The results presented here reveal a novel role for the small scaffold protein MP1 in ER-positive breast cancer cells. Although MP1 is expressed in both ER-positive and ER-negative breast cancer cells, its depletion using RNAi-mediated gene silencing leads to detachment and death of several ER-positive cell lines, including one (LCC9) with acquired estrogen independence and antiestrogen resistance. In contrast, MP1 gene silencing had no detectable effect in three ER-negative breast cancer cell lines or a non-tumorigenic mammary epithelial cell line. Although this is a limited sample, MP1 has also been depleted in rat fibroblasts and human prostate cancer cells, and cell detachment or death was not reported in either case [7, 8]. Therefore, MP1 expression seems to be required for survival in a subset of cell types, including ER-positive breast cancer cells. The mechanism of cell death that occurs as a result of inhibiting MP1 expression in MCF-7 cells was shown to be apoptosis, as demonstrated by decreased Bcl-2 expression, increased PARP cleavage, Annexin V staining, and rescue of the death phenotype by treatment with the pan-caspase inhibitor z-VAD-FMK.
Several interesting questions are raised by these results. One is what pro-survival pathways are affected by loss of MP1 expression in MCF-7 cells. Depletion of MP1 did not result in decreased ERK activation, indicating that its pro-survival functions are not mediated by the ERK pathway. The lack of an effect on ERK activity was somewhat surprising, since MP1 was originally identified as a scaffold protein that increases ERK signaling , but is consistent with results obtained in prostate cancer cells . In contrast, inhibition of MP1 expression resulted in a greater than two fold decrease in Akt phosphorylation. The extent of Akt inhibition may be an underestimate, since by 48 h a majority of cells were dead, and the remaining live cells might represent ones with the lowest extent of MP1 knockdown. Akt plays a known pro-survival function in breast cancer cells, where it relays signals from upstream molecules including integrins, growth factor receptors, PI3K and mTORC1 to downstream molecules such as Bcl-2 and NF-κB [21–26]. In MCF-7 cells, the activity of Akt1 can be modulated by estradiol and IGF [27–29]. The fact that this kinase likely also plays a role in MP1 mediated survival is supported by the fact that expression of a constitutively active Akt1 partially rescued the cell death phenotype observed upon MP1 knockdown. However, the partial effect observed indicates that there are likely to be additional survival pathways involved as well.
Depletion of MP1 in MCF-7 cells also resulted in decreased levels of ERα protein, mRNA (data not shown) and transcriptional activity on ERE-containing promoters. The apoptosis observed is unlikely to be due solely to a loss of ER signaling, since we and others have found that inhibition of ER expression using siRNA does not result in MCF-7 cell death . The fact that LCC9 cells, which are estrogen independent and antiestrogen resistant, die in response to MP1 gene silencing also supports a model in which loss of ER expression is not the sole cause of cell death. Several studies indicate that ER may be implicated in breast cancer cell survival via cross-talk with the PI3K/Akt pathway , or by regulating the activity of NF-κB [32, 33], Bcl-2 [34, 35], or IAP family members . We therefore cannot rule out the possibility that decreased ER expression may contribute in some way to the apoptosis observed in MCF-7 cells.
A second question raised by these results is the molecular basis for the differential requirement for MP1 for survival of ER-positive vs. ER-negative breast cancer cells. One possibility is that activation of pro-survival proteins such as Akt is not dependent on MP1 in ER-negative cells, and a second is that the ER-negative cells are less dependent on these pro-survival signaling pathways. The fact that LY294002 caused a concentration-dependent apoptotic response in MCF-7 cells, but did not affect MDA-MB-231 cells supports the latter hypothesis. This is in agreement with previous reports describing a differential sensitivity to this compound between the two breast cancer cell lines [19, 20].
A final question is whether the cell death that we have observed is related to the previously identified roles of MP1 in cell spreading and motility. Since the phenotype involves cell rounding and detachment, inhibition of MP1expression may disrupt cell adhesion signals, which could then trigger cell death. Preliminary PCR array experiments indicated that inhibiting MP1 expression leads to decreased expression of molecules involved in cell adhesion in MCF-7 cells, including several integrins (data not shown). Immunoblotting analysis indicated a small but reproducible decrease in beta 1 integrin protein levels upon MP1 silencing (Additional file 2: Figure S2). Since integrins can initiate pro-survival signaling , future experiments will investigate if a loss of integrin expression plays a role in the decreased Akt activation and/or apoptosis observed as a result of MP1 knockdown.