Foxp3 expression in human cancer cells

Twenty two tumor cell lines of various tissue origins, kindly donated by collaborators (Table 1), as well as 3 cancer cell lines recently established, as previously described [9], from tumor cell suspensions collected from lung adenocarcinoma patients submitted to surgery (PGEGE, PKAKI and PINTZ), were studied. All cell lines were kept frozen and upon thawing they were maintained in complete medium (CM) that consisted of Iscove's medium (Gibco Laboratories, Grand Island, NY, USA), 10% FCS (Gibco), 100 μM Minimal Essential Aminoacids (Gibco), 100 U/mL penicillin, 100 μg/mL streptomycin (Gibco), 2 mM glutamine (Gibco) and 50 μM 2-mercaptethanol (Gibco), supplemented with hydrocortisone, 17β-estradiol, sodium selenite, insulin and transferrin (HITES solution) as described by Carney et al [10], at 37°C in 8% CO₂.

To explore whether the expression of Foxp3 by cancer cells is affected by culture with the HITES solution, experiments were undertaken using parallel cultures of tumor cell lines, in the presence and in the absence of HITES solution, i.e. at corticosteroid concentrations used to inhibit development of lymphocytes, for a period of 3 weeks, allowing at least 10–12 cell divisions. All experiments were repeated three times.

Foxp3 mRNA expression was examined in all cases using both conventional PCR (RT-PCR) and quantitative real-time PCR (qRT-PCR), after total RNA isolation from the tumor cell lines and reverse transcription to cDNA, as previously described [11], qRT-PCR was performed using the automated thermocycler RotorGene 6000 (Corbett Life Science, Sydney, Australia), the SYBR Supermix kit (Invitrogen, Paisley UK) and the RT² PCR Primer Set for Foxp3 (SuperArray, USA). β₂-Microglobulin (β₂-M) was used as a reference gene (RT² PCR Primer Set, SuperArray) [12, 13]. The qRT-PCR thermocycling conditions for Foxp3 were: 10 min at 95°C initial hold, followed by 40 cycles of denaturation at 95°C, annealing at 60°C and extension at 72°C all for 15 sec. The qRT-PCR thermocycling conditions for β₂-M were: 10 min at 95°C initial hold, followed by 40 cycles of denaturation at 95°C for 15 sec, annealing/extension at 60°C for 60 sec. Relative expression was analyzed using the Rotor Gene software (Ver. 6) and is presented as a multiple of the gene expression in one fibroblast line isolated during the development of the cancer cell lines. EBV-transformed B cells were used as negative controls, whereas a CD4+ Treg clone (kindly provided by Sophie Lucas, Brussels, Belgium) and PHA blasts were used as positive controls. For RT-PCR amplifications, 20 μM of the following primer sets were used for β-actin, forward 5' GGCATCGTGATGGACTCCG 3' and reverse 5' GCTGGAAGGTGGACAGCGA 3' and the RT² PCR Primer Set for Foxp3 (SuperArray), in a total reaction volume of 25 μL. Thermocycling conditions (PTC-200, MJ Research, Watertown-Mass., USA) included for β-actin 21 cycles of denaturation at 94°C, annealing at 68°C and extension at 72°C, all for 1 min, and for Foxp3, 31 cycles of denaturation at 95°C, annealing at 60°C and extension at 72°C, all for 15 sec.

IL-10 and TGFb1 mRNA expression was examined using both conventional PCR (RT-PCR) for IL-10 and quantitative real-time PCR (qRT-PCR) for TGFb1 using cDNA prepared as above. For TGFb1, the RT² PCR Primer Set from SuperArray was used, and the qRT-PCR thermocycling conditions were: 10 min at 95°C initial hold, followed by 40 cycles of denaturation at 95°C for 15 sec and annealing/extension at 60°C for 60 sec. β₂-M was used as a reference gene as for Foxp3 and the relative expression was analyzed and presented as above. For RT-PCR amplifications of IL-10, 20 μM of the RT² PCR Primer Set for IL-10 (SuperArray) was used in a total reaction volume of 25 μL. Thermocycling conditions included 30 cycles of denaturation at 95°C for 15 sec, annealing at 60°C for 15 sec and extension at 72°C for 30 sec. IL-10 expression is presented as a semiquantitative measurement obtained by calculating the intensity quotient for the gene and β-ac tin [14] and then normalizing to that of a fibroblast cell line.

To determine whether the expression of Foxp3 mRNA results in the production of a mature protein, tumor cell lines were examined by immunohistochemistry and by flow cytometry for intracellular expression of Foxp3, using different antibody clones. Immunostaining was performed according to a previously published protocol with slight modifications [15]. Sections were fixed in 4% paraformaldehyde, blocked in 5% normal mouse serum (eBioscience), and incubated with a mouse monoclonal biotinylated anti-Foxp3 antibody against the amino terminus of the human Foxp3 protein (dilution 1/250) (clone 236A/E7, eBioscience, San Diego, USA), followed by goat anti-mouse secondary antibody and dextran coupled with peroxidase molecules (EnVision Detection System, Peroxidase/DAB, Rabbit/Mouse, Dako). DAB (EnVision Detection System, Dako) was used as peroxidase substrate, while endogenous peroxidase activity was blocked directly after the fixation step with 1% of H₂O₂ in 1xTBS. Sections were counterstained with Mayer hematoxylin (Merck).

Flow cytometric detection of intracellular Foxp3 expression was performed using an anti-Foxp3 monoclonal antibody (clone PCH101, eBioscience) labeled with phycoerythrin and the corresponding IgG2a isotypic antibody according to the manufacturer. In each analysis, approximately 10⁶ tumor cells were used, and approximately 50–100 × 10³ events were acquired and analyzed on the basis of Foxp3 positivity using the CXP program (Beckman Coulter, USA). Positivity was determined on the basis of the Mean Fluorescence Index (MFI) for staining with Foxp3, against the staining observed with the corresponding isotypic antibody.

Expression values are presented as raw values or as mean ± SD. Spearman's bivariate correlation was used to identify correlations (Pearson's correlation and significance is presented) with the statistical software SPSS for Windows (version 11.5).