Sorafenib is an oral multikinase inhibitor that was originally developed as a Raf kinase inhibitor. We hypothesized that sorafenib would also have inhibitory effects on cytokine signaling pathways in immune cells. PBMCs from normal donors were treated with varying concentrations of sorafenib and stimulated with IFN-α or IL-2. Phosphorylation of STAT1 and STAT5 was measured by flow cytometry and confirmed by immunoblot analysis. Changes in IFN-α– and IL-2–stimulated gene expression were measured by quantitative PCR, and changes in cytokine production were evaluated by ELISA. Cryopreserved PBMCs were obtained from cancer patients before and after receiving 400 mg sorafenib twice daily. Patient PBMCs were thawed, stimulated with IL-2 or IFN-α, and evaluated for phosphorylation of STAT1 and STAT5. Pretreatment of PBMCs with 10 μM sorafenib decreased STAT1 and STAT5 phosphorylation after treatment with IFN-α or IL-2. This inhibitory effect was observed in PBMCs from healthy donors over a range of concentrations of sorafenib (5–20 μM), IL-2 (2–24 nM), and IFN-α (101–106 U/ml). This effect was observed in immune cell subsets, including T cells, B cells, NK cells, regulatory T cells, and myeloid-derived suppressor cells. Pretreatment with sorafenib also inhibited PBMC expression of IFN-α– and IL-2–regulated genes and inhibited NK cell production of IFN-γ, RANTES, MIP1-α, and MIG in response to IFN-α stimulation. PBMCs from patients receiving sorafenib therapy showed decreased responsiveness to IL-2 and IFN-α treatment. Sorafenib is a Raf kinase inhibitor that could have off-target effects on cytokine-induced signal transduction in immune effector cells.

Sorafenib (BAY43-9006, Nexavar) is an oral multikinase inhibitor that was originally developed as a Raf kinase inhibitor (1). It was shown to have inhibitory effects on the wild-type B-Raf and oncogenic b-raf V600E serine/threonine kinases, proangiogenic receptor tyrosine kinases such as vascular endothelial growth factor receptor-1, -2, and -3, platelet-derived growth factor receptor-β, and fibroblast growth factor receptor-1. Sorafenib also has inhibitory effects on other receptor tyrosine kinases such as c-Kit, FLT-3, p38, and RET (1, 2). These molecular targets are implicated in the etiology of several human cancers (1).

In hepatocellular carcinoma, sorafenib has been shown to inhibit the RAF/MEK/ERK pathway, inducing apoptosis and reducing tumor-induced angiogenesis (3). Activating mutations in B-RAF and genetic rearrangements in RET are critical to the development of thyroid cancers, and phase II trials have shown that sorafenib leads to stabilization of disease in this setting (46). Sorafenib leads to reduced tumor cell growth and vascularization in renal cell carcinoma (7) and induces apoptosis in imatinib mesylate–resistant human leukemia cells by inhibiting STAT5 in myeloid cell leukemia-1 (8). In human glioblastoma, medulloblastoma, and neuroblastoma, sorafenib contributes to growth arrest and apoptosis by inhibition of STAT signaling (911).

Sorafenib is the only Food and Drug Administration–approved medication for use in advanced hepatocellular carcinoma (HCC). The phase III Sorafenib Hepatocellular Carcinoma Assessment Randomized Protocol trial demonstrated that sorafenib significantly prolonged time to progression and overall survival in patients with unresectable HCC (12, 13). Sorafenib has also been approved by the Food and Drug Administration for use in renal cell carcinoma (RCC) based on its ability to increase progression-free survival in patients with metastatic disease. The activity of sorafenib in this setting is attributed to its ability to abrogate the effects of secreted proangiogenic factors (2).

We hypothesized that sorafenib would have off-target inhibitory effects on cytokine signaling pathways (e.g., IL-2 and IFN-α) in immune effector cells. Negative effects of sorafenib on these pathways would likely inhibit the actions of these and other immune hormones whether administered exogenously with therapeutic intent or produced endogenously by immune effector cells in response to tumor Ags or infectious agents. It was also hypothesized that sorafenib could have inhibitory effects on the actions of immune suppressor cells, such as regulatory T cells or myeloid-derived suppressor cells (MDSCs).

rIL-2 was obtained from Roche Pharmaceuticals (Nutley, NJ). rIFN-α-2b (2 × 108 IU/mg) was obtained from Schering-Plough (Kenilworth, NJ). rIFN-γ was purchased from R&D Systems (Minneapolis, MN). rIL-6 was purchased from PeproTech (Rocky Hill, NJ). Anti–phospho-STAT5 (p-STAT5, Tyr694)- and anti–phospho-STAT1 (p-STAT1, Tyr701)-conjugated Ab and isotype control Ab were obtained from BD Biosciences (San Jose, CA). Anti-STAT5, anti-STAT1, and anti-STAT3 mouse monoclonal immunoblot Abs were obtained from BD Biosciences. Anti–phospho-STAT5 (p-STAT5, Tyr694), anti–phospho-STAT1 (p-STAT1, Tyr701), and anti–phospho-STAT3 (p-STAT3, Tyr705) rabbit immunoblot Abs were obtained from Cell Signaling Technology (Danvers, MA). Allophycocyanin–annexin V and propidium iodide staining solution was obtained from BD Biosciences. IFN-γ, MIP-1α, RANTES, MIP-1α, and MIG ELISA kits were obtained from R&D Systems. Allophycocyanin-conjugated mouse anti-human mAb to CD3, CD21, V450-conjugated CD3, and PE-conjugated CD25, CD122, MICA/B, and NKG2D were obtained from BD Biosciences. FITC-conjugated CD56, CCR7, and IFNAR, allophycocyanin-conjugated CD45RO and CD27, VioGreen-conjugated CD45RA and CD3, and V450-conjguated CD11b were obtained from Miltenyi Biotec (Auburn, CA). Allophycocyanin-conjugated mouse anti-human mAbs to CD4, CD8, CD14, and an NKH-1 RD1 mouse anti-human mAb were obtained from Beckman Coulter (Brea, CA). FITC-conjugated goat anti-rabbit IgG were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Normal mouse and normal goat IgG were obtained from Sigma-Aldrich (St. Louis, MO). Sorafenib was obtained from Bayer-Onyx (West Haven, CT), reconstituted with DMSO to a final concentration of 10 mM, and diluted further in PBS supplemented with 0.1% human albumin to a working concentration of 1 mM.

Normal PBMCs were obtained from healthy adult blood donors (source leukocytes, American Red Cross, Columbus, OH). PBMCs were separated from source leukocytes by density gradient centrifugation with Ficoll-Paque (Amersham Pharmacia Biotech, Uppsala, Sweden) as previously described, and used immediately (14).

This study was approved by the Institutional Review Board at The Ohio State University. Patients enrolled on a National Cancer Institute–sponsored phase II study of sorafenib (BAY 43-9006) in patients with metastatic medullary thyroid carcinoma (The Ohio State University no. 06054) received this drug at a dose of 400 mg orally twice daily (5). Mononuclear cells were isolated from peripheral blood of patients prior to sorafenib therapy and after 8 and 16 wk of therapy and cryopreserved. Clinical samples were processed within 1 h of being drawn. Peripheral blood from patients with renal cell carcinoma enrolled on a biological specimens banking protocol (The Ohio State University no. 0390) was used to isolate MDSCs.

The lymphoma cell lines 697 and Ramos were obtained from Dr. John Byrd (The Ohio State University). The SK-RC-45 human renal cell carcinoma cell lines were obtained from Dr. Charles Tannenbaum (Cleveland Clinic and Foundation, Cleveland, OH). NK-92 is a CD3CD56+ NK cell leukemia cell line (15) provided by Dr. Hans G. Klingermann (Vancouver, BC, Canada). The Caki human renal cell carcinoma, A375 and Hs294T human melanoma cell lines, and K562 human chronic myelogenous leukemia cell line were obtained from the American Type Culture Collection (Manassas, VA).

Detection of activated transcription factors in PBMCs was evaluated by intracellular flow cytometry as previously described by our group (14, 16). Briefly, PBMCs were resuspended in 100 μl RPMI 1640 without serum, treated with sorafenib or DMSO control, and stimulated with PBS, IL-2, IFN-α-2b, or IFN-γ and stained for p-STAT1 or p-STAT5. Analyses were performed as previously described using a FACSCalibur flow cytometer (see Fig. 1) or an LSR II flow cytometer (remaining figures) (Becton Dickinson, Franklin Lakes, NJ) (16). Data were expressed as specific fluorescence (Fsp = FtFb), where Ft represents the median value of total staining and Fb represents the median value of background staining with an isotype control Ab. Additionally, the percentage of positive cells (i.e., cells expressing any level of phosphorylated STAT protein) was determined from quadrants set with isotype control Abs.

FIGURE 1.

Phosphorylation of STAT1 and STAT5 is confirmed by flow cytometry and immunoblotting. Freshly isolated PBMCs were treated with multiple concentrations of sorafenib for 30 min and then stimulated with 8 nM IL-2, 105 U/ml IFN-α, 10 ng/ml IFN-γ, or 10 ng/ml IL-6 for 15 min at 37°C. Phosphorylated STAT5 and STAT1 were measured by intracellular flow cytometry (A) and confirmed by immunoblotting whole-cell lysates for p-STAT5, p-STAT1, STAT5, and STAT1. Phosphorylation of STAT3 was also evaluated by immunoblot analysis (B). Data are representative of two independent experiments with similar results.

FIGURE 1.

