Breast cancer: Two-face ER

By Luisa Robbez-Masson
Barts Cancer Institute, Queen Mary University, London, UK

This summary was highly commended by the judges for Access to Understanding 2013


Oestrogen is a female hormone, produced in the ovaries, that stimulates the formation of the female sexual characteristics at puberty. It also triggers the growth of the breast tissues during the reproductive cycle and during pregnancy. However, oestrogen exposure in a woman’s lifetime has been linked to breast cancer risk for many years.

During the development of breast cancer, oestrogen feeds the breast tissues and the tumour indiscriminately and consequently helps its progression. Treatments that block oestrogen receptors have been successfully used in the clinic, however the response is difficult to predict and some patients eventually develop resistance to those drugs. Understanding how oestrogen and its receptor work is key to developing new effective therapies against breast cancer.

Scientists at Cancer Research UK in Cambridge have discovered that the oestrogen receptor (ER) activates a different set of genes in breast cancer patients who died compared to patients who survived. This important study could help scientists understand why a disease that appears superficially identical, has such differing outcomes in terms of response to treatment and, ultimately, survival.

ER belongs to the family of transcription factors, which means that it can travel to the cell nucleus, when partnered with oestrogen, and reach the DNA molecules that harbour all our genes. Like all transcription factors, ER has the capacity to attach to some specific “target” genes, and lead to their activation or inhibition, which will in turn tell the cells how to behave. ER is aided in reaching its nuclear targets by the pathfinder transcription factor FOXA1, which opens up the DNA molecule for easy access. It is known that this collaboration between ER and FOXA1 involves many genes and is very dynamic, but the importance of its role in cancer development is still unclear.

Using molecular biology tools, scientists can “capture” a transcription factor together with the DNA to which it is bound. It is then possible to read the DNA sequence obtained from these binding events in order to identify target genes of specific transcription factors. This technique, previously used in cancer cell line models, was applied for the first time by the team in Cambridge to biopsies from breast cancer patients that all possessed ER and FOXA1. These patients were carefully chosen according to how aggressive their cancer was: eight patients with good prognosis, seven patients with bad prognosis and three who had a relapse and developed metastases were selected.

The team identified the target genes of ER and compared them between the different groups. The results showed that 484 binding sites (i.e. spots on the DNA where ER is bound) were common to most of the patients, but that the strength of the binding was more important in poor outcome and relapse patients compared to good outcome ones. The experiment also indicated that the different outcome patients can be well characterised by a specific set of ER bound genes: 1,192 binding sites were predominantly found in the poor prognosis group and another, distinct 599 binding sites were representative of the good outcomes. The authors also observed that the pathfinder FOXA1 was more often binding near the poor outcome ER binding sites than the good ones.

The experiments revealed that ER was responsible for a double whammy; activating the poor outcome genes but also inhibiting the genes in the good outcome group, meaning that the binding of ER at those specific binding sites was biologically relevant in term of regulating cell behaviour.

In order to validate these finding they turned to well established cancer cell line models that were either drug sensitive or drug resistant, reproducing what is seen in patients. However, contrary to what they had hypothesised, the drug sensitive cell model was more similar to the poor outcome patients group, in term of ER binding sites, possibly suggesting that these models constitute an intermediate category of tumours, with an ER binding profile resembling more advanced tumours, on their way to acquired drug resistance. The events that led to this intermediate category might have been caused by the presence of other stimuli, such as other hormones or growth factors for instance, in the patients from whom the cell line models were established. By treating drug sensitive cells with some of these chemical stimuli, they were able to shift the ER binding profile, and observed an increase in new FOXA1 binding sites.

We have known for many years that ER can play a critical role in orchestrating the behaviour of breast cancer cells, but the emergence of FOXA1 as the principal conductor sets the stage for novel approaches in targeted therapy.

This article describes the research published in:

Differential oestrogen receptor binding is associated with clinical outcome in breast cancer (2012) C. S. Ross-Innes, R. Stark, A. E. Teschendorff, K. A. Holmes, H. Raza Ali, M. J. Dunning, G. D. Brown, O. Gojis, I. O. Ellis, A. R. Green, S. Ali, S.-F. Chin, C. Palmieri, C. Caldas, J. S. Carroll Nature 481(7381), 389-393

This article was selected for inclusion in the competition by Cancer Research UK.