Syntrophin isoforms play specific functional roles in the α_1D-adrenergic receptor/DAPC signalosome

Abstract

α_1D-Adrenergic receptors, key regulators of cardiovascular system function, are organized as a multi-protein complex in the plasma membrane. Using a Type-I PDZ-binding motif in their distal C-terminal domain, α_1D-ARs associate with syntrophins and dystrophin-associated protein complex (DAPC) members utrophin, dystrobrevin and α-catulin. Three of the five syntrophin isoforms (α, β₁ and β₂) interact with α_1D-ARs and our previous studies suggest multiple isoforms are required for proper α_1D-AR function in vivo. This study determined the contribution of each specific syntrophin isoform to α_1D-AR function. Radioligand binding experiments reveal α-syntrophin enhances α_1D-AR binding site density, while phosphoinositol and ERK1/2 signaling assays indicate β₂-syntrophin augments full and partial agonist efficacy for coupling to downstream signaling mechanisms. The results of this study provide clear evidence that the cytosolic components within the α_1D-AR/DAPC signalosome significantly alter the pharmacological properties of α₁-AR ligands in vitro.

1. Introduction

α₁-Adrenergic receptors (AR) are members of the G-protein coupled receptor (GPCR) superfamily that respond to the catecholamines norepinephrine and epinephrine during periods of intense physical activity, stress or injury [1]. Clinically, the α₁-ARs are important drug targets for cardiovascular disease, benign prostatic hypertrophy and post-traumatic stress disorder [1,2]. Of the three α₁-AR subtypes, the physiological roles, pharmacological properties and signal transduction mechanisms of the α_1A- and α_1B-AR subtypes have been the best characterized in vitro and in vivo [1]. Interestingly, few studies exist examining the α_1D-AR subtype because of the inability to obtain significant expression in cultured cells due to retention within the endoplasmic reticulum [3–5]. Given the importance of this receptor in the cardiovascular system, central nervous system and urinary tract [6,7], understanding the mechanisms controlling α_1D-AR functional expression are of critical importance for the development of new therapeutic agents targeting this receptor.

We recently used tandem affinity purification followed by mass spectrometry to reveal α_1D-ARs are expressed as a multi-protein complex, or “signalosome” at the plasma membrane [5,8]. α_1D-ARs interact with the syntrophin family of scaffolding proteins through a PSD95/DlgA/Zo-1 (PDZ)-domain mediated interaction between the embedded PDZ domain in syntrophins and the PDZ-binding motif in the α_1D-AR distal C-terminus [9]. Formation of this complex is obligate forα_1D-AR function, as destruction of the PDZ-binding motif in the α_1D-AR C-terminus results in loss of α_1D-AR drug binding, signal transduction and plasma membrane localization [5,9]. Syntrophins anchor α_1D-ARs to the dystrophin-associated protein complex (DAPC), which includes dystrophin, utrophin, dystrobrevins and α-catulin. Through this complex, necessary signaling molecules PLC-β2 and Rho-GEF are recruited in close proximity to the receptor [8]. This signaling complex is not mimicked by the α_1A or α_1B-AR subtypes, and has the potential to be used by any of other GPCRs containing PDZ-binding motifs in their C-terminal domains (i.e. 5-HT2, β-ARs) [10].

Of the five syntrophin isoforms identified (α, β₁, β₂, γ₁, γ₂), three isoforms α-, β₁- and β₂-syntrophin interact with α_1D-ARs [9]. However, the function of each syntrophin isoform within the α_1D-AR/DAPC signalosome remains to be uncovered. Previous studies provide clues; α_1D-AR functional responses are unaffected in α- or β₂-syntrophin knock-out (KO) mice, but are completely ablated in double α/β₂-syntrophin KO mice [5]. Additionally, syntrophin isoforms demonstrate selectivity in their protein–protein interactions [8]. Taken together, these findings suggest that multiple syntrophin isoforms are clustered within the α_1D-AR/DAPC signalosome, and that each performs a specific function within the complex.

Towards addressing this hypothesis, this study examined the possibility that syntrophin isoforms alter the pharmacological and signaling properties of the α_1D-AR signalosome. We calculated equilibrium dissociation constants, potencies and intrinsic activities of a number of α₁-AR selective ligands in cells expressing α_1D-ARs with α-, β₁- or β₂-syntrophin. Syntrophin isoforms had indistinguishable effects on ligand affinity. However, syntrophin isoforms displayed a consistent rank order on binding site density of α > β₂ > β₁, and on agonist potency with β₂ > α > β₁. Our data support the hypothesis that syntrophin isoforms have unique functional roles within the α_1D-AR/DAPC signalosome.