Phosphorylation of STAT1 and STAT5 is confirmed by flow cytometry and immunoblotting. Freshly isolated PBMCs were treated with multiple concentrations of sorafenib for 30 min and then stimulated with 8 nM IL-2, 105 U/ml IFN-α, 10 ng/ml IFN-γ, or 10 ng/ml IL-6 for 15 min at 37°C. Phosphorylated STAT5 and STAT1 were measured by intracellular flow cytometry (A) and confirmed by immunoblotting whole-cell lysates for p-STAT5, p-STAT1, STAT5, and STAT1. Phosphorylation of STAT3 was also evaluated by immunoblot analysis (B). Data are representative of two independent experiments with similar results.

Close modal

Detection of activated transcription factors in PBMCs was evaluated by immunoblot. Briefly, PBMCs were resuspended in 100 μl RPMI 1640 without serum, treated with sorafenib, and stimulated with IL-2, IFN-α-2b, IFN-γ, or IL-6. Cells were lysed in RIPA buffer and postnuclear lysates were boiled in an equal volume of 2× SDS sample buffer (125 mM Tris-HCl [pH 6.8], 4% SDS, 10% glycerol, 0.01% bromphenol blue, and 2% 2-ME) for 10 min. Proteins were separated by SDS-PAGE, transferred to nitrocellulose filters, probed with p-STAT5, p-STAT1, and p-STAT3 primary Abs, and developed by ECL. Total STAT Abs were used to confirm equal loading.

Phosphatidylserine exposure and cell wall integrity were assessed in PBMCs treated with sorafenib or DMSO vehicle control by flow cytometry using allophycocyanin–annexin V and propidium iodide (BD Biosciences) as previously described (17).

Total PBMCs were depleted of CD3-, CD4-, CD19-, CD33-, and CD66b-expressing cells using RosetteSep (StemCell Technologies, Vancouver, BC, Canada) and purified over a Ficoll gradient. CD34+ cells were enriched from the remaining mononuclear cells using an indirect CD34 isolation kit (Miltenyi Biotec). All CD34+ fractions were >95% pure by flow cytometry using an allophycocyanin-conjugated anti-human CD34 Ab (BD Biosciences).

CD4+ and CD8+ T cells were negatively enriched from fresh leukopacks (American Red Cross, Columbus, OH) by 30 min incubation with RosetteSep CD4 T (or CD8 T) cell enrichment mixture (StemCell Technologies), followed by Ficoll-Paque (Amersham Pharmacia Biotech, Uppsala, Sweden) density gradient centrifugation. CD4+ (or CD8+) cells were then isolated to purity by positive selection using an EasySep human CD4 (or CD8) positive selection kit (Miltenyi Biotec). Purity of isolated CD4+ and CD8+ cells was evaluated by flow cytometry using CD3-V450 and CD4-allophycocyanin, or CD8-allophycocyanin, and purity was validated to be >95%. NK cells were negatively enriched from fresh peripheral blood leukopacks (American Red Cross, Columbus, OH) by 30 min incubation with RosetteSep mixture (StemCell Technologies) followed by density gradient centrifugation as previously described (18) and used immediately. Purity of the isolated NK cells was evaluated using CD56-allophycocyanin and CD3-V450 and validated to be >81%.

T cells and NK cells were obtained as described above. Cells were incubated with fluorochrome-labeled Abs at 4°C. Specific Abs for the T cell subsets include CD3-V450, CCR7-FITC, CD45RO-allophycocyanin, and CD45RA-VioGreen (Miltenyi Biotec). Specific Abs for the NK cell subsets include CD56-FITC, CD11b-V450, CD3-VioGreen, and CD27-allophycocyanin (Miltenyi Biotec). Flow cytometry was performed as previously described (19).

Sorafenib-treated K562 cells and NK cells were analyzed by flow cytometry for surface expression of MHC class I A, B, and C, MICA/B, NKG2D, and the receptors for IL-2 (IL-2Rα and IL-2Rβ) and IFN-α (IFNAR2). Cells were incubated with fluorochrome-labeled Abs at 4°C. Specific Abs for the T cell subsets include PE-conjugated CD25, CD122, MICA/B, NKG2D (BD Biosciences), IFNAR2, and FITC-conjugated HLA-ABC (Miltenyi Biotec).

CD4+ T cells were isolated directly from fresh leukopacks (American Red Cross, Columbus, OH) by 30 min incubation with RosetteSep CD4 T cell enrichment mixture (Stem Cell Technologies), followed by Ficoll-Paque (Amersham Pharmacia Biotech, Uppsala, Sweden) density gradient centrifugation. CD4+CD25+ cells were then isolated by positive selection using CD25 microbeads (Miltenyi Biotec). Purity of isolated CD4+CD25+ cells and Foxp3 expression were evaluated by flow cytometry using a regulatory T cell staining kit (CD4-allophycocyanin, CD25-FITC, and Foxp3-PE; eBioscience, San Diego, CA).

Murine MDSCs were isolated from BALB/c mice bearing 4T1 murine breast cancer tumors. Spleens were harvested aseptically from tumor-bearing mice, filtered through 70-μm cell strainers, washed with PBS, and resuspended in RPMI 1640 with 10% FBS (complete media). GR1+/CD11b+ MDSCs were isolated on a magnetic column using anti-GR1 biotinylated beads (Miltenyi Biotec) with purity >95% by flow cytometry. Human MDSCs were isolated from peripheral blood of patients with RCC enrolled on a biological specimens banking protocol (The Ohio State University no. 0390). Briefly, myeloid cells were first isolated using an EasySep myeloid isolation kit (StemCell Technologies). Cells were then labeled with anti–HLA-DR magnetic microbeads (Miltenyi Biotec) and negatively selected using an EasySep magnet.

MDSCs were obtained as described above. Cells were incubated with fluorochrome-labeled Abs at 4°C. Specific Abs include CD33-PE, HLA-DR–PC7, CD11b-allophycocyanin, CD14–Pacific Blue, and CD15-FITC (Beckman Coulter). Flow cytometry was performed as previously described (20).

NK cell coculture assays were performed as previously described (21). Normal NK cells were obtained from healthy adult blood donors (source leukocytes, American Red Cross, Columbus, OH) using an NK cell enrichment RosetteSep (StemCell Technologies). K562 cell lines were cultured in the wells of a 96-well flat-bottom culture plate. Purified human NK cells were subsequently added to the wells (2 × 105 cells/well) in 200 μl 10% HAB medium supplemented with IFN-α (103 U/ml) and sorafenib (20 μM) or DMSO. Control conditions consisted of NK cells with or without tumor cells treated with medium alone, sorafenib alone, DMSO alone, or cytokine alone. Cell-free culture supernatants were harvested after 48 h and analyzed for IFN-γ, RANTES, MIP-1α, and MIG by ELISA according to the manufacturer’s protocol (R&D Systems) (22).

Real-time PCR was performed to evaluate the expression of cytokine-responsive genes as previously described (14). Briefly, total RNA was isolated from the cultured PBMCs with the use of an RNeasy RNA isolation kit (Qiagen, Valencia, CA) and quantitated using the Ultrospec 3100 Pro spectrophotometer (Amersham Pharmacia Biotech, Piscataway, NJ). Reverse transcription was performed using 2 μg total RNA and random hexamers (PerkinElmer, Norwalk, CT) as primers for first-strand synthesis of cDNA. The resulting cDNA (2 μl) was used as a template to measure the levels of mRNA for 2′,5′-oligoadenylate synthetase 1 (OAS1), IFN-induced protein with tetratricopeptide repeats 2 (IFIT2), IFN-γ, cytokine-inducible SH2 domain–containing protein (CIS), CD69, suppressor of cytokine signaling 1 (SOCS1), CXCL10, and PIM-1 genes by real-time PCR using predesigned primer/probe sets (Applied Biosystems, Foster City, CA) and 2× TaqMan Universal PCR Master Mix (Applied Biosystems). Predesigned primer/probe sets for human β-actin or GAPDH, housekeeping genes, were used as an internal control in each reaction well.

Pairwise comparisons between sorafenib treatment and vehicle control for the healthy adult donor and cell line data were performed using two-sided paired t tests, at a significance level of p = 0.05. For the patient samples, linear mixed models with repeated measurement were used for hypothesis testing to take into account the dependency of the observations at different times from the same donor cells or cells isolated from same patient. Data were log transformed when necessary to satisfy the normality assumption in the model and stabilize the variance. All analyses were performed using SAS v9.2 (SAS Institute, Cary, NC).

Because sorafenib has been shown to have potent effects on a variety of cellular kinases, we tested whether sorafenib would also influence phosphorylation of signal transduction proteins normally activated by IL-2, IFN-α, IFN-γ, and IL-6 in immune effector cells. Freshly isolated PBMCs were treated with increasing concentrations of sorafenib for 30 min and subsequently stimulated with doses of IL-2, IFN-α, IFN-γ, and IL-6 that were felt to be maximally activating. IL-2 was administered at a dose of 8 nM, IFN-α was used at a concentration of 105 U/ml, IFN-γ was employed at a dose of 10 ng/ml, and IL-6 was given at a dose of 10 ng/ml. The phosphorylation status of STAT5 and STAT1 was evaluated by flow cytometry and confirmed by immunoblot analysis (Fig. 1, Supplemental Fig. 1). Phosphorylation of STAT3 was evaluated by immunoblot analysis (Fig. 1B). Decreased phosphorylation of STAT5 following IL-2 stimulation was seen at doses of sorafenib >2.5 μM. Phosphorylation of STAT1 in response to IFN-α and IFN-γ was also inhibited by sorafenib pretreatment at doses >2.5 μM (all p values < 0.01). Decreased phosphorylation of STAT3 in response to IL-6 was seen at all doses of sorafenib tested. At doses of ≤2.5 μM, sorafenib had a detectable but nonsignificant effect on cytokine-induced activation of STAT1 and STAT5 proteins. Almost complete inhibition of cytokine signaling was achieved when sorafenib was used at concentrations of ≥10 μM, a level seen in the plasma of patients receiving standard oral doses of sorafenib (23). Basal levels of unphosphorylated STAT5, STAT1, and STAT3 were unaffected by sorafenib treatment (Fig. 1B). Sorafenib-treated PBMCs also exhibited a 2-fold decrease in phosphorylated ERK following PMA stimulation (data not shown). Thus, sorafenib exhibits broad effects on kinases in PBMCs. No difference in cell viability between vehicle-treated and sorafenib-treated PBMCs was observed in this experiment (98.1 and 97.6% viable cells, respectively). PBMCs were further examined at 24 and 48 h for viability, and no significant difference between vehicle-treated and sorafenib-treated cells was found at all doses tested (Supplemental Fig. 2). Additionally, PBMCs stimulated with IL-21 after sorafenib pretreatment also showed a significant decrease in STAT1 phosphorylation as compared with controls (data not shown).

