PGRMC1 Inhibits Progesterone-Evoked Proliferation and Ca2+ Entry Via STIM2 in MDA-MB-231 Cells

Progesterone receptor membrane component 1 (PGRMC1) has been shown to regulate some cancer hallmarks. Progesterone (P4) evokes intracellular calcium (Ca2+) changes in the triple-negative breast cancer cell lines (MDA-MB-231, MDA-MB-468, and BT-20) and in other breast cancer cell lines like the luminal MCF7 cells. PGRMC1 expression is elevated in MDA-MB-231 and MCF7 cells as compared to non-tumoral MCF10A cell line, and PGRMC1 silencing enhances P4-evoked Ca2+ mobilization. Here, we found a new P4-dependent Ca2+ mobilization pathway in MDA-MB-231 cells and other triple-negative breast cancer cells, as well as in MCF7 cells that involved Stromal interaction molecule 2 (STIM2), Calcium release-activated calcium channel protein 1 (Orai1), and Transient Receptor Potential Channel 1 (TRPC1). Stromal interaction molecule 1 (STIM1) was not involved in this novel Ca2+ pathway, as evidenced by using siRNA STIM1. PGRMC1 silencing reduced the negative effect of P4 on cell proliferation and cell death in MDA-MB-231 cells. In line with the latter observation, Nuclear Factor of Activated T-Cells 1 (NFAT1) nuclear accumulation due to P4 incubation for 48 h was enhanced in cells transfected with the small hairpin siRNA against PGRMC1 (shPGRMC1). These results provide evidence for a novel P4-evoked Ca2+ entry pathway that is downregulated by PGRMC1.


Introduction
The expression of nuclear progesterone receptor (nPR) is needed for cell proliferation and maturation of breast tissue, and further, nPR expression is found under the regulation of the estrogens [1]. Interestingly, activation of alveologenesis and, subsequently, milk production relies on the balance between progesterone receptor and prolactin/Stat5 signaling [1,2]. Reduction of circulating progesterone (P 4 ) concentration facilitates gene expression of the prolactin receptor as well as the activation of its downstream effector, the signal transducer and activator of transcription complex 5 (STAT5) [1][2][3].
Physicians have used the lack or presence of the classical nuclear estrogen receptor (nER) or nPR, in order to histologically characterize the different breast cancer subtypes [4]. Thus, ESR1, PGR, and ERBB2 mRNAs are not detected in most of the basal tumor samples and, therefore, also known as triple-negative breast cancer subtype [5].
Regarding the role of P 4 in cell proliferation, it has been reported that P 4 impairs normal breast epithelial cell proliferation in vivo, as well as in some types of breast cancer cells in vitro [6]. The effect of P 4 on cell proliferation has been described depending on nPR activation; thus, the triple-negative breast cancer (TNBC) cell lines should remain unaltered in the presence of P 4 ; however, recent studies have claimed that P 4 evokes alteration of the TNBC cells [7,8]. These negative effects in MDA-MB-231 cells are suggested to be mediated by membrane progesterone α (mPRα) [7], which might be modulated by progesterone receptor membrane component 1 (PGRMC1) [1]. Nonetheless, the evidence against mPRα as the P 4 receptor has also been reported by others [9]. More recent studies have been performed to identify alternative P 4 receptors, leading to the identification of two additional protein families that are able to transduce the P 4 effects in the neuroimmune system and other tissues [10,11]: (1) Class II progestin and adipoQ receptor family members (PAQR) (such as mPRα is also known as PAQR7)); (2) B5-like heme/steroid-binding protein family that groups proteins like the PGRMC1,PGRMC2, and others [10]. PGRMC1 expression is enhanced in patients suffering from different types of tumors, resulting in poor prognosis [12,13]. In line with this observation, PGRMC1 has been linked with lipid metabolism, which would lead to enhanced breast cancer progression [14,15]. It is worth mentioning that the role of PGRMC1 is mainly evidenced in the luminal type A breast cancer cell lines (MCF7 and T47-D); meanwhile, MDA-MB-231 cell proliferation remains unaltered, even when PGRMC1 expression is artificially altered [14].
Finally, changes in the cytosolic free Ca 2+ concentration ([Ca 2+ ] c ) have been reported to be relevant for cell proliferation [17]. In this sense, although P 4 evokes changes in the intracellular calcium homeostasis, scientists disagree in the possible cellular Ca 2+ sources. Thus, in retinal Muller glia cells, a non-permeable P 4 analog evokes changes in the intracellular Ca 2+ concentration, and further, authors reported a substantial calcium entry from the extracellular medium downstream of progesterone membrane receptors activation, but they did not show intracellular store calcium release [18]. Contrarily, human spermatozoa exhibit Ca 2+ release from intracellular stores in response to P 4 [19]. In neurons, P 4 administration enhances Ca 2+ release by favoring inositol trisphosphate receptor (IP 3 R) permeability via enhanced phosphorylation by the serine/threonine-protein kinases (AKT) [20]. Similarly, in oral squamous cancer cells (PE/CA-PJ15 cells), it has been described that P 4 (5 and 10 µM) evokes both Ca 2+ release and Ca 2+ entry, which is inhibited by adding the PGRMC1 antagonist AG205 [21]. Contrarily, in the breast cancer luminal A type, T47-D, ATP evokes a transient increase in the [Ca 2+ ] c that is downregulated by P 4 , medroxyprogesterone acetate (MPA), and also by the progesterone non-permeable analog, progesterone-CMO-BSA. The negative effect of P 4 on ATP-evoked calcium mobilization is non-detected in CHO cells, and experimental data has revealed the absence of mPRs in these cells [22].
Here, we aimed to extend the knowledge regarding the possible regulation of P 4 on calcium homeostasis in breast cancer cells. We provided evidence of a novel P 4 -activated pathway in TNBC (MDA-MB-231, BT-20, and MDA-MB-468 cells). P 4 evoked Ca 2+ entry in MDA-MB-231 cells and MCF-7 cells by the activation of a novel calcium pathway involving the Calcium release-activated calcium channel protein 1 (Orai1), Transient receptor potential channel 1 (TRPC1), and Stromal interaction molecule 2 (STIM2); meanwhile, STIM1 silencing had no effect in this mechanism. Interestingly, activation of PGRMC1 by P 4 attenuated this Ca 2+ signaling pathway, which led to a decreased Ca 2+ -dependent Nuclear Factor of Activated T-Cells 1 (NFAT1) nuclear accumulation and, therefore, negatively regulated MDA-MB-231 cells proliferation.

