One of the most often diagnosed malignancies in women globally is breast cancer (BC), being now the second cause of death because of cancer ^[1][2][1,2]. The biological activity and treatment response of BC are influenced by a variety of histological and molecular abnormalities [3]. Despite improvements in the development of diagnostic methods and treatments, the incidence and mortality rate of breast cancer-bearing patients are rising internationally [4]. Age, family history, histological differentiation and grading, and the local and systemic advancement of the disease have all been studied to evaluate the patient risk and choose the best course of action ^[5][6][5,6]. The three main types of breast cancer are classified based on the hormone receptors’ status. The first group consists of tumors that have either tested positive for the estrogen receptor (ER) or the progesterone receptor (PR). The second group consists of tumors that have either tested positive for the human epidermal growth factor receptor 2 (HER2) with or without ER and PR positivity, whereas the third one is called triple-negative breast cancer (TNBC), since these types of tumors lack expression of all three receptors (ER, PR, HER2) [7]. Receptor status, among other variables, has been demonstrated as the one of most important factors in estimating the prognosis and therapeutic response [8]. Furthermore, breast cancer classification based on intrinsic molecular subtypes as a result of the microarray expression profiling has been distinguished ^[9][10][9,10]. These are termed luminal A (ER+PR+ tumors, expressing luminal genes such as ESR1, GATA3, XBP1, and FOXA1; characterized by the low expression of Ki-67), luminal B (ER+ with lower expression of luminal genes, e.g., PGR and FOX1 and a high expression of Ki-67, >20%), HER2-enriched (characterized by the HER2 positivity; however, not all clinically classified HER+ tumors are of these molecular subtype and intermediate expression of luminal genes), basal-like (increased expression of EGFR and basal cytokeratins with low expression of the luminal A-type genes), and claudin-low (ER-, PR-, and HER- tumors are also negative for claudin 3/4/7 and E-cadherin (reviewed in: ^[11][12][13][11,12,13]).

ERs are activated by estrogens and play important roles in the development of several cancers; in particular, breast [14], endometrial [15], and ovarian cancers [16]. Estrogens are a group of low molecular weight lipophilic molecules that occur in three forms: estrone (E1), estradiol (E2; the term estrogen is used in relation to E2, due to its predominant role in physiology), and estriol (E3) [17]; the fourth form produced during pregnancy, namely estetrol (E4), is a fetal estrogen with selective tissue actions [18]. These hormones contain in their structure a steroid skeleton made of four aromatic rings. One of them is the phenolic A ring, which is responsible for binding to the ER [19]. Estrogens, like other steroid hormones, are synthesized at the rough endoplasmic reticulum from its precursor—cholesterol, which is described in detail by Fuentes and Silvera (2019) [20]. Briefly, they are synthesized from androstenedione in the presence of oxygen and NADPH. The crucial enzyme involved in this process is aromatase (CYP19A1), an enzyme that participates in the final stage of E1 and E2 synthesis. The synthesis of estrogens takes place in the gonads (predominantly in the ovaries—granulosa cells), adrenal cortex, and adipose tissue, in smaller amounts also in other tissues, including breast and placenta [21], or fetal liver, in the case of E4 [18]. E1 and E2 can arise from testosterone in peripheral tissues (mainly adipose tissue) in the enzymatic reaction catalyzed via aromatase, which has a significant impact on the level of estrogen synthesis in postmenopausal women [22].