A dose-response experiment (i.e., use of increasing concentrations of cytokine) revealed that pretreatment of PBMCs with sorafenib (10 μM for 30 min) rendered them insensitive to even very high concentrations of cytokine. In fact, there was no tested dose of IL-2, IFN-α, or IFN-γ that could overcome the effects of sorafenib pretreatment and induce JAK–STAT signal transduction (Fig. 2A–C). In a separate experiment, PBMCs were treated with sorafenib for 30 min, washed, rested for varying periods of time, treated with 8 nM IL-2, and then analyzed for activation of STAT5. As shown in Fig. 2D, the sorafenib pretreatment had a statistically significant effect on STAT5 phosphorylation for at least 16 h following removal of the drug (p < 0.05). This same pattern of results was seen when sorafenib- and DMSO-treated PBMCs were compared with respect to the actual percentage of cells that exhibited positive staining for phosphorylated STAT5 after exposure to IL-2 (data not shown).

FIGURE 2.

Sorafenib-treated PBMCs stimulated with cytokines show differences in specific fluorescence when measuring p-STAT1 and p-STAT5 levels via flow cytometry versus control DMSO. Freshly isolated PBMCs were treated with DMSO or 10 μM sorafenib for 30 min and then stimulated with increasing doses of IL-2, IFN-α, or IFN-γ for 15 min at 37°C. Phosphorylated STAT5 or STAT1 levels were measured by intracellular flow cytometry and the results are expressed as specific fluorescence (AC). PBMCs were treated with DMSO or 10 μM sorafenib for 30 min. The media were changed and the PBMCs were allowed to rest for varying amounts of time before being stimulated with 8 nM IL-2 for 15 min at 37°C. Phosphorylated STAT5 was measured by intracellular flow cytometry. Data are expressed as specific fluorescence (D). Data shown are the means ± SEM of three (A–C) or five independent experiments (D). There was a statistically significant difference between DMSO and sorafenib treatment (A–D) (p < 0.05).

FIGURE 2.

Sorafenib-treated PBMCs stimulated with cytokines show differences in specific fluorescence when measuring p-STAT1 and p-STAT5 levels via flow cytometry versus control DMSO. Freshly isolated PBMCs were treated with DMSO or 10 μM sorafenib for 30 min and then stimulated with increasing doses of IL-2, IFN-α, or IFN-γ for 15 min at 37°C. Phosphorylated STAT5 or STAT1 levels were measured by intracellular flow cytometry and the results are expressed as specific fluorescence (AC). PBMCs were treated with DMSO or 10 μM sorafenib for 30 min. The media were changed and the PBMCs were allowed to rest for varying amounts of time before being stimulated with 8 nM IL-2 for 15 min at 37°C. Phosphorylated STAT5 was measured by intracellular flow cytometry. Data are expressed as specific fluorescence (D). Data shown are the means ± SEM of three (A–C) or five independent experiments (D). There was a statistically significant difference between DMSO and sorafenib treatment (A–D) (p < 0.05).

Close modal

To further evaluate the effects of sorafenib treatment on immune cells, IFN-α– and IL-2–induced gene expression after sorafenib exposure was evaluated. PBMCs were isolated from normal donors, treated with 10 μM sorafenib for 30 min, and then stimulated with IFN-α or IL-2 for 4 h. RNA was isolated from stimulated PBMCs, and real-time PCR for cytokine-induced genes was performed. OAS1, IFIT2, and CXCL10 are STAT-1–regulated genes. CD69 is a protein kinase regulator and SOCS1 is a cytokine-inducible negative regulator of signaling. As shown in Fig. 3A–D, IFN-α–induced expression of these genes following pretreatment with sorafenib was dramatically decreased as compared with PBMCs that were pretreated with vehicle control (fold induction, 54.4 ± 10.5 versus 24.0 ± 5.3, 8454 ± 1253 versus 3422 ± 829.5, 69.1 ± 0.3 versus 0.04 ± 0.4 [p < 0.05], 13.1 ± 0.3 versus 4.5 ± 0.4, and 11.0 ± 0.6 versus 4.0 ± 0.3, respectively). Real-time PCR also revealed a marked decrease in PIM-1, CIS, and IFN-γ expression following IL-2 stimulation of sorafenib-treated PBMCs as compared with vehicle-treated cells (p < 0.05) (Fig. 3F). To demonstrate that sorafenib inhibits actual immune cell effector function, NK cells were isolated from normal donors, cocultured with K562 cells, treated with sorafenib, and then stimulated with IFN-α. Cytokine production was evaluated by ELISA. Sorafenib pretreatment significantly inhibited NK cell production of IFN-γ, RANTES, MIP1-α, and MIG in response to IFN-α stimulation (all p < 0.005) (Fig. 3G). There was minimal cytokine production by NK cells in the absence of tumor cells (data not shown). Experiments to examine how sorafenib affects various receptors on K562 and NK cells used in this coculture were also conducted. Results are described below.

FIGURE 3.

Quantitative PCR and ELISA results show differences in gene expression and cytokine production for IFN-α– and IL-2–responsive genes after treatment with sorafenib. Freshly isolated PBMCs were treated with 10 μM sorafenib for 30 min and stimulated with either 105 U/ml IFN-α or 8 nM IL-2 for 4 h at 37°C. Total RNA was isolated from treated cells and converted to cDNA. Real-time PCR was performed for the IFN-α–responsive genes OAS1 (A), IFIT2 (B), CD69 (C), CXCL10 (D), and SOCS1 (E). The IL-2–responsive genes PIM-1, CIS, and IFN-γ (F) were also examined. Induction of gene expression was calculated relative to β-actin and GAPDH. Data shown are representative of two independent experiments with similar results (C–E) or one independent experiment (A, B, and F). The mean and SD of triplicate determination are shown for all panels. Freshly isolated NK cells were cocultured with K562 cells, pretreated with 20 μM sorafenib, and then stimulated with 103 U/ml IFN-α. Cytokine production of IFN-γ, RANTES, MIP1-α, and MIG was evaluated by ELISA (G). Data shown are representative of two experiments with similar results. The means and SD of triplicate determination are shown for all panels.

FIGURE 3.

Quantitative PCR and ELISA results show differences in gene expression and cytokine production for IFN-α– and IL-2–responsive genes after treatment with sorafenib. Freshly isolated PBMCs were treated with 10 μM sorafenib for 30 min and stimulated with either 105 U/ml IFN-α or 8 nM IL-2 for 4 h at 37°C. Total RNA was isolated from treated cells and converted to cDNA. Real-time PCR was performed for the IFN-α–responsive genes OAS1 (A), IFIT2 (B), CD69 (C), CXCL10 (D), and SOCS1 (E). The IL-2–responsive genes PIM-1, CIS, and IFN-γ (F) were also examined. Induction of gene expression was calculated relative to β-actin and GAPDH. Data shown are representative of two independent experiments with similar results (C–E) or one independent experiment (A, B, and F). The mean and SD of triplicate determination are shown for all panels. Freshly isolated NK cells were cocultured with K562 cells, pretreated with 20 μM sorafenib, and then stimulated with 103 U/ml IFN-α. Cytokine production of IFN-γ, RANTES, MIP1-α, and MIG was evaluated by ELISA (G). Data shown are representative of two experiments with similar results. The means and SD of triplicate determination are shown for all panels.

Close modal

The IL-2 receptor is present on several immune subsets, including T cells and NK cells. IFN-α stimulates JAK–STAT signal transduction in these subsets as well as in monocytes and B cells. To determine whether sorafenib could suppress STAT phosphorylation in each of these immune subsets, dual parameter flow cytometry was employed. Freshly isolated PBMCs from normal donors were treated with sorafenib for 30 min and then stimulated with IL-2 or IFN-α for 15 min. The level of STAT5 phosphorylation was then assessed in the CD3-, CD4-, CD8-, and CD56-gated populations. Pretreatment of PBMCs with sorafenib significantly inhibited the IL-2–induced activation of STAT5 in T cells and NK cells (p < 0.01) (Fig. 4A). As previously shown by our group, CD14+ and CD21+ cells do not phosphorylate STAT5 in response to IL-2 stimulation and therefore they were not examined in this context (14). Similarly, sorafenib pretreatment markedly inhibited the phosphorylation of STAT1 in response to IFN-α in the CD3, CD4, CD8, CD21, CD14, and CD56 immune subsets (Fig. 4B). This same pattern of results was observed when sorafenib- and DMSO-treated immune cells were compared with respect to the percentage of cells that exhibited positive staining for phosphorylated STAT5 and STAT1 after cytokine exposure (data not shown). STAT1 phosphorylation was also inhibited in CD34+ cells isolated from PBMCs (Fig. 4B).

