The Ewing sarcoma family of tumors (ESFT) represents an aggressive spectrum of malignant tumour types with common defining histological and cytogenetic features. To evaluate the functional activation of signal transducer and activator of transcription 3 (STAT3) in ESFT, we evaluated its activation in primary tissue sections and observed the functional consequences of its inhibition in ESFT cell lines. STAT3 was activated (tyrosine 705-phosphorylated) in 18 out of 31 primary tumours (58%), either diffusely (35%) or focally (23%). STAT3 was constitutively activated in 3 out of 3 ESFT cell lines tested, and its specific chemical inhibition resulted in complete loss of cell viability. STAT3 inhibition in ESFT cell lines was associated with several consistent changes in chemokine profile suggesting a role of STAT3 in ESFT in both cell survival and modification of the cellular immune environment. Together these data support the investigation of STAT3 inhibitors for the Ewing family of tumors.
The Ewing sarcoma family of tumors (ESFT) is an aggressive malignancy of childhood comprising a spectrum of tumour types with common defining histological and cytogenetic features [
Paraffin-embedded tumor samples were randomly selected from hospital histology archived material. Haematoxylin and eosin stained sections were examined to identify areas of viable tumour. Tissue cores were taken from these areas using a skin biopsy punch (diameter 4 mm) and reembedded to produce 3 tissue arrays, comprising a total of 31 tumour cores.
Immunohistochemical analysis was performed on 4
The ESFT cell lines A673 and TC32 were kind donations of Professor Sue Burchill, University of Leeds, UK. The ESFT cell line SKNMC, the prostate cancer cell line PC3, and HeLa were obtained from ATCC. Cell lines were maintained in the following media by Gibco/Invitrogen (Pasley, UK): DMEM (cell lines A673, SKNMC, HeLa), RPMI (cell line TC32), or F12 (cell line PC3) media containing 10% foetal calf serum. For fixing of 20 ng/mL IL-6-treated HeLa cells (E-Bioscience), supernatant was aspirated after 10 minutes activation, and cell pellets were fixed in 10% formaldehyde for 10 minutes before undergoing processing as for routine histopathology specimens. Following processing, each sample was embedded in paraffin.
For production of lysates the following experimental protocol was followed for ESFT and the PC3 cell lines. 500,000 cells were incubated in 2 mL media for 16 hours. Media was then removed and replaced by 1 mL of fresh media in the presence or absence of the pharmacological STAT3 inhibitor, S3i-201 (Calbiochem, Darmstadt, Germany), at a concentration of 100
25,000 cells were plated in 96-well flat-bottom plates suspended in 100
Supernatants for chemokine array analysis were produced by following the experimental protocol for production of lysates, as outlined above. Before harvesting and lysing cells, spent media was removed and stored at −80°C, after centrifugation to remove debris. The supernatants were analysed at a later stage using the Ray Biotech (Norcross, USA) Human Chemokine Array 1 according to the manufacturer’s instructions, using overnight incubations. The spots on the developed films were scanned with a BIO-RAD GS-800 Calibrated Densitometer (Hercules, USA) and quantified using ImageQuant software (GE Healthcare, Little Chalfont, UK). From each spot that represented a chemokine, the background (negative control spots) was subtracted. The chemokine spots were then normalised against positive control spots. The values that were obtained thus were density values expressed as percentage of positive control spots. Further analysis was then carried out using Microsoft Excel 2007 (Redmond, USA). Chemokines which had changed in control experiments (PC3 cell line) by more than 10% were excluded from further analysis, to exclude potential off-target effects. A change in chemokine level was arbitrarily defined as an increase or decrease by more than 10%, observed in arrays from two independent experiments.