Estrogens, including E2 (the predominant circulating estrogen in humans) are transported in the blood along with specific proteins. They sequentially cross biological membranes by diffusing to the target sites, where they primarily act by attaching to specific nuclear ER. Receptor–ligand complexes can directly silence/activate gene expression or act indirectly by interacting with intracellular signaling molecules. The mechanism of action of estrogens is very diverse, and the nature of the response depends on both the genetic and physiological predisposition of the target cells. Estrogens are synthesized in both sexes; however, at different concentrations and with different functions [23]. These hormones play a significant role in the proliferation and growth of cells associated with reproduction and have a myriad of other cellular functions; for instance, carbohydrate and lipid metabolism, and the regulation of energy homeostasis ^[17][24][17,24]. Importantly, estrogens affect the cardiovascular [25] and central nervous system [26]. The effect of estrogens on the cardiovascular system may be protective, as shown by several studies, including large-scale clinical trials ^[27][28][29][27,28,29], but have also been associated with the risk of coronary heart disease [30]. Furthermore, estrogen-related malfunctions result in several autoimmune, metabolic, or degenerative pathologies and cancers, including the development of breast cancer [17].

The ER plays a key role in the development, progression, and invasion of ER-expressing BC [31]. ER-positive tumors have a more favorable prognosis compared to other BC types and are usually responsive to hormonal treatment. In the absence of ERα expression, BC exhibits more aggressive phenotypes [32].

2. Estrogen Receptors

The ER family includes the nuclear ER (nER) and G protein-coupled estrogen receptor 1 (GPER1) [33]. nER is characterized by conserved domain structures, such as the DNA-binding domain (DBD) and the ligand-binding domain (LBD) [34]. Two major nER isoforms, ERα and Erβ, are responsible for the regulation of the female reproductive system development, the preservation of bone mass, and the protection of the central nervous system, among other physiologically important processes [35]. The evolutionary origin of the estrogen-signaling system remains unclear; however, the research on invertebrates provided insight into the vertebrate pathway. Interestingly, the ER homologs have been identified in amphioxus ^[36][37][36,37], mollusks ^[38][39][38,39], and annelids [40]. Regarding the functional insights, the ERs from amphioxus and mollusks are not activated by estrogens ^[38][41][42][38,41,42], while in two annelid species, transcription is activated in response to the low doses of estrogens upon ER binding [40]. Based on the phylogenetic context, it was hypothesized the ER possibly originated in the bilateralian lineage [43]. In humans, the nERs are encoded by two different genes (ESR1 for ERα [44] and ESR2 for ERβ [45]) as a result of gene duplication in the early vertebrate lineage [46] that are located on different chromosomes—ESR1 is located on chromosome 6 and ESR2 on chromosome 14. The nER is composed of six homologous A-F domains (Figure 1) representing the receptors’ structural regions and having unique functional characteristics. Domains A and B are located at the amino terminus (N-terminal domain) and contain the so-called activation of function domain 1 (AF-1), whose function is to activate the transcription of target genes [20]. Domain C possesses a zinc-finger motif and corresponds to the DBD domain, namely the DNA-binding domain. This domain is responsible for receptor dimerization and binding to the estrogen-dependent genes promoters’ sequences, called estrogen-response elements (ERE) [47]. The D domain is characterized by the presence of a nuclear localization signal (NLS), which, after the binding of a specific ligand, followed by the conformational change caused by this interaction, is exposed, and it is necessary for translocation to the nucleus. Domain D is the so-called hinge region (H), which is responsible for the functional synergy between fragments AF-1 and the second transcriptional activation domain—the AF-2 fragment located at the carboxyl terminus (C-terminus) [48]. The E domain is the ligand-binding domain (LBD), which contains the ligand-binding site (L). The F domain located at the end of the C-terminus probably acts as a modulator of transcriptional activity and is involved in the interaction with the coactivators ^[49][50][49,50]. ERα and ERβ show high homology in the LBD and DBDs, while they differ in the transcription-activating domain (AF-1) [20]. Due to alternative splicing, both receptor subtypes occur in isoforms ^{[20][51][52][53][54]}[20,51,52,53,54]; five shorter isoforms for ERα, and three shorter isoforms and one longer isoform for ERβ [20]. They are also differentially expressed throughout the body ^[55][56][55,56]: ERα predominance is shown by the endometrial cells, ovary, hypothalamus and outgoing ducts’ testicles, while ERβ is expressed mainly in the kidney cells, brain, heart, bones, lungs, intestinal mucosa, prostate, and vascular endothelium. The deregulation of ERα expression and function is closely related to the carcinogenesis process in ovarian, uterine and breast cancer epithelial cells. On the other hand, ERβ inhibits ERα-mediated transcription and estradiol-induced cell proliferation, which is probably the reason why it is associated with benign forms of breast cancer ^[57][58][59][57,58,59]. The ERα/ERβ cellular ratio plays a key role in regulating E2 activity; for instance, in human T47D BC cells [60]. However, approximately 75% of breast tumors are ER-positive [61] and aberrations in the function are associated with ERα. Hence, ERα is one of the main clinical drug targets [62]. The primary function of both receptors is the downstream regulation of gene transcription upon E2 binding to control the cell proliferation and differentiation activated by the ER-dependent signal transduction [63]. GPER1 (also known as GPR30), is the second type of estrogen-dependent receptor and is a member of the transmembrane metabotropic receptors family, which was originally detected in breast cancer tissue [64]. The GPER1 coding gene is located on chromosome 7 [65]. It is created via a single polypeptide with an α-helical structure strongly folded and immersed in the cell membrane, through which the polypeptide chain passes seven times, forming a hydrophobic transmembrane domain [66]. The GPER1 is present in many cells and tissues. mRNA expression was confirmed, e.g., in the ovaries, prostate, thymus, bone marrow, skeletal muscles, liver, lungs, heart, kidney, pancreas, small intestine, and brain [67]. In response to the extracellular signal by its predominant ligand—E2, the GPER1 regulates many cellular processes via a rapid non-genomic dependent mechanism. Compared to normal tissues, GPER1 is detected with a higher expression in breast cancer cells [68].