FIGURE 4.

Differences in p-STAT1 and p-STAT5 levels between control-treated and sorafenib-treated cells via flow cytometry. Freshly isolated PBMCs were treated with DMSO or 10 μM sorafenib for 30 min at 37°C and stimulated with 8 nM IL-2 or 105 U/ml IFN-α for 15 min at 37°C. Phosphorylated STAT5-specific fluorescence (A) was measured by intracellular flow cytometry within the CD3, CD4, CD8, and CD56 immune subsets. Phosphorylated STAT1-specific fluorescence was measured by intracellular flow cytometry within the CD3, CD4, CD8, CD21, CD14, CD56, and CD34 immune subsets (B). CD4+ (C) and CD8+ T cells (D) (central, effector, naive) were purified from fresh leukopacks and pretreated for 30 min with 20 μM sorafenib, stimulated for 15 min with 8 nM IL-2 or 105 mU/ml IFN-α, and levels of activated STAT1 or STAT5 (respectively) were measured by intracellular flow cytometry. A similar experiment was conducted with NK cells (tolerant, regulatory, cytotoxic) (E). Raw gating examples are shown in Supplemental Fig. 3. There was a statistically significant difference between DMSO and sorafenib treatment (A–E) (p < 0.05). Proliferation of stimulated CD4+ T cells is indicated by the presence of daughter peaks on the histogram, which represents the CFSE-labeled cells that become more dilute with each generation. Pretreatment of CD4+ T cells with sorafenib (10 or 20 μM) led to marked inhibition of proliferation. Identical results were obtained with CD8+ T cells (F). NK-92 proliferation was inhibited by 97% with sorafenib pretreatment versus a vehicle control (G). T cells were cocultured for 48 h with 10 or 20 μM sorafenib, stimulated with CD3/CD28 beads, and IFN-γ levels in culture supernatants were measured at 72 h by ELISA (H). There was a statistically significant difference between DMSO and 20 μM sorafenib treatment (p < 0.0002). Data shown are the means ± SEM of three (A–E) or two (F and H) independent experiments. A similar experiment was conducted with NK-92 cells stimulated with IL-2 plus IL-15 (I). Data shown are the results of one independent experiment (I).

FIGURE 4.

Differences in p-STAT1 and p-STAT5 levels between control-treated and sorafenib-treated cells via flow cytometry. Freshly isolated PBMCs were treated with DMSO or 10 μM sorafenib for 30 min at 37°C and stimulated with 8 nM IL-2 or 105 U/ml IFN-α for 15 min at 37°C. Phosphorylated STAT5-specific fluorescence (A) was measured by intracellular flow cytometry within the CD3, CD4, CD8, and CD56 immune subsets. Phosphorylated STAT1-specific fluorescence was measured by intracellular flow cytometry within the CD3, CD4, CD8, CD21, CD14, CD56, and CD34 immune subsets (B). CD4+ (C) and CD8+ T cells (D) (central, effector, naive) were purified from fresh leukopacks and pretreated for 30 min with 20 μM sorafenib, stimulated for 15 min with 8 nM IL-2 or 105 mU/ml IFN-α, and levels of activated STAT1 or STAT5 (respectively) were measured by intracellular flow cytometry. A similar experiment was conducted with NK cells (tolerant, regulatory, cytotoxic) (E). Raw gating examples are shown in Supplemental Fig. 3. There was a statistically significant difference between DMSO and sorafenib treatment (A–E) (p < 0.05). Proliferation of stimulated CD4+ T cells is indicated by the presence of daughter peaks on the histogram, which represents the CFSE-labeled cells that become more dilute with each generation. Pretreatment of CD4+ T cells with sorafenib (10 or 20 μM) led to marked inhibition of proliferation. Identical results were obtained with CD8+ T cells (F). NK-92 proliferation was inhibited by 97% with sorafenib pretreatment versus a vehicle control (G). T cells were cocultured for 48 h with 10 or 20 μM sorafenib, stimulated with CD3/CD28 beads, and IFN-γ levels in culture supernatants were measured at 72 h by ELISA (H). There was a statistically significant difference between DMSO and 20 μM sorafenib treatment (p < 0.0002). Data shown are the means ± SEM of three (A–E) or two (F and H) independent experiments. A similar experiment was conducted with NK-92 cells stimulated with IL-2 plus IL-15 (I). Data shown are the results of one independent experiment (I).

Close modal

Examination of different subsets of T lymphocytes and NK cells showed that sorafenib pretreatment markedly inhibited the phosphorylation of STAT1 (in response to IFN-α) and STAT5 (in response to IL-2) in different subsets of each cell type. CD4+ (Fig. 4C) and CD8+ (Fig. 4D) T cells (central, effector, naive) were isolated and purified from fresh leukopacks, pretreated for 30 min with sorafenib, and stimulated for 15 min with IL-2 or IFN-α. Levels of activated STAT1 or STAT5 (respectively) were measured by intracellular flow cytometry in gated populations. A similar experiment was conducted with NK cells (Fig. 4E) to examine the effect of sorafenib on the tolerant, regulatory, and cytotoxic subsets.

Pretreatment of CD4 T cells and CD8 T cells with 10 or 20 μM sorafenib led to a >90% decrease in proliferation following stimulation with CD3/CD28 beads versus a control treatment (Fig. 4F). Human NK cell proliferation in response to cytokine stimulation was minimal and only slightly affected by sorafenib pretreatment (data not shown). This proliferation experiment was repeated using the IL-2–dependent NK cell line NK-92. Pretreatment of NK-92 cells with sorafenib (10 or 20 μM) led to a 97% decrease in IL-2–induced proliferation as compared with a vehicle control (Fig. 4G). Gating examples are given in Supplemental Figs. 3 and 4.

Sorafenib pretreatment also significantly inhibited the ability of CD4+ and CD8+ T cells to secrete IFN-γ at 72 h in response to stimulation with CD3/CD28 beads (Fig. 4H). Likewise, sorafenib pretreatment markedly inhibited IFN-γ production by the NK-92 cell line at 72 h (Fig. 4I).

We next examined the effects of sorafenib on cytokine signaling in regulatory T cells, which are phenotypically defined as cells that express CD4, CD25, and FOXP3. CD4+CD25+ cells were isolated from whole PBMCs, treated with sorafenib for 30 min, and then stimulated with IFN-α or IL-2. Phosphorylation of STAT1 in response to IFN-α and phosphorylation of STAT5 in response to IL-2 were then evaluated by flow cytometry. Cytokine-induced activation of JAK–STAT signal transduction in CD4+CD25+ T cells was completely abrogated by sorafenib pretreatment (Fig. 5A, 5B). Approximately 85% of this cell population expressed FOXP3, as determined by intracellular flow cytometry performed on an aliquot of the purified cells. MDSCs are another type of inhibitory immune cell that may be affected by sorafenib. The effect of sorafenib on IFN-α–induced activation of STAT1 in this population of cells was examined in MDSCs from tumor-bearing mice and also MDSCs from patients with metastatic RCC. Murine MDSCs pretreated with sorafenib exhibited a 10-fold decrease in phosphorylation of STAT1 (as measured by Fsp) compared with controls (p < 0.001) (Fig. 5C). Sorafenib had similar effects on the activation of STAT1 in IFN-α–treated human MDSCs. Whereas the effects of sorafenib were demonstrable in all MDSCs tested, the existence of interpatient variability led to a nonsignificant result (Fig. 5D). A gating example is given in Supplemental Fig. 4.

FIGURE 5.

Sorafenib decreases cytokine signaling in immunosuppressive cells. Freshly isolated PBMCs were treated with DMSO or 10 μM sorafenib for 30 min at 37°C and stimulated with 8 nM IL-2 or 105 U/ml IFN-α for 15 min at 37°C. The specific fluorescence of p-STAT5 (A) and p-STAT1 (B) were measured in the CD4+CD25+ subset. There was a statistically significant difference between DMSO and sorafenib treatment (A and B) (p < 0.05). Murine MDSCs were treated with DMSO or 10 μM sorafenib for 30 min at 37°C and stimulated with PBS, 103 U/ml IFN-α, or 104 U/ml IFN-α for 15 min at 37°C. Phosphorylated STAT1-specific fluorescence was measured by intracellular flow cytometry. There was a statistically significant difference between DMSO and sorafenib treatment (C) (p < 0.001). MDSCs isolated from RCC patients were treated with DMSO or 10 μM sorafenib for 30 min at 37°C and stimulated with PBS, 103 U/ml IFN-α, or 104 U/ml IFN-α for 15 min at 37°C. Phosphorylated STAT1-specific fluorescence was measured by intracellular flow cytometry, and the figure shows the data from one representative patient. Other patients demonstrated a range of 14–66% decrease in phosphorylation of STAT1 following sorafenib treatment compared with DMSO, but the existence of interpatient variability led to the results not being statistically significant (D). Data shown are the means ± SEM of five (C), three (D), or two (A and B) independent experiments.

FIGURE 5.