P 4 Inhibits MDA-MB-231 Cell Proliferation
MDA-MB-231 cells lack the classical nPR [6], while these cells express a less characterized and alternatively spliced nPR, which may also bind to P 4 [23]. Additionally, the effect of P 4 in cell proliferation has been shown to be different according to the cell types investigated [6,8,24]. As previously reported [6], and as depicted in Figure 1A, the treatment of MDA-MB-231 cells with 1 µM of P 4 resulted in a significant reduction of bromodeoxiuridin (BrdU) fluorescence at all analyzed times (24-72 h), as compared to cells treated with the vehicle. Therefore, these data revealed a negative role of P 4 in the proliferation of the MDA-MB-231 cells (n = 6).    Considering the relevant role of the Ca 2+ -homeostasis on cell proliferation, even in MDA-MB-231 cells, as previously demonstrated by our research group [17,25,26], and the controversy regarding the role of P 4 in calcium homeostasis aforementioned, we analyzed the effect of P 4 on the Ca 2+ homeostasis in MDA-MB-231 cells [18]. MDA-MB-231 cells suspended in a Ca 2+ -free medium exhibited a small Ca 2+ entry when it was added back to the extracellular medium (1 mM CaCl 2 ; Figure 1B, dotted line). Incubation of MDA-MB-231 cells with P 4 evoked Ca 2+ entry ( Figure 1B, grey line), which was significantly smaller than the one evoked by thapsigargin in these cells (2 µM; TG, black line, Figure 1B, p < 0.001). TG blocks the sarcoendoplamic reticulum Ca 2+ -ATPase (SERCA), which is a classic maneuver to activate store-operated Ca 2+ entry (SOCE) ( Figure 1B, p < 0.001, n = 6) [27,28]. Interestingly, we were unable to visualize a significant P 4 -induced Ca 2+ release from the internal stores, contrary to the results observed using TG, and therefore, it may explain the smaller activation of the Ca 2+ entry recorded under these experimental conditions if compared to the one evoked using TG.
In order to investigate the mechanism underlying the P 4 -dependent Ca 2+ entry activation, we analyzed possible changes in the intraluminal Ca 2+ concentration induced by P 4 using the endoplasmic reticulum Ca 2+ dye, GEM-CEPIA1er, in the triple-negative breast cancer cell lines-MDA-MB-231 and MDA-MB-468 cells. As depicted in Figure 1C,D, MDA-MB-231 cells suspended in a Ca 2+ -free HBS medium presented changes in the decay of the blue/green fluorescence ratio of GEM-CEPIA1er after cell stimulation with P 4 (1 µM) with respect to the non-stimulated resting cells (compare black and grey lines in Figure 1C). As shown in Figure 1F, similar P 4 -evoked Ca 2+ release occurred in MDA-MB-468 cells (compare black-dotted and black-solid lines in Figure 1F). Internal control experiments consisted of the addition of TG (1 µM) at the end of each experiment, which enhanced the release of Ca 2+ from the endoplasmic reticulum, as detected by the drastic change in the GEM-CEPIA1er fluorescence ratio. Following this, in an attempt to avoid the Ca 2+ release evoked by P 4 , we used a relative specific antagonist of the inositol trisphosphate receptor (IP 3 R), 2-Aminoethoxydiphenyl borate (2-APB), in MDA-MB-231 cells ( Figure 1E) and MDA-MB-468 cells ( Figure 1F) [29]. As shown in Figure 1E,F, treatment of the triple-negative breast cancer cells with 50 µM of 2-APB for 5 min was able to almost completely block the release of Ca 2+ evoked by P 4 . Therefore, these results allowed us to conclude that although not being reported using fura-2, P 4 evokes enough intraluminal Ca 2+ changes to promote the extracellular Ca 2+ entry, and, further, the Ca 2+ release evoked by P 4 occurs through the IP 3 R.

