins in TGF-beta signalling pathways. Nature 383: 832–836.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Feng X. H.,
    2. R. Derynck
    . 2005. Specificity and versatility in tgf-beta signaling through Smads. Annu. Rev. Cell Dev. Biol. 21: 659–693.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Itoh S.,
    2. P. ten Dijke
    . 2007. Negative regulation of TGF-beta receptor/Smad signal transduction. Curr. Opin. Cell Biol. 19: 176–184.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Kavsak P.,
    2. R. K. Rasmussen,
    3. C. G. Causing,
    4. S. Bonni,
    5. H. Zhu,
    6. G. H. Thomsen,
    7. J. L. Wrana
    . 2000. Smad7 binds to Smurf2 to form an E3 ubiquitin ligase that targets the TGF beta receptor for degradation. Mol. Cell 6: 1365–1375.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Goumans M. J.,
    2. C. Mummery
    . 2000. Functional analysis of the TGFbeta receptor/Smad pathway through gene ablation in mice. Int. J. Dev. Biol. 44: 253–265.
    OpenUrlPubMed
    1. Dickson M. C.,
    2. J. S. Martin,
    3. F. M. Cousins,
    4. A. B. Kulkarni,
    5. S. Karlsson,
    6. R. J. Akhurst
    . 1995. Defective haematopoiesis and vasculogenesis in transforming growth factor-beta 1 knock out mice. Development 121: 1845–1854.
    OpenUrlAbstract
    1. Oshima M.,
    2. H. Oshima,
    3. M. M. Taketo
    . 1996. TGF-beta receptor type II deficiency results in defects of yolk sac hematopoiesis and vasculogenesis. Dev. Biol. 179: 297–302.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Goumans M. J.,
    2. G. Valdimarsdottir,
    3. S. Itoh,
    4. A. Rosendahl,
    5. P. Sideras,
    6. P. ten Dijke
    . 2002. Balancing the activation state of the endothelium via two distinct TGF-beta type I receptors. EMBO J. 21: 1743–1753.
    OpenUrlAbstract
  17. ↵
    1. Zhou J.,
    2. J. Zhang,
    3. M. G. Lichtenheld,
    4. G. G. Meadows
    . 2002. A role for NF-kappa B activation in perforin expression of NK cells upon IL-2 receptor signaling. J. Immunol. 169: 1319–1325.
    OpenUrlAbstract/FREE Full Text
    1. Yu T. K.,
    2. E. G. Caudell,
    3. C. Smid,
    4. E. A. Grimm
    . 2000. IL-2 activation of NK cells: involvement of MKK1/2/ERK but not p38 kinase pathway. J. Immunol. 164: 6244–6251.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Zhang J.,
    2. I. Scordi,
    3. M. J. Smyth,
    4. M. G. Lichtenheld
    . 1999. Interleukin 2 receptor signaling regulates the perforin gene through signal transducer and activator of transcription (Stat)5 activation of two enhancers. J. Exp. Med. 190: 1297–1308.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Atkins M. B.,
    2. M. T. Lotze,
    3. J. P. Dutcher,
    4. R. I. Fisher,
    5. G. Weiss,
    6. K. Margolin,
    7. J. Abrams,
    8. M. Sznol,
    9. D. Parkinson,
    10. M. Hawkins,
    11. et al
    . 1999. High-dose recombinant interleukin 2 therapy for patients with metastatic melanoma: analysis of 270 patients treated between 1985 and 1993. J. Clin. Oncol. 17: 2105–2116.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Bright J. J.,
    2. L. D. Kerr,
    3. S. Sriram
    . 1997. TGF-beta inhibits IL-2-induced tyrosine phosphorylation and activation of Jak-1 and Stat 5 in T lymphocytes. J. Immunol. 159: 175–183.
    OpenUrlAbstract
  22. ↵
    1. Nelson B. H.,
    2. T. P. Martyak,
    3. L. J. Thompson,
    4. J. J. Moon,
    5. T. Wang
    . 2003. Uncoupling of promitogenic and antiapoptotic functions of IL-2 by Smad-dependent TGF-beta signaling. J. Immunol. 170: 5563–5570.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Penafuerte C.,
    2. J. Galipeau
    . 2008. TGF beta secreted by B16 melanoma antagonizes cancer gene immunotherapy bystander effect. Cancer Immunol. Immunother. 57: 1197–1206.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Dong M.,
    2. T. How,
    3. K. C. Kirkbride,
    4. K. J. Gordon,
    5. J. D. Lee,
    6. N. Hempel,
    7. P. Kelly,
    8. B. J. Moeller,
    9. J. R. Marks,
    10. G. C. Blobe
    . 2007. The type III TGF-beta receptor suppresses breast cancer progression. J. Clin. Invest. 117: 206–217.
