Nuclear alpha1-adrenergic receptors signal activated ERK localization to caveolae in adult cardiac myocytes

Abstract

We previously identified an alpha1-AR-ERK (alpha1A-adrenergic receptor-extracellular signal-regulated kinase) survival signaling pathway in adult cardiac myocytes. Here, we investigated localization of alpha1-AR subtypes (alpha1A and alpha1B) and how their localization influences alpha1-AR signaling in cardiac myocytes. Using binding assays on myocyte subcellular fractions or a fluorescent alpha1-AR antagonist, we localized endogenous alpha1-ARs to the nucleus in wild-type adult cardiac myocytes. To clarify alpha1 subtype localization, we reconstituted alpha1 signaling in cultured alpha1A- and alpha1B-AR double knockout cardiac myocytes using alpha1-AR-green fluorescent protein (GFP) fusion proteins. Similar to endogenous alpha1-ARs and alpha1A- and alpha1B-GFP colocalized with LAP2 at the nuclear membrane. alpha1-AR nuclear localization was confirmed in vivo using alpha1-AR-GFP transgenic mice. The alpha1-signaling partners Galphaq and phospholipase Cbeta1 also colocalized with alpha1-ARs only at the nuclear membrane. Furthermore, we observed rapid catecholamine uptake mediated by norepinephrine-uptake-2 and found that alpha1-mediated activation of ERK was not inhibited by a membrane impermeant alpha1-blocker, suggesting alpha1 signaling is initiated at the nucleus. Contrary to prior studies, we did not observe alpha1-AR localization to caveolae, but we found that alpha1-AR signaling initiated at the nucleus led to activated ERK localized to caveolae. In summary, our results show that nuclear alpha1-ARs transduce signals to caveolae at the plasma membrane in cardiac myocytes.

Figure 1. Endogenous α1-adrenergic receptors localize to the nucleus in WT adult mouse cardiac myocytes

(a) WT myocytes were fractionated by homogenization and ultracentrifugation. Membrane (M), cytosolic (C), and nuclear (N) fractions (30 μg) were validated by Western blots for caveolin-3, GAPDH, LAP2, and SERCA. α1-AR levels in WT membrane and nuclear fractions were determined by a single-point, maximum-concentration, binding assay with ³H-prazosin (1.2 nM), where 10 mM phentolamine (RBI, Natick, MA) defined non-specific binding. (b) WT andα1ABKO adult mouse cardiac myocytes (AMMC) were cultured for 24 hr then incubated with 50 nM BODIPY-prazosin (non-specific α1-AR antagonist) for 16 hr, fixed with paraformaldehyde and confocal images captured. Myocyte nuclei are identified by the white arrows. Final magnification=600×. Inset: Isolated nuclei were incubated with 50 nM BODIPY-prazosin for 1 hr and confocal images captured. Final magnification=1,200×.

Figure 2. The α1A and α1B subtypes co-localize with the nuclear membrane protein LAP2 in adult mouse cardiac myocytes

(a) Cultured α1ABKO cardiac myocytes were infected with adenoviruses expressing the α1A-GFP (1,000 MOI) or α1B-GFP (3,000 MOI). After 40 hr, myocytes were fixed with 4% paraformaldehyde and stained with an antibody against the nuclear membrane marker LAP2, and a Texas red-conjugated secondary antibody. Fluorescent and transmitted light images were captured by confocal microscopy and co-localization was determined using Imaris software. Final magnification=600×. (b) Cultured α1ABKO cardiac myocytes were infected as above and fractionated as in Figure 1. Western blots were performed to detect the α1-subtypes using an antibody to GFP.

Figure 3. Prazosin induces α1-adrenergic receptors to move off the nuclear membrane in adult mouse cardiac myocytes

Cultured α1ABKO cardiac myocytes were infected as in Figure 2. After 24 hr, myocytes were treated with prazosin (50 nM) or vehicle (control) for 16 hr then fixed with paraformaldehyde and confocal images captured. Final magnification=600×.

