No abstract

The G protein-coupled estrogen receptor 1 (GPER) has been demonstrated to participate in many cellular functions, but its regulatory inputs are not clearly understood. Here we describe a new approach that identifies GPER as a calmodulin-binding protein, locates interaction sites, and characterizes their binding properties. GPER coimmunoprecipitates with calmodulin in primary vascular smooth muscle cells under resting conditions, which is enhanced upon acute treatment with either specific ligands or a Ca2+-elevating agent. To confirm direct interaction and locate the calmodulin-binding domain(s), we designed a series of FRET biosensors that consist of enhanced cyan and yellow fluorescent proteins flanking each of GPER’s submembrane domains (SMDs). Responses of these biosensors showed that all four submembrane domains directly bind calmodulin. Modifications of biosensor linker identified domains that display the strongest calmodulin-binding affinities and largest biosensor dynamics, including a.a. 83–93, 150–175, 242–259, 330–351, corresponding respectively to SMDs 1, 2, 3, and the juxta-membranous section of SMD4. These biosensors bind calmodulin in a strictly Ca2+-dependent fashion and with disparate affinities in the order SMD2>SMD4>SMD3>SMD1, apparent Kd values being 0.44±0.03, 1.40±0.16, 8.01±0.29, and 136.62±6.56 µM, respectively. Interestingly, simultaneous determinations of biosensor responses and suitable Ca2+ indicators identified separate Ca2+ sensitivities for their interactions with calmodulin. SMD1-CaM complexes display a biphasic Ca2+ response, representing two distinct species (SMD1 sp1 and SMD1 sp2) with drastically different Ca2+ sensitivities. The Ca2+ sensitivities of CaM-SMDs interactions follow the order SMD1sp1>SMD4>SMD2>SMD1sp2>SMD3, EC50(Ca2+) values being 0.13±0.02, 0.75±0.05, 2.38±0.13, 3.71±0.13, and 5.15±0.25 µM, respectively. These data indicate that calmodulin may regulate GPER-dependent signaling at the receptor level through multiple interaction sites. FRET biosensors represent a simple method to identify unknown calmodulin-binding domains in G protein-coupled receptors and to quantitatively assess binding properties.

Estrogen exerts many effects on the vascular endothelium. Calmodulin (CaM) is the transducer of Ca 2؉ signals and is a limiting factor in cardiovascular tissues. It is unknown whether and how estrogen modifies endothelial functions via the net-

The G protein-coupled estrogen receptor 1 (GPER, formerly also known as GPR30) modulates many Ca-dependent activities in endothelial cells. However, the underlying mechanisms are poorly understood. We recently reported that GPER acts to prolong cytoplasmic Ca signals by interacting with and promoting inhibitory phosphorylation of the plasma membrane Ca-ATPase. In the present study, we examined the role of GPER activation in modulating store-operated Ca entry (SOCE) via effects on the stromal interaction molecule 1 (STIM1). GPER activation by agonist G-1 reduces the peak but prolongs the plateau of bradykinin-induced Ca signals in primary endothelial cells. G-1 dose-dependently inhibits thapsigargin-induced SOCE measured by the Mn quenching method. GPER heterologous expression reduces SOCE, which is further pronounced by G-1 treatment. Consistently, GPER gene silencing in endothelial cells is associated with an increase in SOCE. Treatment with G-1 reduces puncta formation by STIM1 triggered by the activation of SOCE. The effect of GPER activation to inhibit SOCE is not affected by combined nonphosphorylatable substitutions at serines 486 and 668 on STIM1, but is substantially reduced by similar substitutions at serines 575, 608 and 621. Taken together with our recently reported inhibitory actions of GPER on Ca efflux, the current data contribute to a model in which GPER acts to clamp agonist-induced cytoplasmic Ca signals. Kinetic modeling based on current and reported data is used to estimate the overall effect of GPER activation on point activity of endothelial nitric oxide synthase during the time course of agonist-induced total Ca signals.

The angiotensin II receptor type 1 (ATR) mediates many Ca-dependent actions of angiotensin II (AngII). Calmodulin (CaM) is a key transducer of Ca signals in cells. Two locations on the receptor's submembrane domains (SMD) 3 and 4 are known to interact with CaM. However, the binding sites for CaM, biochemical properties of the interactions, and their functional impact are not fully understood. Using a FRET-based screening method, we identified a new binding site for CaM on SMD2 (a.a. 125-141), in addition to SMD3 and the juxtamembranous region of SMD4 (SMD4, a.a., 309-327). Simultaneous measurements of CaM binding and free Ca show that the interactions are Ca-dependent, with disparate K and EC50(Ca) values within the physiological range of cytoplasmic Ca. Full interaction between CaM and SMD3 requires the entire domain (a.a. 215-242) and has an EC50(Ca) value in the range of resting cytoplasmic Ca, suggesting ATR-CaM interaction can occur in resting conditions in cells. AngII induces robust ERK1/2 phosphorylation in primary vascular smooth muscle cells. This effect is suppressed by ATR inhibitor losartan and virtually abolished by CaM antagonist W-7. AngII-induced ERK1/2 phosphorylation is suppressed in cells expressing mutant ATR with reduced CaM binding at each identified binding domain. AngII triggers transient Ca signals in cells expressing wild-type ATR. These signals are reduced in cells expressing mutant ATR with reduced CaM binding at SMD3 or SMD4, but are very slow-rising, low amplitude signal in cells expressing ATR with reduced CaM binding at SMD2. The data indicate that CaM interactions with ATR can occur at various domains, with different affinities, at different physiological Ca levels, and are important for ATR-mediated signaling.