Activation of PGRMC1 Downregulates P 4 -Dependent Ca 2+ Mobilization in MDA-MB-231 Cells
As shown in Figure 2A, MDA-MB-231 cells presented greater PGRMC1 expression values as compared to the non-tumoral cell line, MCF10A (1.9 ± 0.2-fold increase as compared to MCF10A, p < 0.001, respectively; n = 6). Similarly, enhanced PGRMC1 expression was detected in other breast cancer cell types-the luminal A type MCF7 and triple-negative BT-20 ( Figure 2A, p < 0.01 and p < 0.001, respectively). In addition, considering the discrepancies mentioned above regarding the structural identity of the P 4 -receptor in the MDA-MB-231 cells, we silenced the expression of the PGRMC1 receptor by using the small hairpin RNA against PGRMC1 (shPGRMC1) and an endoribonuclease-prepared siRNAs against PGRMC1 (esiPGRMC1). MDA-MB-231 cell transfection with either shPGRMC1 or esiPGRMC1 efficiently reduced PGRMC1 expression in MDA-MB-231 cells after 48 h of transfection (See WB image in Figure 2B,D). Interestingly, PGRMC1 silencing did not alter Ca 2+ mobilization in resting cells ( Figure 2B, Ca 2+ traces), but an increase in the P 4 -induced Ca 2+ entry by a 35 ± 19% (using ShPGRMC1; Figure 2C) and 60 ± 38% (using esiPGRMC1; Figure 2D) was observed as compared to the non-genetically modified MDA-MB-231 cells (see cells transfected with scramble plasmids vs. shPGRMC1 and esiPGRMC1, respectively, and the histogram at the bottom of Figure 2C,D; p < 0.01 and p < 0.05, n = 6). On the contrary, we did not observe significant changes in the TG-evoked SOCE in MDA-MB-231 cells transfected with shPGRMC1 under these experimental conditions ( Figure 2C, right-hand side graph, 20 ± 15%; p > 0.05, n = 6). According to our results, stimulation of Ca 2+ entry by P 4 was downregulated by PGRMC1 in MDA-MB-231 cells, and the latter results were also reproduced in MCF7 cells (See Figure 3A,B,D). These results agreed with the downregulation of (Ca 2+ ) c concentration by P 4 , as previously described in spontaneously immortalized rat granulosa cells and breast cancer luminal A type, T47-D, but authors did not deeply explore this mechanism [22,30].
STIM1 is a very relevant component of the Ca 2+ entry in non-electrical excitable cells, including breast cancer cells [31]. In order to dig into the mechanism underlying the P 4 -evoked changes in Ca 2+ entry, we knocked-down STIM1 by using esiRNA [28,[32][33][34][35]. esiRNA consists of a commercial combination of siRNAs directed against different sequences of the targeted protein that has been generated by endonuclease activity, and thus, they would present greater efficiency and specificity than siRNAs, which are generated toward a unique sequence of the targeted protein [36].
As depicted in Figure 4A.1, transfection of MDA-MB-231 cells with esiSTIM1 completely silenced STIM1 expression, which did not significantly alter the Ca 2+ homeostasis of MDA-MB-231 cells maintained under normal conditions ( Figure 4A.3). In our hands, esiSTIM1 did not quantitatively modify the Ca 2+ entry induced by the administration of P 4 ( Figure 4B, black-dotted trace; p > 0.05, n = 6). Next, we explored the possible role of STIM2, a less characterized isoform of STIM1 [37,38], which is able to detect smaller changes in the endoplasmic reticulum's Ca 2+ content in response to low physiological agonist concentration, and therefore, it may detect the small P 4 -evoked intraluminal Ca 2+ changes, previously shown in Figure 1 [39][40][41]. The transfection of MDA-MB-231 cells with the esiSTIM2 significantly reduced the expression of the targeted protein ( Figure 4A.2), and as evidenced in Figure 4B, STIM2 silencing attenuated P 4 -evoked Ca 2+ entry by 67.8 ± 3.0% ( Figure 4B, grey trace; p < 0.001, n = 6) but did not significantly modify the Ca 2+ homeostasis in non-stimulated cells ( Figure 4A.4). Similarly, the downregulation of P 4 -evoked Ca 2+ entry was observed in MCF7 lacking STIM2 ( Figure 3C       Following this, we explored the plasma-membrane Ca 2+ channels activated by P 4 in MDA-MB-231 cells. Thus, the transfection of MDA-MB-231 cells with shRNAs against Orai1 and TRPC1 channels resulted in a significant reduction (89.5 ± 3.8% and 84.3 ± 1.2%) in P 4 -evoked Ca 2+ entry as compared to the cells transfected with scrambled plasmids ( Figure 4C; p < 0.001; n = 6). Therefore, our findings indicate that the P 4 -evoked Ca 2+ entry course through a novel pathway where STIM2, Orai1, and TRPC1 are involved. Interestingly, STIM2 silencing did not alter the activation of SOCE in MDA-MB-231 cells, which is mainly conducted by STIM1 and Orai1, as depicted in Figure 5A-C, and it has been previously reported by others [31]. Structural analysis of the Orai1 structure revealed that Orai1 is phosphorylated at the serine-34 residue by the protein kinase A (PKA), which modulates the channel function [42]. Furthermore, it has been described that activation of PGRMC1 leads to the activation of certain protein kinases like PKA, PI3K, and SFK [16]. Therefore, we explored whether the inhibition of P4-evoked Ca 2+ entry might occur through the activation of the PGRMC1/cAMP/PKA signaling pathways [43,44]. As depicted in Figure 6, MDA-MB-231 (6A.1), MDA-MB-468 (6A.2), and BT20 (6A.3) cells were preincubated with 3 μM of the specific PKA antagonist, KT5720 (PKAi), which resulted in a significant increase in the P4-evoked Ca 2+ entry, as summarized in the bar graph in Figure 6A. Additionally, to further demonstrate the activation of PKA in MDA-MB-231 cells in response to P4 Structural analysis of the Orai1 structure revealed that Orai1 is phosphorylated at the serine-34 residue by the protein kinase A (PKA), which modulates the channel function [42]. Furthermore, it has been described that activation of PGRMC1 leads to the activation of certain protein kinases like PKA, PI3K, and SFK [16]. Therefore, we explored whether the inhibition of P 4 -evoked Ca 2+ entry might occur through the activation of the PGRMC1/cAMP/PKA signaling pathways [43,44]. As depicted in Figure 6, MDA-MB-231 (6A.1), MDA-MB-468 (6A.2), and BT20 (6A.3) cells were preincubated with 3 µM of the specific PKA antagonist, KT5720 (PKAi), which resulted in a significant increase in the P 4 -evoked Ca 2+ entry, as summarized in the bar graph in Figure 6A. Additionally, to further demonstrate the activation of PKA in MDA-MB-231 cells in response to P 4 stimulation, we analyzed the changes in the phosphorylation at Ser239 of vasodilator stimulated phosphoprotein (VASP) that is a substrate of PKA and has commonly been used for monitoring the PKA activation. As shown in Figure 6B, P 4 evoked a 1.3 ± 0.06-fold increase in P-VASP as compared to mock resting cells (p < 0.05; n = 3); meanwhile, PGRMC1 silencing reversed P-VASP to similar values than those found in mock resting cells (p < 0.05; n = 3). These data reproduced the observation noted with the shPGRMC1 (n = 6), and therefore, led us to conclude that PGRMC1 impairs P 4 -induced Ca 2+ entry by activating PKA and, subsequently, by inducing Orai1 phosphorylation, as it has been previously demonstrated by others [42].  The image is representative of 3 independent transfections, and the bar graph represents the increase in the phosphorylation rate of VASP with respect to the control mock cells under resting conditions. *: represents p < 0.05; meanwhile, $ : represents p < 0.05 with respect to the cell stimulated with P 4 but non-transfected with the small hairpin siRNA against PGMRC1 (shPGRMC1).

