Metastases at distant sites are the main cause of death in patients with breast cancer . The metastatic process involves a series of events in which changes in cell motility represent the hallmark of invasion and the initial step in metastasis . Over the past years, it has been clearly established that GnRH and its receptor are expressed in many extra-pituitary tissues and malignant tumors from the reproductive system, including the breast [26, 27]. It has also been shown that binding of GnRH to breast malignant tumor cells results in growth modulation  and inhibition of metastatic capacity . Although activation of some signaling pathways and effectors proteins involved in GnRHR-regulated cell motility have been reported [17, 24], the molecular mechanism(s) whereby the GnRH/GnRHR system suppresses cell migration is still unclear.
In the present study we assessed the effect of GnRH on the invasiveness capacity of human breast cancer MDA-MB-231 cells, an aggressive, highly invasive, and estrogen unresponsive cell line . To this end, we overexpressed the GnRHR in MDA cells and analyzed the effects of its cognate ligand on the pathways leading to actin cytoskeleton activation and cell adhesion. MDA cells transfected with the GnRHR (WT and DesK191) cDNAs, specifically bound 125I]-Buserelin and produced higher levels of the second messenger IP in response to the GnRH analog than untransfected cells, overcoming the problem related to the low naturally expressed GnRHR levels in breast cancer cell lines [50, 51]. In fact, the increased expression levels and IP response to GnRHa detected in GnRHR-transfected MDA cells, emulated those previously detected in breast cancer cells exhibiting high GnRHR expression levels . Here we confirm that to detect relevant effects of GnRH on breast cancer cells function, it is necessary to substantially increase cell surface plasma membrane receptor levels, which is an important issue given that the number of GnRHRs is highly variable in malignant breast tumors tissue . In this scenario, measurement of GnRHR density in malignant breast tissue may be useful as a surrogate marker to predict the tumor responsiveness to GnRHa administration.
We have shown that in MDA cells, GnRHRs were able to promote IP production upon activation by agonist. Although we did not detect measurable changes in cAMP levels after exposure to GnRHa in this particular cell line (not shown), previous studies have found that the inhibitory effects of GnRHa on other reproductive cancers (including prostate and endometrial cancer) is mediated by the Gαi protein [53–55]. Our data are consistent with previous studies in MCF-7 breast cancer cells, in which the GnRH/GnRHR system was capable to selectively promote IP production . These data support the idea that in extrapituitary tissues, the GnRHR may couple to different G proteins and activate distinct signaling pathways depending on the cell context, the particular GnRH analog employed to activate the receptor, and also probably the receptor density [56, 57].
Actin polymerization is involved in cell migration and thus is important in determining the invasiveness ability of cancer cells . Our results showed that Buserelin promoted actin polymerization as stress fibers in WT and Desk191 GnRHR-transfected adherent MDA cells, thus suggesting that activation of the GnRH/GnRHR system may be involved in the migratory potential of these malignant cells. Since Buserelin-stimulated MDA cells displayed many stress fibers and high F-actin levels, we analyzed the effects of GnRHa on RhoA, a small GTPase involved in actin polymerization and formation of stress fibers. In fact, previous studies have shown the effect of GnRH on actin cytoskeleton via other members of the Rho GTPases family [22, 23]. We found that in Collagen I-adherent MDA cells, exposure to GnRHa increased RhoA-GTP levels and paralleled the amount of stress fibers. The effects of GnRH in RhoA-GTP were verified by co-transfection assays employing the GAP domain of p190RhoGAP and its dominant negative mutant, GAP-Y1284D . p190RhoGAP is a specific GAP for RhoA and its effect represents more than 60% of the overall GAP activity in the cell [57, 58]. The results showed that GTPase RhoA levels were abolished or unaltered in the presence of the GAP domain or GAP-Y1284D, respectively, thus indicating the specificity of the GnRH/GnRHR system on this particular small G protein. Concurrently, these data indicates that the effects of GnRH on actin polymerization and stress fibers assembly are mediated through activation of RhoA in Collagen I-attached MDA cells.
To demonstrate that GnRH-activated GTP RhoA promotes cell adhesion and thus may represent one of the mechanisms whereby this G protein inhibits cell migration, cell adhesion assays as well as confocal visualization of FA (substrate binding sites) were performed. In fact, previous studies have shown that RhoA activity supports efficient substrate adhesion, reduces cell detachment rate, and attenuates cell locomotion [59–61]. Our results showed that exposure of Collagen-I-adherent MDA cells to the GnRHa promoted cell adhesion to substrate and increased the number of FA. Further, cell invisiveness of these GnRHa-exposed cells was abolished as disclosed by invasion assays in Collagen-I-covered Transwell Chambers.
Actin polymerization leads to membrane protrusion and extracellular cell-matrix adhesion, which are generally considered as markers of the migration capacity of different cell types . In this vein, it was interesting to find that in MDA cells co-transfected with the GnRHR and the GAP domain, stimulation with Buserelin did not promote detectable increments in cell adhesion to substrate - but paradoxically, it increased the levels of F-actin at the periphery of the cells. The observation that GAP domain-cotransfected cells additionally showed membrane protrusions similar to lamellipodia, suggests that continuous activation of other GTPases, such as Rac1, was present in these cells. In fact, previous studies have demonstrated the ability of GnRH to stimulate this particular GTPase .
Our findings in MDA cells exposed to GnRH has also been observed in other cell lines, and apparently the effects of this decapeptide on cell migration depend on the cell context. For example, it has been shown that GnRH-mediated attenuation in migration capacity of DU145 cells (prostate cancer-derived) is associated with an increase in the amount of stress fibers and with RhoA activation as well. By contrast, in TSU-Pr1 cells (also derived from prostate cancer cells) GnRH favors cell migration through mechanisms mediated by the GTPasas Rac1 and Cdc42, and by formation of filipodia and lamellipodia . Our results suggests that in MDA cells transfected with the GAP domain, the absence of active RhoA GTPase promoted loss of the FA and hence in their ability to adhere to substrate as it was observed in response to GnRHa. In this scenario, the loss of migratory capacity of these cells might have resulted from the relative decrease in RhoA GTPase levels, since it is well known that cell invasion requires the concourse of several small G proteins .