br The modulation of ER
The modulation of ERα: from SERMs to TSECs Continued efforts to provide women with efficacious menopausal therapies have generated interest in the development of selective Calphostin C receptor modulators (SERMs). Similar to estrogens, SERMs have been shown to bind to ERs with high affinity, despite lacking the estrogen steroid moiety, and to regulate transcriptional events in a variety of target tissues, and thus are characterized by tissue-specific actions, being agonists and mimicking the effects of estrogen in some tissues, but antagonizing their effects in others (Komm and Mirkin, 2014). The profile of these molecules would selectively preserve the beneficial effects of estrogen in bone tissue (growth and maturation, as well as regulation of bone turnover) and its trophic impact in the urogenital system, while they may reduce their unwanted side effects (mainly increased risk of venous thrombosis and breast cancer). Most of the effects of estrogens are mediated by ERα (Arnal et al., 2017). The key physiological roles of ERβ have been and remain controversial in several respects (Arnal et al., 2017; Bottner, Thelen, and Jarry, 2014). The actual therapeutic strategies are mainly based on the modulation of ERα, although ERβ and GPR30 can play a role in peculiar tissues and pathophysiological conditions (Feldman and Limbird, 2017; Muka et al., 2016; Reslan and Khalil, 2012). ERα belongs to the nuclear receptor superfamily and it exerts its biological functions in several ways. In the classical genomic response, estrogen binds to ERα, leading to conformational changes, dimerization and recruitment of coactivators into the nucleus where they interact with estrogen response elements or other transcription factors to modulate the transcription of target genes (Ascenzi, Bocedi, and Marino, 2006). Ligand-induced transcriptional activity of ERα involves the action of two distinct activation functions (AF), AF1 and AF2. Mouse models targeting ERs were developed to evaluate the relative contribution of each receptor, i.e., ERα (ERαKO mice), ERβ (ERβKO) or activation functions, i.e., AF1 (ERα-AF10) and AF2 (ERα-AF20) (Arnal et al., 2017). Mice deficient in ERαAF1 express a short ERα isoform of 49 kDa, deficient in most of the A/B domain in place of the full length 66 kDa ERα. Studies of this mouse model demonstrated that, whereas ERαAF1 is required for the proliferative action of E2 in the uterus and mammary gland, it is dispensable for vascular effects (including athero-protection), metabolic responses (high fat diet-induced type 2 diabetes (Handgraaf et al., 2013)), and cortical bone protection (Borjesson et al., 2011). To achieve selective inactivation of AF2, the amino acids 543–549, located in helix 12 of ERα, are deleted. ERα-AF20 mice thus appear to express ERα that has no longer any nuclear action, due to the lack of AF-2 (Adlanmerini et al., 2014; Arao et al., 2011; Billon-Gales et al., 2011). The prototypical SERM is tamoxifen, developed in the 1970s for breast cancer treatment. Different mouse models have represented precious tools to understand the mechanisms of action of tamoxifen in physiology and physiopathology. Tamoxifen has been characterized as an AF2 antagonist, which allows ERαAF1 to be active in vitro. The main explanation for its characteristic mechanisms of action relied on ligand-specific induced alterations in the conformation of the ligand-binding domain of ERα (Heery, Kalkhoven, Hoare, and Parker, 1997). Hence, when E2 binds to its ligand binding domain (LBD), helix 12 packs against helices 3, 5/6, and 11, forming the coactivator-binding groove recognized by the LxxLL motifs of coactivators. However, the position of helix 12 in relation to the remainder of the LBD differs when tamoxifen occupies the ligand-binding pocket. This peculiar orientation of helix 12 in the presence of tamoxifen occludes the coactivator-binding groove, whereas it allows the release of the A domain (Metivier et al., 2002) and recruitment of corepressors (Katzenellenbogen et al., 2001). Interestingly, while this distinctive structure of AF-2 allows the recruitment of repressive machineries in the presence of tamoxifen, the agonist activity of this compound requires part of the NH2-terminal region of ERα AF-1. In vivo experiments allowed to definitively demonstrate that tamoxifen acts as an agonist through activation of ERαAF1 since its uterotrophic, athero-protective and metabolic actions are totally abrogated in the ERαAF10 mice (Abot et al., 2013; Fontaine et al., 2013; Guillaume et al., 2017). In women, tamoxifen acts as an estrogen antagonist in breast tissue and it prevents the recurrence of breast cancer when given post-operatively. Tamoxifen mimics the actions of estrogen in bone but also to some extent in the uterus (with an increased risk of endometrial hyperplasia leading to endometrial cancer) and it also increases the risk of venous thrombosis (Cuzick et al., 2011). Despite a favorable cardiovascular profile (especially in the prevention of myocardial infarction (Grainger and Schofield, 2005)), its side effects in the endometrium limit its use to women diagnosed and cured for breast cancer.