PGRMC1 Silencing Reduces the Negative Effect of P 4 in MDA-MB-231 Cell Proliferation
A positive effect of P 4 in proliferation has been reported in cells expressing the classical nPR, such as MCF7 or glioblastoma cells [45][46][47]; meanwhile, the negative effect of P 4 on cell proliferation was observed in cells lacking the nPR like the human BxPC3 pancreatic adenocarcinoma cells [48]. In TNBC cells, we explored the role of PGRMC1 on the P 4-mediated inhibition of cell proliferation, previously described in Figure 1. As shown in Figure 7A, the silencing of PGRMC1 reversed the inhibitory effect of P 4 on MDA-MB-231 cell proliferation, thus suggesting that this negative effect is mediated by the P 4 binding and activation of PGRMC1. Furthermore, our results indicated that the silencing of STIM2 did not alter the proliferation pattern found in MDA-MB-231 cultured in the presence of P 4 for 24 h (See Figure 7B); then, STIM2 would not be involved in the negative role of P 4 on MDA-MB-231 cells proliferation.

PGRMC1 Is Involved in the P 4 -Evoked Cell Death in Breast Cancer Cells
In order to confirm that the negative effect of P 4 on breast cancer cells course by activation of PGRMC1, we analyzed the number of dead cells by using the PI uptake protocol in MDA-MB-231 cells that were previously transfected with the scramble or shPGRMC1 plasmids. As shown in Figure 8

PGRMC1 Is Involved in the P4-Evoked Cell Death in Breast Cancer Cells
In order to confirm that the negative effect of P4 on breast cancer cells course by activation of PGRMC1, we analyzed the number of dead cells by using the PI uptake protocol in MDA-MB-231 cells that were previously transfected with the scramble or shPGRMC1 plasmids. As shown in   After confirmation of the efficiency of the silencing protocol by Western blotting, cells were cultured for an additional 24 h in the absence or presence of P 4 (1 µM). Following this, cells were incubated on the day of the experiment with 4 µM of propidium iodide (PI) for 45 min at 37 • C inside the incubator. Cell images were acquired at 555/624 nm of wavelength (excitation/emission, respectively), by using a 40× objective mounted in a fluorescence microscope. The percentage of positive PI loaded cells with respect to the total counted cells in the fields was estimated, considering the mean ± Standard error of the mean (S.E.M.) of 3-4 independent experiments, and was represented in the bar chart. ***: represents p < 0.001 with respect to the mock resting cells; meanwhile, $$$ : represents p < 0.001 with respect to the cells expressing PGRMC1 and treated with P 4 . The bar in the images represents 50 µm.

PGRMC1 Activation by P 4 Impairs Nuclear Factor of Activated T-Cells 1 (NFAT1) Nuclear Translocation
Finally, we analyzed the effect of P 4 on the activation of NFAT1, an important calcium-dependent and pro-proliferative transcription factor that acts as checkpoints in the cell cycle [49][50][51][52][53].
As depicted in Figure 9, NFAT1-reporter fluorescence was significantly enhanced in cells where PGRMC1 was silenced and, subsequently, stimulated for 48 h with 1 µM of P 4 , perhaps, enhanced NFAT1 reporter visualization results of the biggest Ca 2+ entry previously shown in these cells lacking PGRMC1 under similar experimental conditions (Figure 2), and it has been also reported by others.