    OpenUrlCrossRefPubMed
    1. Wang Z. G.,
    2. W. Zhao,
    3. M. Ramachandra,
    4. P. Seth
    . 2006. An oncolytic adenovirus expressing soluble transforming growth factor-beta type II receptor for targeting breast cancer: in vitro evaluation. Mol. Cancer Ther. 5: 367–373.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Rowland-Goldsmith M. A.,
    2. H. Maruyama,
    3. T. Kusama,
    4. S. Ralli,
    5. M. Korc
    . 2001. Soluble type II transforming growth factor-beta (TGF-beta) receptor inhibits TGF-beta signaling in COLO-357 pancreatic cancer cells in vitro and attenuates tumor formation. Clin. Cancer Res. 7: 2931–2940.
    OpenUrlAbstract/FREE Full Text
    1. Suzuki E.,
    2. V. Kapoor,
    3. H. K. Cheung,
    4. L. E. Ling,
    5. P. A. DeLong,
    6. L. R. Kaiser,
    7. S. M. Albelda
    . 2004. Soluble type II transforming growth factor-beta receptor inhibits established murine malignant mesothelioma tumor growth by augmenting host antitumor immunity. Clin. Cancer Res. 10: 5907–5918.
    OpenUrlAbstract/FREE Full Text
    1. Yakymovych I.,
    2. U. Engström,
    3. S. Grimsby,
    4. C. H. Heldin,
    5. S. Souchelnytskyi
    . 2002. Inhibition of transforming growth factor-beta signaling by low molecular weight compounds interfering with ATP- or substrate-binding sites of the TGF beta type I receptor kinase. Biochemistry 41: 11000–11007.
    OpenUrlCrossRefPubMed
    1. Bakin A. V.,
    2. A. K. Tomlinson,
    3. N. A. Bhowmick,
    4. H. L. Moses,
    5. C. L. Arteaga
    . 2000. Phosphatidylinositol 3-kinase function is required for transforming growth factor beta-mediated epithelial to mesenchymal transition and cell migration. J. Biol. Chem. 275: 36803–36810.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Tzai T. S.,
    2. C. I. Lin,
    3. A. L. Shiau,
    4. C. L. Wu
    . 1998. Antisense oligonucleotide specific for transforming growth factor-beta 1 inhibit both in vitro and in vivo growth of MBT-2 murine bladder cancer. Anticancer Res. 18(3A): 1585–1589.
    OpenUrlPubMed
  31. ↵
    1. Durbin J. E.,
    2. R. Hackenmiller,
    3. M. C. Simon,
    4. D. E. Levy
    . 1996. Targeted disruption of the mouse Stat1 gene results in compromised innate immunity to viral disease. Cell 84: 443–450.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Stagg J.,
    2. J. H. Wu,
    3. N. Bouganim,
    4. J. Galipeau
    . 2004. Granulocyte-macrophage colony-stimulating factor and interleukin-2 fusion cDNA for cancer gene immunotherapy. Cancer Res. 64: 8795–8799.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Bouchentouf M.,
    2. P. Paradis,
    3. K. A. Forner,
    4. J. Cuerquis,
    5. M. N. Boivin,
    6. J. Zheng,
    7. M. R. Boulassel,
    8. J. P. Routy,
    9. E. L. Schiffrin,
    10. J. Galipeau
    . 2010. Monocyte derivatives promote angiogenesis and myocyte survival in a model of myocardial infarction. Cell Transplant. 19: 369–386.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Mosmann T. R.,
    2. T. Yokota,
    3. R. Kastelein,
    4. S. M. Zurawski,
    5. N. Arai,
    6. Y. Takebe
    . 1987. Species-specificity of T cell stimulating activities of IL 2 and BSF-1 (IL 4): comparison of normal and recombinant, mouse and human IL 2 and BSF-1 (IL 4). J. Immunol. 138: 1813–1816.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Saunders B. M.,
    2. C. Cheers
    . 1996. Intranasal infection of beige mice with Mycobacterium avium complex: role of neutrophils and natural killer cells. Infect. Immun. 64: 4236–4241.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Gombert J. M.,
    2. A. Herbelin,
    3. E. Tancrède-Bohin,
    4. M. Dy,
    5. C. Carnaud,
    6. J. F. Bach
    . 1996. Early quantitative and functional deficiency of NK1+-like thymocytes in the NOD mouse. Eur. J. Immunol. 26: 2989–2998.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Pearce E. L.,
    2. A. C. Mullen,
    3. G. A. Martins,
    4. C. M. Krawczyk,
    5. A. S. Hutchins,
    6. V. P. Zediak,
    7. M. Banica,
    8. C. B. DiCioccio,
    9. D. A. Gross,
    10. C. A. Mao,
    11. et al
    . 2003. Control of effector CD8+ T cell function by the transcription factor Eomesodermin. Science 302: 1041–1043.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Lin J. X.,
    2. W. J. Leonard
    . 2000. The role of Stat5a and Stat5b in signaling by IL-2 family cytokines. Oncogene 19: 2566–2576.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Delespine-Carmagnat M.,
    2. G. Bouvier,
    3. J. Bertoglio
    . 2000. Association of STAT1, STAT3 and STAT5 proteins with the IL-2 receptor involves different subdomains of the IL-2 receptor beta chain. Eur. J. Immunol. 30: 59–68.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Yu C. R.,
    2. H. A. Young,
    3. J. R. Ortaldo
    . 1998. Characterization of cytokine differential induction of STAT complexes in primary human T and NK cells. J. Leukoc. Biol. 64: 245–258.