Sorafenib decreases cytokine signaling in immunosuppressive cells. Freshly isolated PBMCs were treated with DMSO or 10 μM sorafenib for 30 min at 37°C and stimulated with 8 nM IL-2 or 105 U/ml IFN-α for 15 min at 37°C. The specific fluorescence of p-STAT5 (A) and p-STAT1 (B) were measured in the CD4+CD25+ subset. There was a statistically significant difference between DMSO and sorafenib treatment (A and B) (p < 0.05). Murine MDSCs were treated with DMSO or 10 μM sorafenib for 30 min at 37°C and stimulated with PBS, 103 U/ml IFN-α, or 104 U/ml IFN-α for 15 min at 37°C. Phosphorylated STAT1-specific fluorescence was measured by intracellular flow cytometry. There was a statistically significant difference between DMSO and sorafenib treatment (C) (p < 0.001). MDSCs isolated from RCC patients were treated with DMSO or 10 μM sorafenib for 30 min at 37°C and stimulated with PBS, 103 U/ml IFN-α, or 104 U/ml IFN-α for 15 min at 37°C. Phosphorylated STAT1-specific fluorescence was measured by intracellular flow cytometry, and the figure shows the data from one representative patient. Other patients demonstrated a range of 14–66% decrease in phosphorylation of STAT1 following sorafenib treatment compared with DMSO, but the existence of interpatient variability led to the results not being statistically significant (D). Data shown are the means ± SEM of five (C), three (D), or two (A and B) independent experiments.

Close modal

We also wanted to examine the effects of sorafenib treatment on target and immune cells. Surface expression of MHC class I A, B, and C, MICA/B, NKG2D, and the receptors for IL-2 (IL-2Rα and IL-2Rβ) and IFN-α (IFNAR1) on K562 and NK cells were analyzed. The results are given in Table I and reveal that levels of MICA/B were diminished in K562 cells (2.1% expression down from 51.5% expression). This change could explain some of the reduced killing of sorafenib-treated cultures of NK cells plus K562 cells, but this effect would not impact IFN signaling in immune cells. Levels of IFNAR1 on K562 cells were also diminished (15.3% expression down from 85.4% expression). However, levels of the other markers on K562 cells were minimally affected. Notably, there was very little expression of MHC class I (HLA-A, -B, and -C) on untreated K562 cells, and its expression was unaffected by treatment with sorafenib. With respect to the subject of the effects of sorafenib on NK cells, there was almost no effect of sorafenib on the expression of these markers. A gating example is given in Supplemental Fig. 3.

Table I.
Effect of sorafenib on surface proteins
Cell TypeConditionIL-2RαIL-2RβMICA/BNKG2DIFNAR
K562 Control 3.78 31.8 51.5 11.7 85.4 
Sorafenib 1.84 3.07 2.07 2.82 15.3 
NK cells Control 15.9 61.8 26.1 71.4 72.2 
Sorafenib 5.83 45.3 13.1 69.1 58.4 
Cell TypeConditionIL-2RαIL-2RβMICA/BNKG2DIFNAR
K562 Control 3.78 31.8 51.5 11.7 85.4 
Sorafenib 1.84 3.07 2.07 2.82 15.3 
NK cells Control 15.9 61.8 26.1 71.4 72.2 
Sorafenib 5.83 45.3 13.1 69.1 58.4 

NK cells were freshly isolated from normal donors, and K562 cells were cultured until at least 80% confluent before use. Each cell type was treated with either nothing, DMSO control, or sorafenib. After 48 h, cells were collected and stained for the following: CD25 (IL-2Rα), CD122 (IL 2Rβ), MICA/B, NKG2D ligands, and the IFN-α receptor (IFNAR). Data are expressed as percentage positive cells and are the result of one independent experiment.

Similar to PBMCs, the pretreatment of the acute lymphoblastic leukemic cell line 697 or the Burkitt’s B cell lymphoma cell line Ramos with sorafenib led to inhibition of phosphorylated STAT1 in response to IFN-α (p < 0.05) (Fig. 6). Cytokine therapy with IL-2 and IFN-α is used in the treatment of advanced stage melanoma and RCC. Therefore, we sought to examine the effects of sorafenib on cytokine signaling in these cell types. The RCC lines Caki and SK-RC-45 and the melanoma cell lines A375 and HS294t were pretreated with sorafenib for 30 min and then stimulated with IFN-α. Sorafenib pretreatment did not inhibit the phosphorylation of STAT1 in response to IFN-α in any of these cell lines (Fig. 6). The lack of effect of sorafenib on cytokine signaling in RCC and melanoma cells is contrary to the results obtained with immune cells and lymphoma cell lines. This finding may represent the existence of constitutive signaling events in solid tumors.

FIGURE 6.

p-STAT1 differences in sorafenib versus control conditions on several cell lines measured via flow cytometry. B cell lymphoma cell lines 697 and Ramos, renal cell carcinoma cell lines Caki and SK-RC-45, and melanoma cell lines A374 and Hs294t were treated with DMSO or 10 μM sorafenib for 30 min at 37°C and then stimulated with 105 U/ml IFN-α for 15 min at 37°C. Phosphorylated STAT1 levels were measured by intracellular flow cytometry. Data are expressed as specific fluorescence (p < 0.05). Data shown are the means ± SEM of three independent experiments.

FIGURE 6.

p-STAT1 differences in sorafenib versus control conditions on several cell lines measured via flow cytometry. B cell lymphoma cell lines 697 and Ramos, renal cell carcinoma cell lines Caki and SK-RC-45, and melanoma cell lines A374 and Hs294t were treated with DMSO or 10 μM sorafenib for 30 min at 37°C and then stimulated with 105 U/ml IFN-α for 15 min at 37°C. Phosphorylated STAT1 levels were measured by intracellular flow cytometry. Data are expressed as specific fluorescence (p < 0.05). Data shown are the means ± SEM of three independent experiments.

Close modal

The in vivo effects of sorafenib on cytokine signaling were evaluated in patients with metastatic differentiated thyroid carcinoma who were receiving 400 mg sorafenib twice daily in the context of a phase II clinical trial (The Ohio State University no. 0441) (5). This is the Food and Drug Administration–approved dose that has been shown to provide adequate steady-state serum levels with a maximum concentration of 9.35 mg/l (23). Cryopreserved PBMCs isolated from patients at baseline and at 8 and 16 wk after the start of therapy were analyzed for levels of phosphorylated STAT5 and STAT1 following ex vivo stimulation with IL-2 and IFN-α. The phosphorylation of STAT5 after IL-2 treatment was reduced from a mean Fsp of 16.7 ± 6.7 at baseline to an Fsp of 12.3 ± 8.6 and 10.5 ± 2.8 at 8 and 16 wk of therapy, respectively (Fig. 7A). However, the linear trend was not statistically significant (p = 0.06). The phosphorylation of STAT1 in response to IFN-α did lead to a statistically significant decrease from a mean Fsp of 42.5 ± 36.2 at baseline to an Fsp of 23.7 ± 12.8 and 21.2 ± 11.5 at 8 and 16 wk of therapy, respectively (p = 0.02 for the linear trend of the log-transformed data) (Fig. 7B). Levels of regulatory T cells and MDSCs were measured in a subset of these patients. However, there was no discernable pattern of change over time (data not shown).

FIGURE 7.

Patient levels of p-STAT1 and p-STAT5 during therapy measured via flow cytometry. Cryopreserved PBMCs isolated from patients prior to therapy and 8 and 16 wk after the initiation of sorafenib therapy were stimulated with 8 nM IL-2 or 105 U/ml IFN-α. Phosphorylated STAT5 (A) and STAT1 (B) levels were measured by intracellular flow cytometry. Data were log transformed and are expressed as specific fluorescence (*p = 0.02). Data shown are the mean of four independent experiments.

FIGURE 7.

Patient levels of p-STAT1 and p-STAT5 during therapy measured via flow cytometry. Cryopreserved PBMCs isolated from patients prior to therapy and 8 and 16 wk after the initiation of sorafenib therapy were stimulated with 8 nM IL-2 or 105 U/ml IFN-α. Phosphorylated STAT5 (A) and STAT1 (B) levels were measured by intracellular flow cytometry. Data were log transformed and are expressed as specific fluorescence (*p = 0.02). Data shown are the mean of four independent experiments.

Close modal

We hypothesized that sorafenib would have off-target effects on IL-2 and IFN-α signal transduction via the JAK–STAT pathway in immune effector cells. Our investigations determined the following: first, therapeutic concentrations of sorafenib applied to PBMCs in vitro were sufficient to inhibit the phosphorylation of STAT1 and STAT5 in response to stimulation with IFN-α and IL-2, respectively, regardless of cytokine concentration. Second, the expression of genes normally stimulated by IFN-α and IL-2 and immune cell cytokine production were decreased by prior treatment with sorafenib. Third, sorafenib had effects on nearly all immune cell compartments, including immune cells with immunosuppressive actions such as regulatory T cells and MDSCs. Finally, sorafenib therapy in vivo led to decreased phosphorylation of STAT1 and STAT5 following stimulation with IFN-α and IL-2.

Although previous studies have demonstrated that sorafenib treatment decreases immune cell activation and function, none of these studies examined specific mechanisms such as the effects of sorafenib on cytokine signaling. Other studies have searched for evidence of the effects of sorafenib on the Raf/Ras/MEK pathway, the designed target of the drug. To our knowledge, the present study is the first to examine the direct effects of sorafenib therapy on the JAK–STAT signaling pathway in immune cells. Using intracellular flow cytometry, it was shown that pretreatment of PBMCs with therapeutic concentrations of sorafenib in vitro results in decreased phosphorylation of IL-2, IFN-α, and IFN-γ signaling intermediates. These findings were confirmed via immunoblot analysis, and decreased IL-6–induced phosphorylation of STAT3 was also demonstrated. Importantly, note that whereas the immunoblot results do not precisely correlate with the flow cytometry findings, these are two different assays with different methods employed to prepare the cells. Furthermore, it has been previously established that total levels of phosphorylated STAT1 as measured by flow cytometry correlated directly with the transcription of IFN-stimulated genes driven by the binding of STAT1 to specific promoter regions (16). Importantly, we have established that this phenomenon holds true for multiple immune subsets in vitro and in a small subset of cancer patients.