Discussion
The role of P 4 on cell proliferation has been widely debated due to the presence of different receptors that may be targeted by this hormone in the different tissues [8,9,35,49,54]. In our hands, stimulation of MDA-MB-231 cells with P 4 (1 µM) resulted in a PGRMC1-dependent reduction of cell proliferation, as demonstrated by the observation that PGRMC1 silencing impairs the inhibitory effect of P 4 in cell proliferation. Furthermore, P 4 -evoked Ca 2+ entry was strongly dependent on the activation of STIM2, Orai1, and TRPC1. Furthermore, P 4 -evoked Ca 2+ entry was bigger in cells lacking PGRMC1 and in those cells pre-incubated with the PKA antagonist KT5720, thus, suggesting that the PGRMC1 plays a negative role in the P 4 -induced Ca 2+ signaling. PGRMC1 silencing reversed the P 4 negative effect on the Ca 2+ entry reported in the MDA-MB-231 cells, which would lead to an enhanced NFAT1 translocation to the nucleus, as evidenced by using the NFAT1-reporter.
MDA-MB-231 cells express at least three different types of P 4 receptors: PAQR family, PGRMC family, and sigma-2R. The latter was initially considered to be the PGRMC1; in fact, current evidence supports this idea [55,56]. Nonetheless, sigma-2R has been recently claimed to be different from PGRMC1 based on genetic and biochemical evidence [57][58][59][60]. In MDA-MB-231 cells, we demonstrated that inhibition of sigma-2R with NO1 reported a different effect on TG-evoked SOCE to the one presented here by silencing PGRMC1 [26]. The different physiological roles of both proteins would confirm the previous observations done by other groups [55]; however, we could not discard that both proteins would be part of a macromolecular complex targeted by P 4 . On the other hand, regarding the possible relationship between the members of the PAQR and PGRMC1 families, controversy still remains. In this sense, it has been reported that mPRα expressed in seatrout oocyte membranes would be activated by P 4 , which would subsequently downregulate cAMP production by impairing the activation of the adenylate cyclase [54,61], but according to the results obtained with the PKA antagonist and further demonstrated with the analysis of P-VASP, we believed that the latter did not take place in our cell models. Additionally, these authors reported that PGRMC1 might be involved in this pathway since PGRMC1 supports the access of mPRα to the plasma membrane [11,60]. Contrarily, artificial human mPRα expressed in HEK293 cells and MDA-MB-231 cells has failed to evoke Ca 2+ mobilization in response to P 4 , and it is not able to interact with G proteins, and therefore, it is not able to activate cAMP or MAPK (ERK1/2 or p38). In fact, these authors concluded that mPRα was not expressed in the plasma membrane [30]. By silencing PGRMC1, we were able to completely abolish the negative effect of P 4 on the proliferation of the MDA-MB-231 cells, which denotes a relevant role of this protein in the physiology of the MDA-MB-231 cells, as previously indicated by others [8,10]. Interestingly, PGRMC1 knockdown in the human colon cancer cell line, HCT116, has reported the contrary effect since authors evidenced a decrease in proliferation and metastasis to the liver in cells lacking this protein [62].
Regarding the role of P 4 in the regulation of the intracellular Ca 2+ homeostasis, it has been reported that the hormone reduces the expression of Ca 2+ channels and Ca 2+ -dependent proteins involved in cell contraction in mouse embryonic stem cells (such as Ryr2, Calm2, Trpv2, and Mylk3) [63]. In line with this observation, spontaneously immortalized granulosa cells exhibit a dramatic reduction in the (Ca 2+ ) c in response to prolonged P 4 administration [30]. In the breast cancer cell line MCF7, P 4 elicits changes in the expression of S100A11, S100A10, calreticulin, VDAC1, SERCA3, and SERCA1, resulting in a barely Ca 2+ efflux from the endoplasmic reticulum as well as impaired membrane potential in the mitochondria [64]. Here, we reported for the very first time that P 4 -evoked Ca 2+ entry requires STIM2 activation in MDA-MB-231 and MCF7 cell lines (Figures 3 and 4), which operates Ca 2+ entry in response to small changes in the amount of Ca 2+ accumulated in the endoplasmic reticulum [28,36]. The latter was evidenced here by using GEM-CEPIA1er, a Ca 2+ dye that was specifically designed to monitor the intraluminal Ca 2+ changes in the transfected cells [65]. Interestingly, the combined use of this Ca 2+ dye and 2-APB allowed us to demonstrate that in the triple-negative breast cancer cells (MDA-MB-231 and MDA-MB-468), the release of Ca 2+ evoked by P 4 occurred through the IP 3 R (Figure 1). 