2449 Ligand independent activation of neuronal nuclear estro
Ligand-independent activation of neuronal nuclear estrogen receptors may also continue to occur in absence of ovarian estrogens. AF-1 requires intracellular kinase cascade-dependent phosphorylation of specific residues in the A/B domain of nuclear estrogen receptors (Le Goff et al., 1994; Kato et al., 1995; Bunone et al., 1996). The insulin-like growth factor-1 (IGF-1) receptor (IGF-1R) and ERα co-localize in the rodent hippocampus (Cardona-Gómez et al., 2000). The two major IGF-1R effector cascades are the phosphoinositide-3 kinase (PI3K) and mitogen-activated protein kinase (MAPK) family cascades (Kadowaki et al., 1996). Both the PI3K and MAPK cascades can regulate the transcriptional output of ERα in neuronal 2449 (Patrone et al., 1998; Mendez and Garcia-Segura, 2006). Therefore, IGF-1R may activate nuclear estrogen receptors in neuronal cells through regulation of the PI3K or MAPK cascade. Multiple lines of evidence indicate a reciprocally interactive relationship exists between ligand-independent kinase cascade signaling and estrogen receptor function. In hippocampal neurons, membrane-associated estrogen receptors can activate the MAPK cascade through interaction with membrane-associated metabotropic glutamate receptors (Boulware et al., 2005), resulting in enhanced hippocampus-dependent memory (Boulware et al., 2013). Likewise, estrogen-independent activation of intracellular kinase signaling can activate nuclear estrogen receptor signaling. In COS-1 renal carcinoma cells, MAPK-dependent activation of ERα was only observed in the presence of the full agonist 17-β-estradiol (17βE2), the partial agonist tamoxifen, or by genetic removal of the ligand-binding domain entirely (Tora et al., 1989; Ali et al., 1993; Kato et al., 1995). In contrast, similar experiments indicate that MAPK activation in absence of applied ligand is sufficient to activate estrogen receptors in the HeLa cervical cancer and SK-BR-3 breast cancer cell lines (Bunone et al., 1996) as well as the SK-N-BE neuroblastoma (Patrone et al., 1998) and Neuro-2A neuroblastoma cell lines (Mendez and Garcia-Segura, 2006). Unlike the nonsteroidogenic COS-1 cell line (Zuber et al., 1986), the latter cell lines were derived from inherently estrogenic tissue. Both HeLa cells and neuronal cultures have since been shown to endogenously synthesize estrogenic compounds in vitro (Ishikawa et al., 2006; Prange-Kiel et al., 2003). Therefore it is possible that endogenous synthesis of estrogenic compounds in these cell lines functions to relieve inhibition from the unbound ligand-binding domain and permit estrogen receptor activation in absence of exogenously applied ligand. Our lab has found multiple lines of evidence suggesting that both brain-derived estrogen synthesis and ligand-independent activation of nuclear estrogen receptors continue in the rodent brain in absence of any circulating ovarian or exogenously applied estrogens. We first found that increased levels of ERα in the hippocampus of ovariectomized rats enhances hippocampus-dependent memory (Rodgers et al., 2010; Witty et al., 2012), through a mechanism dependent on IGF-1R signaling (Witty et al., 2013). We have also found that intrahippocampal infusion of IGF-1 increases phosphorylation of ERα at serine-118 and expression of estrogen receptor-dependent target genes (Grissom and Daniel, 2016). We have separately shown that continuous, unopposed estradiol exposure enhances hippocampus-dependent memory in ovariectomized rats through a mechanism dependent on brain-derived estradiol (Nelson et al., 2016). Finally, we have shown that significant estrogen response element (ERE)-dependent gene expression continues in the hippocampus and cortex long after ovariectomy in the transgenic ERE-LUC mouse line (Pollard et al., 2018). However, the mechanisms through which IGF-1-dependent kinase activation and ligand availability interact in neuronal cells to regulate ERE-dependent gene expression are poorly defined.