The inhibitory effects of sorafenib on IL-2 and IFN-α signaling are notable due to the importance of these cytokines in mounting an effective immune response, particularly as inhibition of signaling was observed at physiologically relevant concentrations of sorafenib. Although only signaling of IL-2, IFN-α, and IFN-γ was evaluated in the present study, inhibition of the activity of other cytokines that rely on signaling through the JAK–STAT pathway is possible. Indeed, exploratory studies revealed that sorafenib also inhibited IL-21–induced JAK–STAT signal transduction in PBMCs. The mechanism of inhibition likely involves effects on JAK activity. Recent observations demonstrated that whereas total levels of JAK1 were not affected by sorafenib, phosphorylation of JAK1 was markedly inhibited in glioblastoma cell lines (9). Similarly, IL-6–induced phosphorylation of STAT3 and JAK2 was inhibited by sorafenib in neuroblastoma cell lines (11). Despite extensive efforts, our analysis of JAK1 activity in sorafenib-treated cells did not produce conclusive evidence for inhibition of JAK1 activation in PBMCs. Based on the aforementioned studies, we propose that sorafenib inhibits cytokine-induced JAK–STAT activation via inhibition of JAK. The action of sorafenib on many of its target kinases has been shown to be reversible (2). However, whereas some immune cells were able to resume cytokine signaling following exposure to sorafenib and a subsequent washing, a smaller fraction of these cells responded to cytokine stimulation than immune cells that had not been pretreated with sorafenib (Fig. 2D). This finding suggested that the inhibition of this pathway does not readily resolve upon removal of sorafenib. It is thus possible that sorafenib may inhibit baseline immune function, as well as cytokine-based immunotherapy. In contrast, the B-raf inhibitor PLX4032 has been shown to have minimal effects on the phosphorylation of STAT1, STAT3, STAT5, and STAT6 in CD8+ T cells (24).

Recent studies exploring the negative effects of pharmacologically relevant concentrations of sorafenib upon T lymphocytes and dendritic cells have shown effects on several pathways. Sorafenib has been found to inhibit the proliferation of CD4+ and CD8+ T cells (25). Studies have shown that in vitro concentrations of sorafenib comparable to levels observed in patients strongly impaired the activation of T cells in both normal donors and patients with metastatic renal cell carcinoma and melanoma (26) (27). However, several studies have also shown that sorafenib can activate T cells. For example, subpharmacologic doses of sorafenib selectively increase CD4+ T cell activation and block the inhibitory effects of regulatory T cells in patients with HCC (28). Chuang et al. (29) showed that serial low doses of sorafenib enhanced the migration of CD8+ T cells through inhibition of STAT3.

Although the latter studies seem to be at odds with the present study, note that this group employed sorafenib at doses that were much lower than those used in the present study and lower than has been reported in patients being treated with sorafenib. According to a previous analysis, sorafenib concentrations in patients normally range from 6 to 12 μM (pharmacologic concentration) (30). In the aforementioned study, concentrations of sorafenib that ranged from 0.1 to 3 μM were used to represent subpharmacologic concentrations. The effects of sorafenib on T effector responses were found to be dose-dependent (28). Subpharmacologic concentrations of sorafenib resulted in effector T cell activation in patients with HCC, but pharmacologic doses of sorafenib decreased T effector activation by downregulating CD25 surface expression.

This is consistent with other data showing that sorafenib treatment has been found to decrease the expression of CD25 and CD69 on T cells following stimulation with PHA, and it also inhibited the expression of class II MHC and CD80 on dendritic cells following stimulation with LPS. The decreased expression of activation markers was accompanied by decreased phosphorylation of Lck and Erk in T cells and decreased activity of the MyD88 pathway in dendritic cells (27). Sorafenib pretreatment also inhibited cytokine production by stimulated T cells (IL-2) and dendritic cells (IL-6, IL-12, IL-10, and TNF-α) (31). Interestingly, in another study, sorafenib was found to reverse the inhibitory effects of vascular endothelial growth factor on the transition of immature myeloid cells into dendritic cells (32). In vivo studies examining the effect of sorafenib on T cell–mediated responses demonstrated a decrease in the severity of contact dermatitis and a decreased induction of Ag-specific CD8+ T cells after immunization, suggesting that the inhibition of signaling in immune cells had functional consequences in vivo (31). Thus, multiple investigations have revealed that sorafenib can alter immune cell functioning. Our study now demonstrates that cytokine signaling is also impaired, thus providing a potential mechanism for these observations.

Several clinical trials have been conducted to evaluate the safety and potential efficacy of combining sorafenib treatment with IFN-α-2b therapy. A recent phase II trial comparing sorafenib therapy to the combination of sorafenib and low-dose IFN-α-2b failed to identify any difference in efficacy between the two arms (33). In the context of a phase I trial combining sorafenib therapy with increasing doses of IFN-α-2b, Escudier et al. (34) observed no significant reduction in the phosphorylation of ERK in lymphocytes obtained from patients receiving sorafenib and stimulated ex vivo with PMA, which is consistent with results seen in a study examining patients with solid tumors who received single agent sorafenib (35). However, as sorafenib is a promiscuous inhibitor, changes in ERK phosphorylation may not be representative of the full range of effects of sorafenib. Houben et al. (26) showed that the presence of sorafenib in culture did not affect ERK phosphorylation following T cell stimulation through the TCR. However, this pretreatment led to impaired T cell activation (as measured by CD25 expression) in response to stimulation via the TCR. Sorafenib may therefore be inhibiting other downstream signaling pathways emanating from the TCR (e.g., phosphorylation of STAT5). Thus, the combination of T cell–based therapies with target agents such as sorafenib may lead to unintended downregulation of the specific immune response.

Much focus has been placed on the presence and function of regulatory T cells in patients with various malignancies. Increased numbers of regulatory T cells in tumors have been associated with decreased patient survival, illustrating the impact that these immunosuppressive cells may have on cancer progression (3638). Previous studies have shown the inhibitory effect of sorafenib on the number of regulatory T cells (CD4+CD25+FOXP3+) in the peripheral blood of patients receiving oral sorafenib therapy and in a murine model of hepatocellular carcinoma (39, 40). Studies have shown that IL-2–induced activation of STAT5 is required for the expansion and maintenance of regulatory T cells in the periphery and that administration of IL-2 leads to expansion of this cell population and increased expression of FOXP3 (4145). Disruption of the IL-2 signaling cascade by sorafenib could abrogate this effect of IL-2 immunotherapy. In the present study, we showed that IL-2–induced phosphorylation of STAT5 is disrupted in regulatory T cells following exposure to sorafenib. Several groups have shown the importance of this signaling pathway in regulatory T cell development, as STAT5a and STAT5b null mice do not develop regulatory T cells, and mice deficient in IL-2 have decreased numbers of CD4+CD25+FOXP3+ cells (4648) The presence of IL-2 and functional JAK–STAT signaling components is also required for the expression of Foxp3 and the enhancement of their suppressive actions (49, 50). Our results suggest that the IL-2–mediated activity of CD25+ regulatory T cells could be inhibited in patients receiving sorafenib therapy, thus potentially relieving some of the immunosuppressive activity of this immune cell subset. Other small molecule inhibitors (e.g., sunitinib) have been shown to have similar effects on regulatory T cells (51). We saw a similar decrease in cytokine signaling in MDSCs, which represent another population of immunosuppressive cells. Similar to regulatory T cells, this population of cells was also decreased in a mouse model of hepatocellular carcinoma treated with sorafenib (40).

Although the number of repetitions testing unique patient samples in our studies is modest, the present study demonstrated that sorafenib inhibits signal transduction in response to exogenous cytokines, which may also extend to endogenous cytokine signaling. Sorafenib also has inhibitory effects on key immune cells, including immunosuppressive immune subsets. The timing of sorafenib administration must therefore be carefully considered if it is to be used in combination with immunostimulatory cytokines. Also, sorafenib could be employed as a means of altering the activities of inhibitory immune cell subsets, such as regulatory T cells and MDSCs.

We thank The Ohio State University Comprehensive Cancer Center Analytical Cytometry Shared Resource and Nucleic Acid Shared Resource. We thank Dr. Chubul M. I. Ahmed as well as Dr. Howard M. Johnson of the University of Florida for assistance.

This work was supported by National Institutes of Health Grants P01 CA095426 (to M.A.C.), P30 CA16058 (to M.A.C.), T32 CA090223 (to W.E.C.), T32 CA009338 (to M.A.C.), T32 GM068412 (to A.C.J.-R. and B.L.M.-B.), K24 CA093670 (to W.E.C.), K22 CA134551 (to G.B.L.), and by a Valvano Foundation for Cancer Research award (to G.B.L.).

The online version of this article contains supplemental material.

Abbreviations used in this article:

CIS

cytokine-inducible SH2 domain–containing protein

Fsp

specific fluorescence

HCC

hepatocellular carcinoma

IFIT2

IFN-induced protein with tetratricopeptide repeats 2

IFNAR

IFN-α receptor

MDSC

myeloid-derived suppressor cell

OAS1

2′,5′-oligoadenylate synthetase 1

RCC

renal cell carcinoma

SOCS1

suppressor of cytokine signaling 1.

