kelly lab

Estrogen, K+ Channels and Homeostasis

The gonadal steroid estrogen is a pleiotropic hormone that has widespread actions not only on reproductive tissues but also has pronounced effects in the central nervous system (CNS).  For over twenty-five years, my laboratory has been studying the actions of estrogen in the brain, specifically in hypothalamic neurons that control homeostasis and behavior, using molecular, electrophysiological and behavioral techniques. We have discovered a novel estrogenic signaling pathway that can better explain the differences between females and males, not only in terms of the reproductive cycle but also differences in stress responses, motivation and mood.  These actions of estrogen are similar to the effects of neurotransmitters like serotonin in the CNS. We use an in vitro slice preparation to do whole-cell patch recording in hypothalamic neurons followed by single cell reverse transcription-polymerase chain reaction (RT-PCR) to identify changes in the expression of receptors and K+ channel transcripts during different physiological states. 

In collaboration with Drs. Oline Ronnekleiv and Tom Scanlan in the Department, we have developed a new non-steroidal compound that mimics the actions of estrogen in POMC neurons called STX (see Figure 1).  We have discovered that estrogen acts directly on these neurons via a novel estrogen G protein-coupled receptor to alter their activity.  This unique membrane-associated estrogen receptor is Gq-coupled to activation of a phospholipase C-protein kinase C-protein kinase A pathway leading to desensitization of m-opioid and GABAB receptors in POMC and GABA neurons (see Figure 2). We use two different experimental models.  One model is the female guinea pig whose reproductive cycle mimics that of the human.  Another model is the transgenic female mouse in which POMC neurons are tagged with green fluorescence protein in order to selectively target these neurons in the brain slice preparation.  Using these models we are working to identify the phosphorylated target proteins (e.g. K+ channel) that are mediating the desensitized response to µ-opioid and GABAB receptor agonists.

As proof of principle, estrogen is known to control feeding behavior and energy homeostasis, and STX mimics these effects in female guinea pigs. The decrease in body weight gain by STX indicates that estrogen controls body weight gain through in part, the membrane estrogen receptor (see Figure 3).

Finally, we also utilize a custom brain-specific guinea pig microarray chip, developed in collaboration with Dr. Oline Ronnekleiv, to characterize the effects of estrogen on gene regulation in relevant hypothalamic nuclei. We have used this gene chip to identify genes important for neuronal excitability (K+ channels) and signal transduction that are regulated by estrogen treatment in female guinea pigs (see Figure 4).


Figure 1. STX is more potent than estrogen to desensitize the GABAB response in POMC neurons

Figure 1A

A. Schematic showing the protocol for drug administration in the whole-cell patch voltage-clamp experiments (Vhold, -60 mV). The first GABAB receptor-mediated response was generated by perfusing baclofen (at EC50 concentration of 5 µM) until a steady-state outward current was obtained (R1). After drug washout, the current returned to its pre-drug resting level. The cells were then treated with E2 and/or other drugs for 15 min, baclofen (5 µM) was perfused again, and R2 was measured. The effects of estrogen or other drugs on the baclofen response are expressed as a percentage of R2 over R1.

Figure 1B

B. Composite dose–response curves for E2 and STX illustrating the difference in potency between E2 and STX in arcuate POMC neurons. Data are presented as mean ± SEM (n = 4–11 cells/data point. The EC50 for STX was 2.6 nM, which is 17-fold lower than that found for E2 (46.0 nM).  (from Qiu et al., J. Neurosci. 26: 5649, 2006).


Figure 2. A cellular model of the rapid signaling of estrogen in hypothalamic neurons.

Figure 2

Schematic overview showing the rapid versus delayed ER-mediated modulation of neurotransmitter-regulated G-protein-coupled receptors via a membrane-associated ER in hypothalamic neurons. Rapid signaling: E2 activates a membrane-associated ER that is G q-coupled to activation of phospholipase C that catalyzes the hydrolysis of membrane-bound phosphatidylinositol 4,5-biphosphate (PIP2) to inositol 1,4,5 triphosphate (IP3) and DAG. Calcium is released from intracellular stores (endoplasmic reticulum) by IP3, and DAG activates PKC . Through phosphorylation, AC VII activity is upregulated by PKC . The generation of cAMP activates PKA, which can rapidly uncouple GABAB and µ-opioid (µ) receptors from their effector system through phosphorylation of a downstream effector molecule (e.g., the inwardly rectifying K+ channel, or GIRK). ER-mediated modulation of kinase pathways either reduces the capacity of neuromodulators such as GABA and -endorphin ( End) to inhibit hypothalamic neuronal excitability, or augments the ability of neurotransmitters such as glutamate to increase neuronal excitability (data not shown). Delayed signaling: rapid ER-mediated activation of PKA can lead to phosphorylation of cAMP-responsive element-binding protein (pCREB), which can then alter gene transcription through its interaction with the cAMP-responsive element (CRE). Moreover, other isoforms of PKC can phosphorylate raf-1 (Raf), leading to activation of the MAP kinase pathway and new gene transcription and protein (e.g., GPCRs) synthesis in an estrogen-response element-independent manner. (from Qiu et al., J. Neurosci. 23: 9529, 2003.)


Figure 3. Estrogen and STX significantly attenuate the body weight gain in female guinea pigs after ovariectomy.

Figure 3AFigure 3B

The female guinea pigs were ovariectomized (on day 0) and allowed to recover for 1 week before being given bi-daily subcutaneous injections of oil (OIL), estradiol benzoate (EB), or STX (see Materials and Methods). A two-way ANOVA (repeated measures) revealed an overall significant effect of both estrogen and STX (p < 0.001), and post hoc Newman–Keuls analysis revealed daily significant differences between estrogen and oil-treated, and STX and oil-treated groups (**p < 0.01). Bars represent the mean ± SEM of six and four animals per group for estrogen and STX treatment, respectively. (from Qiu et al., J. Neurosci. 26: 5649, 2006).


Figure 4. Functional categorization of the identified genes from the guinea pig cDNA library.

Figure 4

The guinea pig, brain-specific cDNA library of estrogen-regulated genes isolated from the SSH were sequenced, identified and categorized. The pie chart represents the distribution of identified cDNA transcripts on the guinea pig microarray chip in 10 categories determined by their Gene Ontology function. The number besides the category title denotes the number of genes in that group. A total of 710 unique genes were identified while over 480 transcripts from the SSH were not identified using BLAST (blastn) search engine. Many transcripts were isolated more than once in the SSH. (from Roepke et al., Endocrinology 2007.)