2-APB was previously reported to be a relatively high specific IP 3 R type 1 antagonist, but it may also affect IP 3 R type 3 when used at very elevated concentrations [29]. P 4 -evoked Ca 2+ entry was also found to be dependent on Orai1 and TRPC1 activation. Therefore, PGRMC1 silencing enhanced P 4 -evoked Ca 2+ entry and reversed the P 4 inhibitory role in cell proliferation. In line with this observation, the antagonist of PKA, KT5720, enhanced the P 4 -evoked Ca 2+ entry in all the triple breast cancer cell types evaluated ( Figure 6A), so it did the silencing of PGRMC1 in MDA-MB-231 cells and MCF7 cells. Additional evidence of the existence of a negative regulatory pathway downstream of PGRMC1, which depends on PKA activation results of the evaluation of the level of P-VASP, being the phosphorylation of VASP, impaired in MDA-MB-231 cells lacking PGRMC1 (See Figure 6B), might support the existence of a P 4 /PGRMC1/cAMP/PKA signaling pathway in breast cancer cells, modulating the P 4 -evoked Ca 2+ entry via phosphorylation of Orai1 by PKA, as previously reported [43,44].
The regulation of intracellular Ca 2+ homeostasis is crucial for the activation of several key checkpoint proteins involved in the cell cycle; hence, alteration of this Ca 2+ fluxes might have negative effects on cell proliferation [66]. It has been demonstrated that transient Ca 2+ currents like those conducted by voltage-operated Ca 2+ channels activation are very effective, activating transcription factors to maintain the cells in a quiescent status [67,68]. Ca 2+ currents through the plasma membrane, such as SOCE, have been reported to evoke the activation of the pro-proliferative factor NFAT1 and its translocation to the nucleus [52,69]. Stimulation of cardiac myocytes with P 4 leads to NFAT1 activation that is reversed in the presence of cyclosporine, an antagonist of the Ca 2+ -dependent phosphatase calcineurin, which is the main protein responsible for NFAT1 dephosphorylation, and subsequently, it is required for NFAT1 migration to the cell nucleus [70]. Activation of this pathway has been widely described downstream of the canonical SOCE Ca 2+ channel, Orai1 [70]. Here, in MDA-MB-231 cells, P 4 -STIM2 dependent Ca 2+ entry conducted the activation of Orai1 and TRPC1, and it led to NFAT1 nuclear translocation, as evidenced by the increase in the fluorescence of the NFAT1 reporter. In line with this, PGRMC1 silencing favored an enhanced NFAT1 accumulation in cells treated with P 4 (Figure 9), thus favoring cell proliferation similar to that in prostate cancer cells [71,72]. In addition, NFAT1 has been reported to be overexpressed in several breast types like pancreatic, lung, and other cancer types [73,74]. Interestingly, Ca 2+ entry through SOCE, a mechanism that relies on Orai1/STIM1 interaction, leads to NFAT1 activation due to a calcineurin-dependent dephosphorylation mechanism, as described above. Lack of STIM2 role in proliferation has been recently described in other cancer cell lines. Contrarily, a recent paper has claimed that STIM2 is responsible for the epithelial-mesenchymal transition (EMT), while others have pointed out that it may be involved in migration and promoting metastasis [75][76][77]. So, during cancer onset, PGRMC1 might be temporally downregulated, and thus, cancer cells might avoid the negative effect of P 4 in proliferation (shown in Figures 7 and 8), and in parallel, these cells might get the benefit of the activation of NFAT1 that would allow the MDA-MB-231 cells to migrate and initiate other cancer sprouts (shown in Figure 9). Figure 10 summarizes the main ideas supported by our experimental data, and according to the experimental evidence shown in the present manuscript, patients with TNBC might get the benefit of P 4 administration if an elevated PGRMC1 expression is previously confirmed in their breast tissues samples.
Although future investigation will be required to confirm this hypothesis, this idea is reinforced by the observations done in mice where P 4 reduces tumor size and brain metastasis evoked by exogenously injected MDA-MB-231 cells [5,8,54]. Although MDA-MB-231 cells present a truncated nPR with the ability of binding P 4 , the reported effects of this hormone should be limited to the presence of alternative receptors to the classic nPR [23]. In fact, P 4 enhances proliferation in T47-D, HER2-overexpressing cells, MCF7, and MCF10A cell lines, exhibiting normal to high nPR expression [78,79]. As previously mentioned, this discrepancy in the P 4 effects among these cell types and the TNBCs can be explained by the different expression of the classic progesterone receptors, and particularly, by the absence or presence of the nPR, which should be considered in future investigations.