1
Wilhelm
S.
,
Carter
C.
,
Lynch
M.
,
Lowinger
T.
,
Dumas
J.
,
Smith
R. A.
,
Schwartz
B.
,
Simantov
R.
,
Kelley
S.
.
2006
.
Discovery and development of sorafenib: a multikinase inhibitor for treating cancer.
Nat. Rev. Drug Discov.
5
:
835
844
.
2
Strumberg
D.
2005
.
Preclinical and clinical development of the oral multikinase inhibitor sorafenib in cancer treatment.
Drugs Today (Barc)
41
:
773
784
.
3
Liu
L.
,
Cao
Y.
,
Chen
C.
,
Zhang
X.
,
McNabola
A.
,
Wilkie
D.
,
Wilhelm
S.
,
Lynch
M.
,
Carter
C.
.
2006
.
Sorafenib blocks the RAF/MEK/ERK pathway, inhibits tumor angiogenesis, and induces tumor cell apoptosis in hepatocellular carcinoma model PLC/PRF/5.
Cancer Res.
66
:
11851
11858
.
4
Soares
P.
,
Trovisco
V.
,
Rocha
A. S.
,
Lima
J.
,
Castro
P.
,
Preto
A.
,
Máximo
V.
,
Botelho
T.
,
Seruca
R.
,
Sobrinho-Simões
M.
.
2003
.
BRAF mutations and RET/PTC rearrangements are alternative events in the etiopathogenesis of PTC.
Oncogene
22
:
4578
4580
.
5
Kloos
R. T.
,
Ringel
M. D.
,
Knopp
M. V.
,
Hall
N. C.
,
King
M.
,
Stevens
R.
,
Liang
J.
,
Wakely
P. E.
 Jr.
,
Vasko
V. V.
,
Saji
M.
, et al
.
2009
.
Phase II trial of sorafenib in metastatic thyroid cancer.
J. Clin. Oncol.
27
:
1675
1684
.
6
Ahmed
M.
,
Barbachano
Y.
,
Riddell
A.
,
Hickey
J.
,
Newbold
K. L.
,
Viros
A.
,
Harrington
K. J.
,
Marais
R.
,
Nutting
C. M.
.
2011
.
Analysis of the efficacy and toxicity of sorafenib in thyroid cancer: a phase II study in a UK based population.
Eur. J. Endocrinol.
165
:
315
322
.
7
Chang
Y. S.
,
Adnane
J.
,
Trail
P. A.
,
Levy
J.
,
Henderson
A.
,
Xue
D.
,
Bortolon
E.
,
Ichetovkin
M.
,
Chen
C.
,
McNabola
A.
, et al
.
2007
.
Sorafenib (BAY 43-9006) inhibits tumor growth and vascularization and induces tumor apoptosis and hypoxia in RCC xenograft models.
Cancer Chemother. Pharmacol.
59
:
561
574
.
8
Rahmani
M.
,
Nguyen
T. K.
,
Dent
P.
,
Grant
S.
.
2007
.
The multikinase inhibitor sorafenib induces apoptosis in highly imatinib mesylate-resistant Bcr/Abl+ human leukemia cells in association with signal transducer and activator of transcription 5 inhibition and myeloid cell leukemia-1 down-regulation.
Mol. Pharmacol.
72
:
788
795
.
9
Yang
F.
,
Brown
C.
,
Buettner
R.
,
Hedvat
M.
,
Starr
R.
,
Scuto
A.
,
Schroeder
A.
,
Jensen
M.
,
Jove
R.
.
2010
.
Sorafenib induces growth arrest and apoptosis of human glioblastoma cells through the dephosphorylation of signal transducers and activators of transcription 3.
Mol. Cancer Ther.
9
:
953
962
.
10
Yang
F.
,
Van Meter
T. E.
,
Buettner
R.
,
Hedvat
M.
,
Liang
W.
,
Kowolik
C. M.
,
Mepani
N.
,
Mirosevich
J.
,
Nam
S.
,
Chen
M. Y.
, et al
.
2008
.
Sorafenib inhibits signal transducer and activator of transcription 3 signaling associated with growth arrest and apoptosis of medulloblastomas.
Mol. Cancer Ther.
7
:
3519
3526
.
11
Yang
F.
,
Jove
V.
,
Buettner
R.
,
Xin
H.
,
Wu
J.
,
Wang
Y.
,
Nam
S.
,
Xu
Y.
,
Ara
T.
,
DeClerck
Y. A.
, et al
.
2012
.
Sorafenib inhibits endogenous and IL-6/S1P induced JAK2-STAT3 signaling in human neuroblastoma, associated with growth suppression and apoptosis.
Cancer Biol. Ther.
13
:
534
541
.
12
Llovet
J. M.
,
Ricci
S.
,
Mazzaferro
V.
,
Hilgard
P.
,
Gane
E.
,
Blanc
J. F.
,
de Oliveira
A. C.
,
Santoro
A.
,
Raoul
J. L.
,
Forner
A.
, et al
SHARP Investigators Study Group
.
2008
.
Sorafenib in advanced hepatocellular carcinoma.
N. Engl. J. Med.
359
:
378
390
.
13
Chong
D. Q.
,
Tan
I. B.
,
Choo
S. P.
,
Toh
H. C.
.
2013
.
The evolving landscape of therapeutic drug development for hepatocellular carcinoma.
Contemp. Clin. Trials
36
:
605
615
.
14
Varker
K. A.
,
Kondadasula
S. V.
,
Go
M. R.
,
Lesinski
G. B.
,
Ghosh-Berkebile
R.
,
Lehman
A.
,
Monk
J. P.
,
Olencki
T.
,
Kendra
K.
,
Carson
W. E.
 3rd
.
2006
.
Multiparametric flow cytometric analysis of signal transducer and activator of transcription 5 phosphorylation in immune cell subsets in vitro and following interleukin-2 immunotherapy.
Clin. Cancer Res.
12
:
5850
5858
.
15
Gong
J. H.
,
Maki
G.
,
Klingemann
H. G.
.
1994
.
Characterization of a human cell line (NK-92) with phenotypical and functional characteristics of activated natural killer cells.
Leukemia
8
:
652
658
.
16
Lesinski
G. B.
,
Kondadasula
S. V.
,
Crespin
T.
,
Shen
L.
,
Kendra
K.
,
Walker
M.
,
Carson
W. E.
 III
.
2004
.
Multiparametric flow cytometric analysis of inter-patient variation in STAT1 phosphorylation following interferon Alfa immunotherapy.
J. Natl. Cancer Inst.
96
:
1331
1342
.
17
Vermes
I.
,
Haanen
C.
,
Steffens-Nakken
H.
,
Reutelingsperger
C.
.
1995
.
A novel assay for apoptosis. Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled Annexin V.
J. Immunol. Methods
184
:
39
51
.
18
Luedke
E.
,
Jaime-Ramirez
A. C.
,
Bhave
N.
,
Roda
J.
,
Choudhary
M. M.
,
Kumar
B.
,
Teknos
T. N.
,
Carson
W. E.
 III
.
2012
.
Cetuximab therapy in head and neck cancer: immune modulation with interleukin-12 and other natural killer cell-activating cytokines.
Surgery
152
:
431
440
.
19
Fu
B.
,
Tian
Z.
,
Wei
H.
.
2014
.
Subsets of human natural killer cells and their regulatory effects.
Immunology
141
:
483
489
.
20
Mundy-Bosse
B. L.
,
Young
G. S.
,
Bauer
T.
,
Binkley
E.
,
Bloomston
M.
,
Bill
M. A.
,
Bekaii-Saab
T.
,
Carson
W. E.
 III
,
Lesinski
G. B.
.
2011
.
Distinct myeloid suppressor cell subsets correlate with plasma IL-6 and IL-10 and reduced interferon-alpha signaling in CD4+ T cells from patients with GI malignancy.
Cancer Immunol. Immunother.
60
:
1269
1279
.
21
Raulet
D. H.
2004
.
Interplay of natural killer cells and their receptors with the adaptive immune response.
Nat. Immunol.
5
:
996
1002
.
22
Parihar
R.
,
Dierksheide
J.
,
Hu
Y.
,
Carson
W. E.
.
2002
.
IL-12 enhances the natural killer cell cytokine response to Ab-coated tumor cells.
J. Clin. Invest.
110
:
983
992
.
23
Strumberg
D.
,
Richly
H.
,
Hilger
R. A.
,
Schleucher
N.
,
Korfee
S.
,
Tewes
M.
,
Faghih
M.
,
Brendel
E.
,
Voliotis
D.
,
Haase
C. G.
, et al
.
2005
.
Phase I clinical and pharmacokinetic study of the novel Raf kinase and vascular endothelial growth factor receptor inhibitor BAY 43-9006 in patients with advanced refractory solid tumors.
J. Clin. Oncol.
23
:
965
972
.
24
Comin-Anduix
B.
,
Chodon
T.
,
Sazegar
H.
,
Matsunaga
D.
,
Mock
S.
,
Jalil
J.
,
Escuin-Ordinas
H.
,
Chmielowski
B.
,
Koya
R. C.
,
Ribas
A.
.
2010
.
The oncogenic BRAF kinase inhibitor PLX4032/RG7204 does not affect the viability or function of human lymphocytes across a wide range of concentrations.
Clin. Cancer Res.
16
:
6040
6048
.
25
Molhoek
K. R.
,
McSkimming
C. C.
,
Olson
W. C.
,
Brautigan
D. L.
,
Slingluff
C. L.
 Jr.
2009
.
Apoptosis of CD4+CD25high T cells in response to sirolimus requires activation of T cell receptor and is modulated by IL-2.
Cancer Immunol. Immunother.
58
:
867
876
.
26
Houben
R.
,
Voigt
H.
,
Noelke
C.