2. Material and methods

2.1. Drugs

[³H]-Prazosin was purchased from Perkin Elmer (Waltham, MS) and [³H]-myo-inositol was purchased from American Radiolabel Chemicals Inc. (St. Louis, MO). 5-MU, A-315456, benoxathian, B MY-7378, cyclazosin, doxazosin, epinephrine, methoxamine, nap hazoline, niguldipine, norepinephrine, phenylephrine, octopamine, synephrine and terazosin were purchased from Sigma–Aldrich (St. Louis, MO).

2.2. Constructs

Mouse β₁-syntrophin and β₂-syntrophin cDNA in pBlu2SKP was provided by Dr. Stanley Froehner (University of Washington, Seattle, WA). Construction of human α_1D-AR and mouse α-syntrophin constructs were previously described in [5].

2.3. Cell culture and transfection

HEK293 cells were propagated in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum, 10 mg/mL streptomycin and 100 units/mL penicillin at 37 °C in 5% CO₂. Constructs were transfected using FuGENE HD transfection reagent (Roche Applied Science, Indianapolis, IN) when cells were ~80% confluent. Stable cells were generated by selection with 100 µg/mL geneticin and/ or 25 µg/mL hygromycin.

2.4. Radioligand binding

[³H]-Prazosin saturation and competition assays were performed as previously described [5]. Briefly, whole cell membranes were harvested from HEK293 cells expressing α_1D-AR with or without various syntrophin isoforms. Fixed membrane concentrations were incubated with [³H]-prazosin for 30 min at 37 °C, samples were then harvested with a cell-harvester (Brandel, Gaithersburg, MD) and counted with a Packard Tri-Carb 2200 CA liquid scintillation analyzer (Packard Instrument Co. Inc., Rockville, MD). Non-specific binding was determined with 10 µM phentolamine. Competition assays were formed with varying concentrations of antagonist and 1 nM [³H]-prazosin. Data were analyzed with Prism Software (GraphPad Software, San Diego, CA).

2.5. ERK1/2 activation

ERK1/2 activation was assayed in a 96-well plate as previously described [5]. Briefly, HEK293 cells expressing α_1D-AR with or without syntrophin isoforms were seeded into a 96-well plate. Cells were serum starved for 24 h, and then stimulated with either phenylephrine (PE) or epidermal growth factor (EGF, positive control) for 10 min. Cells were then fixed with 4% paraformaldehyde, washed with blocking buffer and incubated with anti-ERK and anti-phospho-ERK (Santa Cruz Biotechnology, Santa Cruz, CA) overnight at 4 °C. Cells were then washed, incubated with secondary antibody (IRDye 680 and 800 antibodies, LiCor Biotechnology, Lincoln, NE) and imaged on a LiCor Odyssey Scanner. Data were analyzed with Prism GraphPad software.

2.6. [³H]-Phosphoinositol (PI) hydrolysis

PI hydrolysis was measured as previously described [8]. Briefly, cell expressing α_1D-AR with or without syntrophin isoforms were seeded into 24-well plates at a density of 1,000,000 cells/well with [³H]-myo-inositol at a concentration of 2 µCi/mL. Cells were incubated overnight at 37 °C, washed with warmed Krebs buffer (w/LiCl) and stimulated with drugs as indicated for 1 h at 37 °C. Buffer was removed, and 800 µl ice-cold methanol was added to each well. Five hundred µl of H₂O was added to each well, and cells were sonicated for 5 s. Samples were transferred to eppendorf tubes and 500 µl chloroform was added. Samples were re-sonicated and then centrifuged at 10,000g for 5 min at 4 °C. One mL of the top, aqueous layer was then added to a pre-washed PI column, washed with 40 mL 5 mM myo-inositol and then eluted with 4 mL Formate/Formic acid mix. Samples were incubated for 1 h at room temperature, than counted on a Packard Tri-Carb 2200 CA liquid scintillation analyzer. Data were analyzed with Prism GraphPad software.

3. Results and discussion

3.1. Effect of syntrophin isoforms on α_1D-AR ligand affinity

α-Syntrophin increases α_1D-AR localization to the plasma membrane, binding site formation and coupling to signaling responses [5]. However, α-syntrophin KO mice have normal resting blood pressure and intact α_1D-AR responses, whereas α-/β₂-syntrophin double KO mice are significantly hypotensive and display disrupted α_1D-AR function. Since β₁-/β₂-syntrophin both interact with theα_1D-AR through a PDZ interaction [9],we wondered if β₁- and/or β₂-syntrophin were necessary for α_1D-AR functional responses, and if so, how they contributed to signalosome mechanics.