Cell Lines and Genetic Modification
MCF10A cell line was provided by Potier-Cartereau (Université François Rabelais, Tours, France). MCF7 and the triple-negative breast cancer cell lines-MDA-MB-231, MDA-MB-468, and BT20, were collected from ATCC ® (Manassas, VA, USA). Cells were cultured following the manufacturer's instructions or the protocols described elsewhere [25]. MCF7 and the TNBC cells were cultured at 37 • C with 5% CO 2 in DMEM medium supplemented with FBS and penicillin/streptomycin; meanwhile, MCF10A cells were cultured in DMEM F-12 supplemented with horse serum, insulin, hydrocortisone, EGF, cholera toxin, and penicillin/streptomycin. Cell transfection was achieved by cell permeabilization with DharmaFECT transfection reagent, and subsequently, cells were incubated for 48 h either with 1-3 µg/mL of esiRNAs against STIM1 and STIM2 or siRNA. Alternatively, we used 2 µg/mL of scrambled plasmid, shPGRMC1, shOrai1, and shTRPC1 plasmids. In order to ascertain the efficiency of silencing protocols, Western blotting using the specific antibodies was done, as described below. Finally, cell transfection was done using the YFP-NFAT1-reporter overexpression plasmid (2 µg/mL), which was evaluated upon 72 h of transfection in order to ascertain NFAT1 translocation to the nucleus due to Ca 2+ entry activation [80].

Cell Proliferation Assay
Non-genetically modified MDA-MB-231 cells (scrambled cells) or MDA-MB-231 cells transfected with the respective genetic agents were seeded at 5 × 10 5 cells/mL in 96-well plates. Following this, cells were cultured for 72 h either in the absence (control) or presence of P 4 (1 µM). Proliferation cell rate at time 0, 24, 48, and 72 h was analyzed by using a commercial BrdU-based proliferation kit from Abcam [25]. Briefly, cells were incubated at the time indicated with BrdU for 2 h and subsequently fixed. Upon cell permeabilization, samples were incubated with an anti-BrdU antibody and with the reporter dye for 30 min. Immediately after the addition of the stop solution, the light emitted at 650 nm by the reporter dye was monitored by using a Plate-reader (Epoch, Bioteck ® ; Swindon, UK).

Analysis of the Changes in the Cytosolic Free Ca 2+ Concentration (Ca 2+ ) c
MDA-MB-231 cells were either transfected for 48 h with the indicated genetic material or were kept untreated. The day before the experiments, cells were detached, the medium was replaced, and finally, they were seeded again at 1 × 10 5 cells/mL in a 6-well plate containing a sterile coverslip that was deposited at the bottom of the wells. Cells were allowed to spread onto the coverslips for an additional 24 h. The next day, cells were incubated for 30 min with 2 µM of fura-2/AM at 37 • C in the incubation chamber.
Coverslips were mounted in a perfusion chamber and were maintained in a Ca 2+ -free HBS medium (in mM: 125 NaCl, 25 HEPES, 5 KCl, 1 MgCl 2 , pH 7.4, and supplemented with 0.1% (w/v) bovine serum albumin and 5 mM of D-glucose). Fluorescence emitted by the samples at 505 nm, as a result of the alternative excitation at 340/380 nm wavelengths, was recorded using a cooled sCMOS camera (Zyla 4.2, Andor, Belfast, UK). Fura-2 fluorescence was represented as the changes in the ratio of fluorescence (F n /F 0 ) upon normalizing the ratio by considering the basal values on each experiment, which result from removing the extracellular free Ca 2+ by supplementation of the medium with 75 µM of EGTA (F 0 ). To compare among different experimental conditions, the areas under the curves were calculated by using the integral of the rise in the ratio of fura-2 fluorescence for 2-4 min after the addition of the stimuli to the cells. Ca 2+ entry evoked by emptying the intracellular Ca 2+ stores was visualized upon the addition of 1 mM of CaCl 2 to the extracellular medium, as widely reported elsewhere [25].
Additionally, Ca 2+ mobilization from the endoplasmic reticulum was determined by using the endoplasmic reticulum Ca 2+ dye, GEM-CEPIA1er, which can be excited at 380 nm and presents two maximum emission wavelengths: at 510 nm in the absence of Ca 2+ (green) and at 462 nm in the presence of Ca 2+ (blue). Therefore, the ratio between both emission wavelength (blue/green) can be used to determine the intraluminal Ca 2+ changes within the endoplasmic reticulum, as previously reported by others [65].

