A study of transcription factors STAT3, SP1 and NFkB in breast cancer

Abstract

Background and Aims: Breast cancer is the second most common cause of cancer deaths in women. It is a tumour which has been extensively studied at a molecular level and, compared to other solid tissue tumours, our understanding of its biology is extensive. There are however some patients who are considered to have good prognostic feature of their tumours who go on to die from their disease. Transcription factors are the end point of many cell signalling pathways. They form the link between exogenous hormones and growth factors and DNA transcription. For the purpose of this study 3 different transcription factors have been selected for investigation. STAT3 is activated by various growth factors and cytokines including EGF. It is classified as an oncoprotein as its activation can mediate tumorgenisis in nude mice. STAT3 has been shown to confer resistance to apoptosis in breast cancer cells and it is associated with poor outcome in high risk breast cancers. SP1 is a transcription factor which is essential in the expression and the action of estrogen receptors (ER). It is known to be over expressed in other solid tissue tumours but there has been little work into its role in breast cancer. NFkB is activated in many cell survival settings. It is involved in the transcription of anti-apoptotic genes and also plays a role in cell proliferation, angiogenisis and cell adhesion. It is associated in breast cancers with an over expression of the oncogene Bcl-2. It has not been show to be a marker of prognosis but does appear to identify breast cancers with a poor response to chemotherapy. The aim of this study is to investigate the role of these transcription factors in the behaviour of breast cancers and the outcome of the disease. It will also investigate the affect of EGF and estogen stimulation on STAT3 activation in breast cancer cell lines. Methods: This study consists of 2 elements. Firstly an assessment of transcription factor expression in breast cancer samples and secondly a cell model experiment to investigate the stimulation of STAT3 activation. A cohort of 213 patients who presented to the Queen Elizabeth Hospital with invasive breast cancer in 1999 was selected. Tumour samples from these patients were retrieved and using immunohistochemistry were tested for the expression of STAT3, SP1 and NFkB. These results were then correlated with pathological features of the tumours, tumour receptor status (ER, PR HER2 and EGFR) and outcome of the disease. Two cell lines, MCF7 and SKBr3, were cultured in depleted medium. These cells were then stimulated with estrogen and EGF alone and in combination. Flow-cytometry was then used to quantify the levels of phosphorylated STAT3 in the 2 cell lines over a 3 day time course. The level of phosphorylation was then compared to the control lines to asses the effect of stimulation. Results: 209 breast cancers were successfully analysed for the expression of STAT3, 27% of these cancers expressed nuclear STAT3. The results demonstrated a significant correlation of STAT3 expression with cancers of a high grade (p=<0.001), increasing tumour size (p=0.004), vessel space invasion (p=0.034) and lymph node metastases (p=0.015). STAT3 expression was shown to be significantly correlated to high Nottingham prognostic index (NPI) scores. With regards to receptor status it was show that STAT3 expression was significantly associated with ER negative and PR negative cancers (p=0.003), whereas there was no relationship with HER2 status. The results did show that there was a significant relationship between STAT3 expression and EGFR positive cancers (p=0.007). When disease outcome was investigated it was shown that there was a trend towards improved survival in the STAT3 negative group and a significant relationship between STAT3 expression and disease recurrence at 5 years (p=0.04). SP1 expression was determined in 208 of the cancer samples with 33% of the tumours having strong nuclear staining. There was no significant relationship between SP1 expression and any of the pathological features mentioned. SP1 expression was related to ER positive tumours (p=0.015). Though there was no relationship with 5 year survival it appears that SP1 expression does reduce the risk of late (>2yr) disease recurrence (p=0.005). NFkB was over expressed in 15% of the 208 cancers samples. Again a significant correlation was shown with high grade tumours (p=0.001) and large tumours (p=0.014). NFkB expression was also shown to be more prevalent in ER negative cancers (p=0.006) and EGFR positive tumours (p=0.007). There was no significant relationship between NFkB expression and disease outcome. The cell model results showed that in the EGFR positive ER negative cell line (SKBr3), EGF stimulation resulted in a biphasic response of STAT3 phosphorylation, whereas estrogen had no effect on phosphorylation. In the ER positive MCF7 cells, which express low levels of EGFR, again EGF stimulation resulted in a biphasic response curve. Estrogen stimulation does cause an increase in activation but when estrogen is added to EGF stimulation there is an inhibition of STAT3 phosphorylation. Conclusions: This study has demonstrated that STAT3 and SP1 expression is important in disease outcome in breast cancer patients. Though there are differences in levels of expression, NFkB does not appear to have a role in breast cancer outcome. The cell model has show that EGF stimulation of EGFR positive cell lines results in increased STAT3 activation and also that this effect is inhibited by the addition of estrogen stimulation. These results raise important questions which are discussed in the study and suggest areas for further investigation.