    OpenUrlAbstract
  41. ↵
    1. Ulloa L.,
    2. J. Doody,
    3. J. Massagué
    . 1999. Inhibition of transforming growth factor-beta/SMAD signalling by the interferon-gamma/STAT pathway. Nature 397: 710–713.
    OpenUrlCrossRefPubMed
  42. ↵
    1. Zhang S.,
    2. T. Fei,
    3. L. Zhang,
    4. R. Zhang,
    5. F. Chen,
    6. Y. Ning,
    7. Y. Han,
    8. X. H. Feng,
    9. A. Meng,
    10. Y. G. Chen
    . 2007. Smad7 antagonizes transforming growth factor beta signaling in the nucleus by interfering with functional Smad-DNA complex formation. Mol. Cell. Biol. 27: 4488–4499.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Yu J.,
    2. M. Wei,
    3. B. Becknell,
    4. R. Trotta,
    5. S. Liu,
    6. Z. Boyd,
    7. M. S. Jaung,
    8. B. W. Blaser,
    9. J. Sun,
    10. D. M. Benson Jr..,
    11. et al
    . 2006. Pro- and antiinflammatory cytokine signaling: reciprocal antagonism regulates interferon-gamma production by human natural killer cells. Immunity 24: 575–590.
    OpenUrlCrossRefPubMed
    1. Letterio J. J.,
    2. A. B. Roberts
    . 1998. Regulation of immune responses by TGF-beta. Annu. Rev. Immunol. 16: 137–161.
    OpenUrlCrossRefPubMed
    1. Horwitz D. A.,
    2. J. D. Gray,
    3. K. Ohtsuka
    . 1999. Role of NK cells and TGF-beta in the regulation of T-cell-dependent antibody production in health and autoimmune disease. Microbes Infect. 1: 1305–1311.
    OpenUrlCrossRefPubMed
  46. ↵
    1. Dennler S.,
    2. M. J. Goumans,
    3. P. ten Dijke
    . 2002. Transforming growth factor beta signal transduction. J. Leukoc. Biol. 71: 731–740.
    OpenUrlAbstract/FREE Full Text
  47. ↵
    1. Gu H.,
    2. H. Maeda,
    3. J. J. Moon,
    4. J. D. Lord,
    5. M. Yoakim,
    6. B. H. Nelson,
    7. B. G. Neel
    . 2000. New role for Shc in activation of the phosphatidylinositol 3-kinase/Akt pathway. Mol. Cell. Biol. 20: 7109–7120.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Hand T. W.,
    2. W. Cui,
    3. Y. W. Jung,
    4. E. Sefik,
    5. N. S. Joshi,
    6. A. Chandele,
    7. Y. Liu,
    8. S. M. Kaech
    . 2010. Differential effects of STAT5 and PI3K/AKT signaling on effector and memory CD8 T-cell survival. Proc. Natl. Acad. Sci. USA 107: 16601–16606.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    1. Stopa M.,
    2. D. Anhuf,
    3. L. Terstegen,
    4. P. Gatsios,
    5. A. M. Gressner,
    6. S. Dooley
    . 2000. Participation of Smad2, Smad3, and Smad4 in transforming growth factor beta (TGF-beta)-induced activation of Smad7. THE TGF-beta response element of the promoter requires functional Smad binding element and E-box sequences for transcriptional regulation. J. Biol. Chem. 275: 29308–29317.