Quantification of Cell Death
MDA-MB-231 cells were transfected with shPGRMC1 or the scrambled plasmids, as aforementioned. Cells were cultured for 24 h in the absence or presence of P 4 (1 µM) and, subsequently, they were loaded with propidium iodide (4 µM) for 45 min at 37 • C inside the incubator. Cell middle plane images were acquired at 555/624 of wavelength (Ex/Em, respectively) using a cooled digital CMOS camera (Zyla 4.2, Andor, Belfast, UK) and a fluorescence 40x objective mounted in an inverted microscope (Nikon Eclipse Ti-2, Amsterdam, The Netherlands) that is controlled by the NIS-Elements AR software (Nikon, Amsterdam, The Netherlands). Cell death was expressed as a percentage of positive PI cells with respect to the total cell account in each field.

Visualization of NFAT1 Translocation to the Nucleus
MDA-MB-231 cells (1 × 10 5 cells/mL) were transfected for 48 h with the YFP-NFAT1-reporter overexpression plasmid and shPGRMC1 or with the NFAT1-reporter plasmid alone. Upon confirmation of the positive expression of NFAT1, which was done by visualizing the YFP fluorescence derived from the NFAT1-reporter in control proliferating cells, they were seeded at the appropriate concentration (1 × 10 5 cells/mL). The following day, cells were stimulated for 48 h with P 4 (1 µM) and were maintained growing under regular culture conditions (Mock). Alternatively, as an internal control, we artificially forced the internalization of the YFP-NFAT1-reporter into the nuclei by stimulating the cells for 24 h with 1 µM of thapsigargin (TG). This maneuver should activate SOCE and, subsequently, facilitate downstream NFAT1 constant translocation to the nucleus [69]. The analysis of the fluorescence emitted by the NFAT1-reporter was done by using an inverted fluorescence microscope at 24 h and 48 h after the beginning of the cell stimulation with P 4 . Data were normalized by comparing with the NFAT1 fluorescence with respect to the one emitted by the mock cells growing under regular conditions.

Western Blotting
MCF10A, MCF7, and MDA-MB-231 cells (1 × 10 6 cells/mL) were lysed by mixing with an equal volume of NP40 buffer. The protein quantification was done using the BCA-based protocol, and all samples were normalized to 1 mg/mL, before proteins were denatured by mixing with an equal volume of Laemmli's buffer (supplemented with SDS, 5%), following which, they were heated at 95 • C for 5 min. Proteins were separated by SDS-PAGE and were subsequently transferred to nitrocellulose membranes. Non-specific binding sites were blocked by incubating the membranes with the blocking buffer (10 % of BSA). Western blotting was done by incubating the membranes overnight either with the specific monoclonal anti-STIM1, anti-STIM2, anti-PGRMC1, and anti-VASP antibodies diluted in TBST (supplemented with BSA at 1:1000). The excess of the primary antibody was removed by washing for 30 min with TBST. Following this, the membranes were incubated for an additional hour with the appropriate HRP-conjugated secondary antibodies (diluted 1:10,000 in TBST). Finally, the membranes were exposed to the dura-ECL solution for 5 min. Chemiluminescence images were acquired using a CD-digit chemiluminescence device (Licor ® , Bonsai Lab, Madrid, Spain), and optic densitometry of the images was analyzed by using the Image J (Fiji plugging) (free-software from NIH). Membranes were reprobed with anti-β-actin antibody as a loading control.

Statistical Analysis
A student's t-test was used for establishing a comparison between two groups. Meanwhile, for multiple comparisons, the analysis of the variance (ANOVA), combined with post hoc Tukey's test and Dunnett's tests, was used. p < 0.05 was considered a significant difference between groups.

5.
The silencing of PGRMC1 results in an elevated Ca 2+ entry and, subsequently, increases NFAT1 translocation to the cell nucleus. 6.
PGRMC1 could be targeted to prevent uncontrolled cell proliferation in TNBC.
Author Contributions: C.C. had done most of the experiments. G.M.S. and J.A.R. discussed the results, designed the experiments, and edited the manuscript. P.C.R. designed the experiments, analyzed, presented, and discussed the results, and finally, wrote the manuscript. All authors have read and agreed to the published version of the manuscript.