3. Estrogen Signaling

3.1. Genomic Action of ER

The ER-dependent signaling can be classified as genomic and non-genomic with different activities and pathways involved, respectively (Figure 2). Genomic signaling (Figure 2; bottom panel) depends on the transcriptional activities via the gene expression, while non-genomic (Figure 2; top panel) depends on the activation of various signaling cascades, as reviewed in: ^[20][69][20,69]. In the genomic ER signaling, the complexes of estrogen and the estrogen receptor (ER) are translocated to the nucleus. There, they can indirectly bind to the DNA-binding transcription factors (TFs) via the TF response elements, using protein–protein interactions. By interactions with the coactivator proteins, ER can control the activation of TFs [70]. Nuclear ER can, for example, interact with specificity protein 1 (Sp1) and nuclear factor kappa B (NF-κB) via the so-called “non-classical” activity [71]. The target genes to be modified by the indirect action of ER do not contain the estrogen-response elements (EREs) in their promoters’ regions. The expression of genes that contain EREs can be changed via the direct genomic action of ER. The receptor undergoes a ligand-specific conformational shift after ligand attachment to the ER, enabling the receptor to be released from the heat shock protein complex (HSP90) ^[72][73][72,73]. HSP90 is a molecular chaperone, which protects unbound ER from degradation [74]. Eckert and colleagues have shown nearly 40 years ago [75] that ERα without a ligand is a constantly degraded, short-lived protein (a half-life of 4–5 h). The ERα synthesis and turnover rates were determined in the MCF-7 breast cancer cells. For complete ER-mediated transcriptional activation, histone acetyltransferases (HATs) are necessary. HATs activities enable nucleosome repositioning, chromatin opening, and engagement with the general transcription machinery centered on RNA polymerase II. For example, the p300/CBP acetylates elements of the basal transcription machinery and interacts with other HATs, such as PCAF ^[76][77][78][76,77,78]. Importantly, there is functional crosstalk between the estrogen receptor and other steroid hormone receptors, such as the progesterone receptor (PR), glucocorticoid receptor (GR), and androgen receptor (AR) in breast cancer cells ^{[79][80][81][82][83][84][85]}[79,80,81,82,83,84,85], as well as other cancer cell types, like endometrial ^[86][87][86,87]. These hormones have similar DNA-binding preferences and their genomic binding orchestrates the recruitment of other TFs and chromatin remodeling complexes ^[88][89][90][88,89,90]. Clearly, ER does not function on its own, and its action can be altered by other receptors. For instance, while co-expressed in BC cells, PR is not only an ERα-induced target gene but also an ERα-associated protein, which redirects ERα-associated chromatin binding events ^[81][84][81,84]. This, in turn, results in a unique gene expression in BC cells and is associated with patients’ outcome [81]; however, the mechanistic insight into ER modulation via PR for better BC management needs to be elucidated [84]. AR has also been shown to play a role in ER genomic binding in breast cancer [82] and its function and targeted therapies across BC subtypes have recently been reviewed in [91]. Additionally, in breast cancer cells, the liganded glucocorticoid receptor represses an ERα-regulated transcriptional program [92]. Tonsing-Carter and colleagues [93] have shown that GR modulation decreases ER-positive BC cells’ proliferation and suppresses ER (both wild-type and mutant) chromatin association.

