> Research & Grants
> Grants Program
> Research Grants
> Research Grants Awarded
SWI/SNF: Mediator of the ER+/PR-, Tamoxifen-Resistant Phenotype?
The goal of this project is to identify new molecular determinants of response to hormonal therapies. Currently, the best predictor of response to tamoxifen is the presence of estrogen receptor alpha (ER). ER- patients do not benefit from tamoxifen therapy. However, only 60-65 % of patients with ER+ tumors will respond to tamoxifen. If in addition to ER, the estrogen-induced progesterone receptor (PR) gene is also assessed as a measure of a functional ER, then the expression of the two together becomes an even better predictor of response to tamoxifen where 75% of ER+/PR+ tumors will respond. However, if a tumor is ER+/PR-, the response rate drops to ~35-40%. This intermediate response rate indicates that only in some cases is the tumor behaving as one that is truly ER+, while in others, the tumor is behaving as ER- even though ER is present. For these latter tumors, a likely common mechanism involves a defect in the functional activity of ER.
We hypothesize that the defect in functional ER activity in ER+/PR- tumors that fail to respond to tamoxifen results from the loss of components of the SWI/SNF signaling complex. This chromatin remodeling complex is a multimeric global transcription regulator which functions by shifting the position of histones and its function is clearly essential for a variety of key cellular proteins linked to breast cancer including BRCA1, Rb, p53, and ER. Thus, loss of SWI/SNF function could potentially impact many growth regulating pathways important in breast cancer development. In breast cancer, the subunits BAF57 and BAF180 have been found to be lost in cell lines. Expression of the subunit BRM is inhibited by mitogenic signaling via Ras. Many investigators have shown that ER function requires the SWI/SNF complex, and recently, the BAF57 subunit been found to functionally link ER to the SWI/SNF complex. Thus, loss of SWI/SNF function via loss of BAF57, as shown in breast cancer cell lines, loss of other subunits, or alternative mechanisms such as decreased protein expression of subunits could abrogate ER function without loss of its expression.
The overall hypothesis of this proposal is that loss of functional SWI/SNF interaction with ER results in an ER+/PR- phenotype that will be resistant to anti-estrogen therapies. The 3 specific aims directly test this hypothesis. Specifically, we will: 1) use a dominant negative BRM (dnBRM) to knock out SWI/SNF complexes in ER+ breast cancer cell lines to confirm the role of SWI/SNF in regulating ER transcriptional activity. We will assess the effects of this knock out both on known SWI/SNF targets like p53 or Rb and on ER transcriptional targets like PR and pS2; 2) use either dnBRM or small inhibitory RNAs (siRNA) against BRM or BRG-1 to knock out SWI/SNF function followed by a) immunohistochemistry to assess ER expression and localization, b) assays of MAPK activation to assess non-genomic ER action, c) growth assays to assess effects on estrogen-induced growth and anti-estrogen inhibited growth; and 3) perform comparative microarray analyses on cells in which knock out of SWI/SNF function has resulted in abrogation of ER transcriptional activity and thus expression of estrogen-induced genes such as PR and estrogen-induced or anti-estrogen inhibited growth to establish a gene expression profile characteristic of this phenotype and then test this profile in a small cohort of ER+/PR- tumor samples.
Establishing the role of SWI/SNF in the ER pathway leading to tamoxifen response will allow us to address the relevance of tamoxifen therapy in a subset of patients who express ER but do not have a functional ER pathway.
The overall goal of this project is to identify new factors that can be used to determine whether a patient's tumor will respond to hormonal therapies like tamoxifen. Today, the key factor that is used to decide on the use of tamoxifen is the estrogen receptor status of the tumor. The estrogen receptor (ER) is the protein that binds estrogen and mediates its action as well as the protein that binds tamoxifen and mediates its action. Of tumors expressing ER, ~ 60-65% will respond to tamoxifen. If, however, in addition to ER we measure another protein that tells us whether the ER is functional or not, in this case the progesterone receptor (PR), now we get a better prediction of response where 75% of these ER+/PR+ tumors will respond to tamoxifen. If a tumor is ER+ but PR-, the response rate drops to ~ 35-40% suggesting that in some cases the tumor is behaving as if it were a true ER+ tumor, while in others, it is behaving as an ER- tumor. For this group of ER+/PR- tumors that behave as if they were ER- in response to tamoxifen, it seems likely that a common mechanism for this would involve a defect in ER function such that even though it is present, you don't see the expression of an ER regulated gene, PR, nor a response to tamoxifen.
The SWI/SNF complex is a major regulator of transcriptional activation by several key proteins, including ER. It is composed of several subunits and some of these subunits have been found to be lost in certain types of tumors, including breast cancer. Loss of specific subunits would abrogate the activity of the SWI/SNF complex and thereby inhibit the activity of its interacting partners like ER. In addition to loss, however, the expression of several subunits can be regulated by growth factor signaling, and to complicate matters further, the actual activity of the subunits can be regulated as well. Therefore, it is possible to abrogate the function of the SWI/SNF complex by several mechanisms and many of these may be occurring in breast cancer.
It is our hypothesis that the ER+/PR-, tamoxifen resistant form of breast cancer may occur through abrogation of SWI/SNF complex function and the experiments proposed here are designed to test this hypothesis. We will determine whether loss of SWI/SNF function results in an inactivation of ER such that the result is ER+/PR-. We will further determine if this results in resistance to tamoxifen. Finally, since SWI/SNF can be inactivated by multiple mechanisms, we will determine if we can establish a gene profile that will predict loss of SWI/SNF function and by extension, ER+/PR- and tamoxifen resistance.