Figure 4. α1-adrenergic receptors localize to the nucleus in adult mouse cardiac myocytes in vivo

Sagittal sections from hearts of α1A-GFP transgenic mice (α1A-GFP Tg) or WT mice were stained with anti-GFP antibody (Santa Cruz) to detect α1A-GFP and visualized using the Mouse-on-Mouse Kit with a FITC-conjugated secondary antibody (Vector Laboratories).

Figure 5. The α1A and α1B subtypes co-localize with Gαq and Phospholipase Cβ1 (PLCβ1) in the nuclear membrane in adult mouse cardiac myocytes

α1ABKO cardiac myocytes were cultured, infected, and fixed as in figure 2 then stained with an antibody against (a) Gαq or (b) PLCβ1 and a Texas red conjugated secondary antibody and confocal images captured and prepared as in figure 2. Final magnification=600×. (c) Western blots were performed on WT cardiac myocyte cell fractions to detect Gαq and PLCβ1.

Figure 6. α1-Adrenergic receptors do not localize to caveolae in adult mouse cardiac myocytes

Cultured α1ABKO cardiac myocytes were infected as in figure 2, fixed with 4% paraformaldehyde, and stained with an antibody against caveolin-3 and an Alexa Fluor-594-conjugated secondary antibody. Fluorescent and transmitted light images were captured by confocal microscopy and co-localization determined using Imaris software. Final magnification=600×.

Figure 7. Catecholamine uptake and activation of nuclear α1-adrenergic receptors in adult mouse cardiac myocytes

(a) Catecholamine uptake assay (Molecular Devices, Sunnyvale, CA) was used to determine norepinephrine transport into WT cardiac myocytes. Specificity for norepinephrine uptake was confirmed by addition of norepinephrine (50 nM or 10 μM) 15 min prior to catecholamine uptake measurement. The results represent triplicate measurements from 4–5 myocyte isolations. (b) Cultured WT cardiac myocytes were treated with CGP-12177A (0–200 μM, membrane impermeable α1-AR antagonist) or prazosin for 15 min, and then with phenylephrine (20 μM, PE) for 15 min. Phospho- and total ERK levels were determined by Western blot. (c) Cultured WT cardiac myocytes were harvested for Western blot detection of OCT3 or fixed with 4% paraformaldehyde, and stained with an antibody against OCT3 (Santa Cruz) and an Alexa Fluor-594-conjugated secondary antibody. Fluorescent images were captured by confocal microscopy. Final magnification=600×. Cultured WT cardiac myocytes were treated with corticosterone (1 μM) for 15 min, then PE (20 μM) for 15 min. Phospho- and total ERK levels were determined by Western blot.

Figure 8. Phosphorylated ERK localizes to caveolae at the plasma membrane in adult mouse cardiac myocytes

(a) Cultured α1ABKO cardiac myocytes were infected as in Figure 2. After 40 hr, myocytes were treated with 20 μM phenylephrine (PE) for 15 min. Myocytes were fixed with 4% paraformaldehyde and stained with an antibody against P-ERK and a Texas red-conjugated secondary antibody. Fluorescent images were captured by confocal microscopy. Final magnification=600×. (b, c) The percentage of myocytes positive for P-ERK (n=100–124 myocytes/culture) was determined comparing (c) α1A-GFP and α1B-GFP (n=3) or (d) theα1A-GFP time course (n=2). Groups were compared by one-way ANOVA with a Tukey’s post-test (P<0.05). (d) Isolated α1ABKO cardiac myocytes were cultured and infected as above then treated with filipin (5–10 μg/ml) or vehicle for 1 hr followed by 20 μM PE for 15 min. Myocytes were fixed and stained with antibodies against P-ERK and caveolin-3, and images were captured by confocal microscopy and pseudocolored using Imaris software (GFP pseudocolored magenta and P-ERK green). Final magnification=600×.