3.2. Non-Genomic Action of ER

In the non-genomic ER signaling (Figure 2; top panel), estrogen binds to the receptor (mbER, i.e., the ER that is situated at the plasma membrane [94] or GPER1, the G-protein-coupled estrogen receptor 1 [95]). This mechanism starts outside of the nucleus and is unrelated to the transcription. The estrogen and ER complexes predominantly activate the kinase pathways. These include MAPK (mitogen-activated protein kinase) via the so-called Ras-Raf-MEK-MAPK pathway and PI3K (phosphatidylinositide 3-kinase)/AKT (serine/threonine kinase) via the PI3K-AKT-mammalian target of rapamycin (mTOR) pathway. The activation of the MAPK signaling pathway by estrogen has been studied in various cell types, including breast cancer [96], neuroblastoma [97], and endothelial [98] cells. Upon estrogen binding to the receptor, the small guanine nucleotide-binding protein—Ras (GTPase) is activated. Next, another protein kinase—Raf is activated, which then phosphorylates the MEK protein. This in turn leads to the phosphorylation and activation of MAPK. As a consequence, several TFs of the activating protein 1 family, e.g., c-Jun and c-Fos, are activated. These then regulate the transcription of the target genes ^{[99][100][101]}[99,100,101]. An alternate pathway—the PI3K-AKT-mTOR, activated by mbER, relies on the direct contact of ER with different proteins; first, the tyrosine kinase Src, then the phosphatidylinositol 3-kinase (PI3K), and the AKT proteins that regulate the mTOR pathway. The AKT-dependent mechanisms of mTOR regulation is a key intracellular system that signals cellular growth and survival, and the hyperactivation of it is involved in the carcinogenesis of the ER-positive BC as well as the resistance to endocrine therapy [102]. The activation of receptors connected to G-proteins is another well-known non-genomic effect of sex hormones. GPER1 is a transmembrane receptor, which, once activated by estrogen or its derivatives, triggers the downstream signaling pathways that can affect a variety of physiological processes [95], such as cell proliferation, angiogenesis, and inflammation. The action of GPER1 generates cyclic adenosine monophosphate from the activation of the adenylate cyclase enzyme. Moreover, upon activation of a receptor by estrogen, the PLC (phospholipase C) enzyme is activated. The activated PLC cleaves phosphatidylinositol 4,5-bisphosphate (PIP2) into two secondary messengers, inositol 1,4,5-trisphosphate (IP3) and diacylglycerol (DAG). IP3 diffuses into the cytoplasm and binds to the IP3 receptors on the endoplasmic reticulum, leading to the release of Ca²⁺ from the endoplasmic reticulum into the cytoplasm. This results in a rapid increase in intracellular Ca²⁺ concentration, which can trigger a variety of downstream signaling events. DAG, on the other hand, remains in the plasma membrane and activates protein kinase C (PKC), another downstream signaling molecule that can regulate various cellular processes [103].