    OpenUrlAbstract/FREE Full Text
  50. ↵
    1. Sotomayor M. G.,
    2. H. Yu,
    3. S. Antonia,
    4. E. M. Sotomayor,
    5. D. M. Pardoll
    . 2002. Advances in gene therapy for malignant melanoma. Cancer Contr. 9: 39–48.
    OpenUrl
  51. ↵
    1. Battle T. E.,
    2. R. A. Lynch,
    3. D. A. Frank
    . 2006. Signal transducer and activator of transcription 1 activation in endothelial cells is a negative regulator of angiogenesis. Cancer Res. 66: 3649–3657.
    OpenUrlAbstract/FREE Full Text
  52. ↵
    1. Giese N. A.,
    2. Z. Raykov,
    3. L. DeMartino,
    4. A. Vecchi,
    5. S. Sozzani,
    6. C. Dinsart,
    7. J. J. Cornelis,
    8. J. Rommelaere
    . 2002. Suppression of metastatic hemangiosarcoma by a parvovirus MVMp vector transducing the IP-10 chemokine into immunocompetent mice. Cancer Gene Ther. 9: 432–442.
    OpenUrlCrossRefPubMed
  53. ↵
    1. Angiolillo A. L.,
    2. C. Sgadari,
    3. D. D. Taub,
    4. F. Liao,
    5. J. M. Farber,
    6. S. Maheshwari,
    7. H. K. Kleinman,
    8. G. H. Reaman,
    9. G. Tosato
    . 1995. Human interferon-inducible protein 10 is a potent inhibitor of angiogenesis in vivo. J. Exp. Med. 182: 155–162.
    OpenUrlAbstract/FREE Full Text
  54. ↵
    1. Mikhak Z.,
    2. C. M. Fleming,
    3. B. D. Medoff,
    4. S. Y. Thomas,
    5. A. M. Tager,
    6. G. S. Campanella,
    7. A. D. Luster
    . 2006. STAT1 in peripheral tissue differentially regulates homing of antigen-specific Th1 and Th2 cells. J. Immunol. 176: 4959–4967.
    OpenUrlAbstract/FREE Full Text
  55. ↵
    1. Wendel M.,
    2. I. E. Galani,
    3. E. Suri-Payer,
    4. A. Cerwenka
    . 2008. Natural killer cell accumulation in tumors is dependent on IFN-gamma and CXCR3 ligands. Cancer Res. 68: 8437–8445.
    OpenUrlAbstract/FREE Full Text
  56. ↵
    1. Bodnar R. J.,
    2. C. C. Yates,
    3. M. E. Rodgers,
    4. X. Du,
    5. A. Wells
    . 2009. IP-10 induces dissociation of newly formed blood vessels. J. Cell Sci. 122: 2064–2077.
    OpenUrlAbstract/FREE Full Text
  57. ↵
    1. Grant D. S.,
    2. K. Tashiro,
    3. B. Segui-Real,
    4. Y. Yamada,
    5. G. R. Martin,
    6. H. K. Kleinman
    . 1989. Two different laminin domains mediate the differentiation of human endothelial cells into capillary-like structures in vitro. Cell 58: 933–943.
    OpenUrlCrossRefPubMed
  58. ↵
    1. Strieter R. M.,
    2. S. L. Kunkel,
    3. D. A. Arenberg,
    4. M. D. Burdick,
    5. P. J. Polverini
    . 1995. Interferon gamma-inducible protein 10 (IP-10), a member of the C-X-C chemokine family, is an inhibitor of angiogenesis. Biochem. Biophys. Res. Commun. 210: 51–57.
    OpenUrlCrossRefPubMed
  59. ↵
    1. Luster A. D.,
    2. S. M. Greenberg,
    3. P. Leder
    . 1995. The IP-10 chemokine binds to a specific cell surface heparan sulfate site shared with platelet factor 4 and inhibits endothelial cell proliferation. J. Exp. Med. 182: 219–231.
    OpenUrlAbstract/FREE Full Text
  60. ↵
    1. Yang Y. A.,
    2. O. Dukhanina,
    3. B. Tang,
    4. M. Mamura,
    5. J. J. Letterio,
    6. J. MacGregor,
    7. S. C. Patel,
    8. S. Khozin,
    9. Z. Y. Liu,
    10. J. Green,
    11. et al
    . 2002. Lifetime exposure to a soluble TGF-beta antagonist protects mice against metastasis without adverse side effects. J. Clin. Invest. 109: 1607–1615.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section