,
Hofmeister
V.
,
Becker
J. C.
,
Schrama
D.
.
2009
.
MAPK-independent impairment of T-cell responses by the multikinase inhibitor sorafenib.
Mol. Cancer Ther.
8
:
433
440
.
27
Zhao
W.
,
Gu
Y. H.
,
Song
R.
,
Qu
B. Q.
,
Xu
Q.
.
2008
.
Sorafenib inhibits activation of human peripheral blood T cells by targeting LCK phosphorylation.
Leukemia
22
:
1226
1233
.
28
Cabrera
R.
,
Ararat
M.
,
Xu
Y.
,
Brusko
T.
,
Wasserfall
C.
,
Atkinson
M. A.
,
Chang
L. J.
,
Liu
C.
,
Nelson
D. R.
.
2013
.
Immune modulation of effector CD4+ and regulatory T cell function by sorafenib in patients with hepatocellular carcinoma.
Cancer Immunol. Immunother.
62
:
737
746
.
29
Chuang
H. Y.
,
Chang
Y. F.
,
Liu
R. S.
,
Hwang
J. J.
.
2014
.
Serial low doses of sorafenib enhance therapeutic efficacy of adoptive T cell therapy in a murine model by improving tumor microenvironment.
PLoS One
9
:
e109992
.
30
Blanchet
B.
,
Billemont
B.
,
Cramard
J.
,
Benichou
A. S.
,
Chhun
S.
,
Harcouet
L.
,
Ropert
S.
,
Dauphin
A.
,
Goldwasser
F.
,
Tod
M.
.
2009
.
Validation of an HPLC-UV method for sorafenib determination in human plasma and application to cancer patients in routine clinical practice.
J. Pharm. Biomed. Anal.
49
:
1109
1114
.
31
Hipp
M. M.
,
Hilf
N.
,
Walter
S.
,
Werth
D.
,
Brauer
K. M.
,
Radsak
M. P.
,
Weinschenk
T.
,
Singh-Jasuja
H.
,
Brossart
P.
.
2008
.
Sorafenib, but not sunitinib, affects function of dendritic cells and induction of primary immune responses.
Blood
111
:
5610
5620
.
32
Alfaro
C.
,
Suarez
N.
,
Gonzalez
A.
,
Solano
S.
,
Erro
L.
,
Dubrot
J.
,
Palazon
A.
,
Hervas-Stubbs
S.
,
Gurpide
A.
,
Lopez-Picazo
J. M.
, et al
.
2009
.
Influence of bevacizumab, sunitinib and sorafenib as single agents or in combination on the inhibitory effects of VEGF on human dendritic cell differentiation from monocytes.
Br. J. Cancer
100
:
1111
1119
.
33
Jonasch
E.
,
Corn
P.
,
Pagliaro
L. C.
,
Warneke
C. L.
,
Johnson
M. M.
,
Tamboli
P.
,
Ng
C.
,
Aparicio
A.
,
Ashe
R. G.
,
Wright
J. J.
,
Tannir
N. M.
.
2010
.
Upfront, randomized, phase 2 trial of sorafenib versus sorafenib and low-dose interferon alfa in patients with advanced renal cell carcinoma: clinical and biomarker analysis.
Cancer
116
:
57
65
.
34
Escudier
B.
,
Lassau
N.
,
Angevin
E.
,
Soria
J. C.
,
Chami
L.
,
Lamuraglia
M.
,
Zafarana
E.
,
Landreau
V.
,
Schwartz
B.
,
Brendel
E.
, et al
.
2007
.
Phase I trial of sorafenib in combination with IFN α-2a in patients with unresectable and/or metastatic renal cell carcinoma or malignant melanoma.
Clin. Cancer Res.
13
:
1801
1809
.
35
Tong
F. K.
,
Chow
S.
,
Hedley
D.
.
2006
.
Pharmacodynamic monitoring of BAY 43-9006 (sorafenib) in phase I clinical trials involving solid tumor and AML/MDS patients, using flow cytometry to monitor activation of the ERK pathway in peripheral blood cells.
Cytometry B Clin. Cytom.
70
:
107
114
.
36
Fu
J.
,
Xu
D.
,
Liu
Z.
,
Shi
M.
,
Zhao
P.
,
Fu
B.
,
Zhang
Z.
,
Yang
H.
,
Zhang
H.
,
Zhou
C.
, et al
.
2007
.
Increased regulatory T cells correlate with CD8 T-cell impairment and poor survival in hepatocellular carcinoma patients.
Gastroenterology
132
:
2328
2339
.
37
Curiel
T. J.
,
Coukos
G.
,
Zou
L.
,
Alvarez
X.
,
Cheng
P.
,
Mottram
P.
,
Evdemon-Hogan
M.
,
Conejo-Garcia
J. R.
,
Zhang
L.
,
Burow
M.
, et al
.
2004
.
Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival.
Nat. Med.
10
:
942
949
.
38
Siddiqui
S. A.
,
Frigola
X.
,
Bonne-Annee
S.
,
Mercader
M.
,
Kuntz
S. M.
,
Krambeck
A. E.
,
Sengupta
S.
,
Dong
H.
,
Cheville
J. C.
,
Lohse
C. M.
, et al
.
2007
.
Tumor-infiltrating Foxp3CD4+CD25+ T cells predict poor survival in renal cell carcinoma.
Clin. Cancer Res.
13
:
2075
2081
.
39
Busse
A.
,
Asemissen
A. M.
,
Nonnenmacher
A.
,
Braun
F.
,
Ochsenreither
S.
,
Stather
D.
,
Fusi
A.
,
Schmittel
A.
,
Miller
K.
,
Thiel
E.
,
Keilholz
U.
.
2011
.
Immunomodulatory effects of sorafenib on peripheral immune effector cells in metastatic renal cell carcinoma.
Eur. J. Cancer
47
:
690
696
.
40
Cao
M.
,
Xu
Y.
,
Youn
J. I.
,
Cabrera
R.
,
Zhang
X.
,
Gabrilovich
D.
,
Nelson
D. R.
,
Liu
C.
.
2011
.
Kinase inhibitor sorafenib modulates immunosuppressive cell populations in a murine liver cancer model.
Lab. Invest.
91
:
598
608
.
41
Almeida
A. R.
,
Legrand
N.
,
Papiernik
M.
,
Freitas
A. A.
.
2002
.
Homeostasis of peripheral CD4+ T cells: IL-2Rα and IL-2 shape a population of regulatory cells that controls CD4+ T cell numbers.
J. Immunol.
169
:
4850
4860
.
42
Furtado
G. C.
,
Curotto de Lafaille
M. A.
,
Kutchukhidze
N.
,
Lafaille
J. J.
.
2002
.
Interleukin 2 signaling is required for CD4+ regulatory T cell function.
J. Exp. Med.
196
:
851
857
.
43
Klebb
G.
,
Autenrieth
I. B.
,
Haber
H.
,
Gillert
E.
,
Sadlack
B.
,
Smith
K. A.
,
Horak
I.
.
1996
.
Interleukin-2 is indispensable for development of immunological self-tolerance.
Clin. Immunol. Immunopathol.
81
:
282
286
.
44
Malek
T. R.
,
Yu
A.
,
Vincek
V.
,
Scibelli
P.
,
Kong
L.
.
2002
.
CD4 regulatory T cells prevent lethal autoimmunity in IL-2Rβ-deficient mice. Implications for the nonredundant function of IL-2.
Immunity
17
:
167
178
.
45
Antov
A.
,
Yang
L.
,
Vig
M.
,
Baltimore
D.
,
Van Parijs
L.
.
2003
.
Essential role for STAT5 signaling in CD25+CD4+ regulatory T cell homeostasis and the maintenance of self-tolerance.
J. Immunol.
171
:
3435
3441
.
46
Antony
P. A.
,
Paulos
C. M.
,
Ahmadzadeh
M.
,
Akpinarli
A.
,
Palmer
D. C.
,
Sato
N.
,
Kaiser
A.
,
Hinrichs
C. S.
,
Klebanoff
C. A.
,
Tagaya
Y.
,
Restifo
N. P.
.
2006
.
Interleukin-2-dependent mechanisms of tolerance and immunity in vivo.
J. Immunol.
176
:
5255
5266
.
47
Burchill
M. A.
,
Yang
J.
,
Vogtenhuber
C.
,
Blazar
B. R.
,
Farrar
M. A.
.
2007
.
IL-2 receptor beta-dependent STAT5 activation is required for the development of Foxp3+ regulatory T cells.
J. Immunol.
178
:
280
290
.
48
Yao
Z.
,
Kanno
Y.
,
Kerenyi
M.
,
Stephens
G.
,
Durant
L.
,
Watford
W. T.
,
Laurence
A.
,
Robinson
G. W.
,
Shevach
E. M.
,
Moriggl
R.
, et al
.
2007
.
Nonredundant roles for Stat5a/b in directly regulating Foxp3.
Blood
109
:
4368
4375
.
49
Fontenot
J. D.
,
Rasmussen
J. P.
,
Gavin
M. A.
,
Rudensky
A. Y.
.
2005
.
A function for interleukin 2 in Foxp3-expressing regulatory T cells.
Nat. Immunol.
6
:
1142
1151
.
50
Shevach
E. M.
,
DiPaolo
R. A.
,
Andersson
J.
,
Zhao
D. M.
,
Stephens
G. L.
,
Thornton
A. M.
.
2006
.
The lifestyle of naturally occurring CD4+ CD25+ Foxp3+ regulatory T cells.
Immunol. Rev.
212
:
60
73
.
51
Finke
J. H.
,
Rini
B.
,
Ireland
J.
,
Rayman
P.
,
Richmond
A.
,
Golshayan
A.
,
Wood
L.
,
Elson
P.
,
Garcia
J.
,
Dreicer
R.
,
Bukowski
R.
.
2008
.
Sunitinib reverses type-1 immune suppression and decreases T-regulatory cells in renal cell carcinoma patients.
Clin. Cancer Res.
14
:
6674
6682
.

The authors have no financial conflicts of interest.

Supplementary data