Previous work has described a high incidence of STAT3 activation in ESFT [
PSTAT3 staining identifies two distinct distribution patterns of PSTAT3 tumour cells in ESFT specimens. (a) Hela cells were cultured in the presence or absence of IL-6 prior to formation of cell pellets for generating lysate or embedding in paraffin following formalin fixation. A phospho-STAT3-specific antibody was used to probe lysates by Western blot or to stain cells by immunohistochemistry, magnification ×200. (b) Representative staining patterns of primary ESFT samples. In the left panel, tumor cells are P-STAT3 negative whereas endothelial cells stained positive for PSTAT3 (acting as an internal positive control, black arrows), magnification ×200.
We assessed the role of P-STAT3 in contributing to tumour growth by performing
Phosphorylated STAT3 is present in Ewing’s cell lines, but not PC3, which can be diminished by S3i-201 and reduces viability in ESFT cell lines. (a) Western blot analysis; Lysates from IL-6-stimulated and -unstimulated HeLa cells served as positive controls for P-STAT3 and total STAT3, respectively. The GADPH band is representative of loading in each lane. Data are representative of at least three independent experiments per cell line. (b) Cell viability is determined by MTT assay. Data are representative of at least three independent experiments per cell line.
We next studied the effects of STAT3 inhibition on chemokine secretion patterns of ESFT cell lines and of the STAT3-negative PC3 cell line. Chemokine arrays were performed on supernatants obtained from cells cultured in the presence or absence of pharmacological P-STAT3 inhibition. For each cell line, two independent experiments were performed. Prior to analysis by chemokine arrays, we confirmed by Western blotting that P-STAT3 had been blocked in the cells from which the supernatants had been harvested (data not shown). Known STAT3-regulated chemokines (e.g., IL-8 [
STAT3 inhibition in ESFT cells alters the levels of a limited number of chemokines. Upper panel the gray up-pointing triangle denotes an increase, and the black down-pointing triangle denotes a decrease, of more than 10%, observed in two independent experiments for each cell line. The changes represent a ratio of density values, comparing arrays of STAT3-inihibited cells with arrays of DMSO-treated cells. Lower panel examples of arrays performed on supernatants of STAT3 null control cells (PC3) and the ESFT cell line SKNMC, cultured in the absence (DMSO treatment) or presence (S3i-201 treatment) of STAT3 inhibition. C
We sought to confirm that STAT3 is present in a subset of ESFT tumours and to investigate the role of STAT3 in ESFT. We were particularly interested in the role of STAT3 in contributing to tumour growth and immune regulation, which are classical roles described for STAT3 in adult tumours.
In our series, we demonstrated P-STAT3 positive tumour cells in 18/31 ESFT cores. In comparison, Lai et al. previously found P-STAT3 positive tumour cells in 25 out of 49 tumour cores [
We studied whether STAT3 contributes to tumour growth by performing viability assays of cells in growth conditions in a panel of three ESFT cell lines, which harbour constitutively activated STAT3. As a STAT3 negative control cell line, we used in our experiments the prostate cancer line PC3, which is homozygous STAT3 null [
Next we studied the effects of STAT3 inhibition on chemokine secretion patterns of ESFT cell lines through chemokine arrays performed on supernatants of ESFT cells. We confirmed STAT3 inhibition in the cells that produced the supernatants through Western blotting, and we used a STAT3 null cell line PC3 to exclude nonspecific effects. We found changes in chemokine levels of known STAT3 targets such as IL-8 whose secretion was increased by STAT3 inhibition in all three ESFT cell lines, but not in PC3. In addition, STAT3 inhibition induced a distinct the pattern of changes in chemokine secretion in each ESFT cell line. The significance of these findings is twofold. First, they suggest that STAT3 plays a role in regulating the chemokine environment of ESFT cell lines. Second, our observations highlight that in each ESFT cell lines STAT3 may be integrated into different immunological pathways.
In this study we confirm the finding that a subset of ESFT contains P-STAT3 positive tumour cells. Furthermore, we describe two novel distribution patterns of PSTAT3 positive tumour cells. Our
The authors were supported by a grant from the Bone Cancer Research Trust.