It has been well documented that protein–protein interactions can modify GPCR ligand-binding properties and signaling mechanisms [10]. Thus, we questioned if syntrophin isoforms impart pharmacological differences to the α_1D-AR/DAPC signalosome. To test this hypothesis, we measured equilibrium dissociation constants (K_I) for an array of α₁-AR antagonists using [³H]-prazosin competition radioligand binding assays on HEK293 cell membranes expressing α_1D-AR with α-, β₁- or β₂-syntrophin. Representative com petition curves are shown for the α_1D-AR selective antagonist B MY7378 (Fig. 1A), the α_1A-AR selective antagonist 5-methylurapidil (5-MU, Fig. 1B) and the non-selective α₁-AR antagonist terazosin (Fig. 1C). In all cases, antagonists recognized a single binding site within the range of previously reported K_I values in vitro and in vivo (Table 1) [11–15]. Thus, these results suggest syntrophin isoforms do not alter the pharmacological properties of the α_1D-AR/DAPC signalosome.

Fig. 1.

Effect of syntrophin isoforms on α_1D-AR ligand affinity. Assays were performed using HEK293 cells stably expressing α_1D-ARs and α-syntrophin (■), β₁-syntrophin (□) or β₂-syntrophin (Δ). Representative competition binding assays are shown for the α_1D-AR antagonist BMY 7378 (A), the α_1A-AR antagonist 5-methylurapidil (B) and the nonselective α₁-AR antagonist terazosin. (C) K_I values were calculated using non-linear regression and are shown in Table 1. Data were normalized to a percent [³H]-prazosin bound. Data points are the mean +/− SEM from 2–3 experiments performed in triplicate.

Table 1.

Equilibrium dissociation constants for α₁-AR antagonists at the α_1D-AR subtype. Data are expressed as pK_I values (M) and are the mean +/− SEM of 3 experiments performed in triplicate.

Antagonist	+α-syntrophin	+β1-syntrophin	+β2-syntrophin
BMY 7378	8.04 ± 0.11	8.46 ± 0.08	9.18 ± 0.16
5-MU	7.56 ± 0.17	7.63 ± 0.17	7.44 ± 0.41
Terazosin	7.89 ± 0.11	8.27 ± 0.07	8.82 ± 0.14
Benoxathian	7.76 ± 0.07	8.41 ± 0.04	8.40 ± 0.23
Cyclazosin	8.83 ± 0.14	9.03 ± 0.05	8.23 ± 0.19
Niguldipine	5.95 ± 0.30	6.30 ± 0.12	6.31 ± 0.10
Doxazosin	8.26 ± 0.10	8.57 ± 0.06	9.02 ± 0.10
A-315456	7.25 ± 0.13	7.27 ± 0.08	7.42 ± 0.21

3.2. Effect of syntrophin isoforms on α_1D-AR binding site density

Next, we compared the ability of each syntrophin isoform to increaseα_1D-AR binding site density.α-, β₁- or β₂-syntrophin were stably co-expressed with α_1D-AR in HEK293 cells, lysed into membranes and used for [³H]-prazosin saturation radioligand analysis. Both β₁- and β₂-syntrophin significantly increased α_1D-AR binding site density over α_1D-AR expressed alone (Table 2, Fig. 2). Surprisingly, syntrophin isoforms displayed significant differences in their Bmax values (α = 285.2 fmol/mg; β₁ = 123.4 fmol/mg; β₂ = 138.4 fmol/mg), with no difference in prazosin affinity (K_D α = 0.8 nM; β₁ = 0.87 nM; β₂ = 0.52 nM). Quantitative RT-PCR and western blotting confirmed syntrophin isoform expression levels were equivalent in each cell line (data not shown). Thus, syntrophin isoforms demonstrate a marked difference in their ability to promote proper folding of α_1D-AR binding sites at the plasma membrane, with rank order α-syntrophin > β₂-syntrophin > β₁-syntrophin.

Table 2.

Effect of syntrophin isoforms on α_1D-AR binding site density and signal transduction coupling. The α_1A-AR +/− α-syntrophin and α_1D-AR +/− α-syntrophin data were previously reported in (REF). Phenylephrine concentration–response curves for stimulating PI hydrolysis and ERK1/2 phosphorylation were generated and used to calculate potency (pEC₅₀) and maximal responses (normalized to maximal α_1A-AR responses). Data shown are means +/− SEM from 3 experiments performed in triplicate.

			PI Hydrolysis		ERK1/2 Phosphorylation
Cell line	Bmax (fmol/mg)	Kd (nM)	pEC50 (M)	Max (%)	pEC50 (M)	Max (%)
α1A-AR	674.9 ± 148.1	1.56 ± 0.62	6.11 ± 0.14	102.0 ± 5.3	6.27 ± 0.35	87.6 ± 9.7
+α-syntrophin	541.7 ± 28.1	1.14 ± 0.12	6.27 ± 0.07	102 ± 2.7	7.00 ± 0.22	81.7 ± 5.5
α1D-AR	26.6 ± 7.5	0.22 ± 0.21	7.31 ± 0.66	32.6 ± 5.8	5.72 ± 0.66	48.4 ± 11.0
+α-syntrophin	285.2 ± 51.7	0.80 ± 0.33	6.44 ± 0.19	95.9 ± 5.71	8.16 ± 0.56	98.8 ± 12.9
+β1-syntrophin	123.4 ± 13.6	0.87 ± 0.21	6.27 ± 1.08	94.6 ± 36.8	4.18 ± 2.36	67.6 ± 107.8
+β2-syntrophin	138.4 ± 27.6	0.52 ± 0.27	6.89 ± 0.14	295.2 ± 12.7	7.41 ± 0.48	177.4 ± 21.6

Fig. 2.

