Increased expression of transcription factor TFAP2α correlates with chemosensitivity in advanced bladder cancer

Abstract

Background: The standard treatment for patients with advanced transitional cell carcinoma of the bladder is platin based chemotherapy. Only approximately 50% of the patients respond to chemotherapy. Therefore, molecular predictive markers for identification of chemotherapy sensitive subgroups of patients are highly needed. We selected the transcription factor TFAP2α from a previously identified gene expression signature for chemotherapy response.

Methods: TFAP2α expression and localization was assessed by immunohistochemistry using a tissue microarray (TMA) containing 282 bladder cancer tumors from patients with locally advanced (pT2-T4(b) and N(1-3)) or metastatic (M(1)) disease. All patients had received cisplatin containing chemotherapy. Furthermore, QPCR analysis of three TFAP2α isoforms was performed on tumor specimens of advanced muscle invasive bladder cancers (T2-4). Using the bladder cell lines T24 and SW780 the relation of TFAP2α and cisplatin and gemcitabine sensitivity as well as cell proliferation was examined using siRNA directed TFAP2α knockdown.

Results: TFAP2α protein expression was analyzed on a TMA with cores from 282 advanced bladder cancer tumors from patients treated with cisplatin based combinational chemotherapy. TFAP2α was identified as a strong independent predictive marker for a good response and survival after cisplatin-containing chemotherapy in patients with advanced bladder cancer. Strong TFAP2α nuclear and cytoplasmic staining predicted good response to chemotherapy in patients with lymph node metastasis, whereas weak TFAP2α nuclear staining predicted good response in patients without lymph node metastasis. In vitro studies showed that siRNA mediated knockdown of TFAP2α increased the proliferation of SW780 cells and rendered the cells less sensitive to cisplatin and gemcitabine. In contrast to that T24 bladder cells with mutated p53 showed to be more drug sensitive upon TFAP2α depletion.

Conclusions: High levels of nuclear and cytoplasmic TFAP2α protein were a predictor of increased overall survival and progression free survival in patients with advanced bladder cancer treated with cisplatin based chemotherapy. TFAP2α knockdown increased the proliferation of the SW780 bladder cells and reduced cisplatin and gemcitabine induced cell death. The inverse effect was observed in the TP53 mutated T24 cell line where TFAP2α silencing augmented cisplatin and gemcitabine sensitivity and did not stimulate proliferation.

Figure 1

Expression of TFAP2α isoforms. (A) Expression of TFAP2α isoform 1, 2 and 3 in advanced muscle invasive bladder cancer (T2-4) were determined using real-time Q-PCR. Analysis was performed on cDNA from 10 tumor specimens and each bar represents the mean from the 10 samples.(B) COS-7 cells were transiently transfected with empty pcDNA3.1/V5-His vector (lane 2, 6), pcDNA3.1/V5-His-TFAP2α isoform 1(lane 3, 7), isoform 2 (lane 4, 8) and isoform 3 (lane 5, 9). Western blot of 30 μg total protein lysate from non-transfected HU609 bladder cells (lane 1) and COS-7 transfected cells (lane 2-9) 48 h post transfection probed with anti TFAP2α antibody (lane 1-5) or anti-V5 antibody (lane 6-9).

Figure 2

Immunohistochemistry of bladder cancer tissue with antibody against TFAP2α. TFAP2α immunoreactivity of muscle invasive bladder cancer tumors. A and D: Negative stain of nucleus and cytoplasm at 5 and 40 fold objective magnification, respectively. B and E: Negative stain of nucleus and positive stain of cytoplasm (5× and 40×, respectively). C and F: Positive stain of nucleus and cytoplasm (5× and 40×, respectively).

Figure 3

The relationship between TFAP2α staining and survival after chemotherapy. TFAP2α immunoreactivity and overall survival (OS) rates and progression free survival (PFS). A and B: separation of OS (overall survival) curves based on TFAP2α nuclear and cytoplasmic staining in the lymph node invasive group, respectively. C and D: separation of PFS curves based on TFAP2α nuclear and cytoplasmic staining in the lymph node invasive group, respectively. C and F: separation of OS curves based on TFAP2α nuclear and cytoplasmic staining in the non lymph node invasive group, respectively. Red curves are high TFAP2α staining and blue curves are low TFAP2α staining.

Figure 4

Single agent dose-dependent cytotoxicity induced by cisplatin or gemcitabine. Human bladder cancer cells T24 (A,C) and SW780 (B,D) were seeded in 96 well plates and treatment with the indicated cisplatin or gemcitabine concentration started 24 h after seeding and continued for 48 h. Cell viability was assessed with MTT-assay. Each drug concentration was tested in six individual wells. Real time growth curves monitoring was performed with the RT-CES (RTCA) system. The T24 or SW780 cells were seeded into 16-well or 96-well E-Plates, which contain electrodes integrated into the bottom surfaces of each well that measure cell index based on impedance. Cell index correlates with the area of cells attached to the bottom of the plate. Cisplatin (E) or gemcitabine (F) at the indicated concentration was added 24 h after seeding.

Figure 5

Cisplatin sensitivity of TFAP2α silenced T24 and SW780. A and B: Transfection of 10-50 nM TFAP2α siRNA or control siRNA in T24 and SW780 cells, respectively. Real-time RT-PCR was used to determine the relative TFAP2α mRNA levels 48 h post transfektion. C and D: Transfection of 25 nM TFAP2α siRNA in T24 and SW780 cells, respectively. After 24 h incubation cisplatin or media was added to the cells. The viability of the cells was determined 96 h after transfection (48 h after the drug was added) by MTT-assay and expressed as the viability compared with the culture media control for both the TFAP2α siRNA or control siRNA transfected cells. E and F: As C and D using gemcitabine instead of cisplatin. (n = 6).

Figure 6

Cell proliferation of TFAP2α silenced T24 and SW780. Real time growth curves monitoring was performed with the RT-CES system. The T24 or SW780 cells were seeded into 16-well or 96-well E-Plates, which contain electrodes integrated into the bottom surfaces of each well that measure cell index based on impedance. Cell index correlates with the area of cells attached to the bottom of the plate. A and B: Transfection of 25 nM TFAP2α siRNA in T24 and SW780 cells, respectively. (n = 3) After 24 h incubation cisplatin or media was added to the cells. C and D: Transfection of 25 nM TFAP2α siRNA in T24 and SW780 cells, respectively. After 48 h incubation CyQuant assay was performed. (n = 8)