Effect of syntrophin isoforms on α_1D-AR binding site density. Assays were performed using HEK293 cells stably expressing α_1D-ARs alone (■) or co-expressed with α-syntrophin (□), β₁-syntrophin (Δ) or β₂-syntrophin (□). Binding site densities were measured using [³H]-prazosin saturation radioligand binding assays. Non-specific binding was calculated as [³H]-prazosin binding with 10 µM phentolamine. Bmax and K_D values were calculated using GraphPad Prism software and are reported in Table 2. Data points are the mean +/− SEM from 2–3 experiments performed in triplicate.

3.3. Effect of syntrophin isoforms on α_1D-AR agonist efficacy

Given syntrophins differ in their ability to promote α_1D-AR binding site density, we suspected these increases would directly correlate with enhanced agonist efficacy. To test our assumption, we generated concentration–response curves for the selective α₁-AR full agonist phenylephrine. We measured total inositol phosphate (IP) production using ion exchange chromatography as a functional output, which is downstream of the canonical Gαq/11 signaling pathway. Again, we used the same HEK293 cell lines used in our radioligand binding studies, which stably express α_1D-ARs with either α, β₁ or β₂ syntrophin. As shown in Fig. 3A, phenylephrine produced sigmoidal increases in IP production in syntrophin overexpressing cells relative to cells expressing α_1D-ARs alone. However, phenylephrine maximal responses (or intrinsic activity) did not correlate with the ability of syntrophin isoforms to increase α_1D-AR binding site density. Instead, β₂-syntrophin caused a dramatic increase in phenylephrine maximal responses, while α-syntrophin and β₁-syntrophin produced only moderate increases in agonist maximal responses.

Fig. 3.

Effect of syntrophin isoforms on α_1D-AR signal transduction. (A) PI hydrolysis and (B) ERK1/2 activation assays were performed using HEK293 cells stably expressing α_1D-ARs alone (■) or co-expressed with α-syntrophin (□), β₁- syntrophin (Δ) or β₂-syntrophin (*). Cells were stimulated with increasing concentrations of the α₁-AR selective agonist phenylephrine (PE) for 10 min. Responses are reported as fold activity over basal. PE potencies and intrinsic activities were calculated using GraphPad Prism software. Data points are the mean +/− SEM from 2–3 experiments performed in triplicate.\

This result was particularly surprising. One would expect that as receptor binding site density increases, agonist efficacy should increase as well. Thus, we felt compelled to perform a more thorough analysis to confirm this perplexing finding. First, we repeated this experiment using the same cells, except we used ERK1/2 phosphorylation as a functional output instead of IP production (Fig. 3B). Briefly, cells were cultured on 96 well plates, serum starved for 24 h and stimulated with increasing concentrations of phenylephrine for 10 min. Cells were probed for total and phospho-ERK and imaged using LiCor Odyssey. In agreement with our IP production experiments, phenylephrine maximal responses were significantly greater in β₂-syntrophin expressing cells compared to α- and β₁-syntrophin expressing cells, indicating this effect occurs regardless of the functional output measured.

As an additional test of this intriguing experimental result, we created concentration–response curves for full (norephinephrine, epinephrine) and partial (naphazoline, (+/−) p-octopamine, (+/−) p-synephrine and methoxamine) α₁-AR agonists. If β₂-syntrophin selectively increases α_1D-AR signal transduction coupling, we expected the efficacy of full and partial agonists would be significantly greater in β₂-syntrophin expressing cells then in α- or β₁-syntrophin expressing cells. Data are compiled in Table 3 and representative concentration–response curves for norepinephrine (Fig. 4A), epinephrine (Fig. 4B), methoxamine (Fig. 4C) and naphazoline (Fig. 4D) are shown. Interestingly, potencies (EC₅₀) for each agonist were not significantly different between α-, β₁- or β₂-syntrophin expressing cells, and all drug potencies were comparable to previously reported values [16–19]. However, agonist intrinsic activities were significantly greater in α_1D-AR/β₂-syntrophin cells (Fig. 4, Table 3) compared to α_1D-AR/α-syntrophin cells. Indeed, partial agonists naphazoline and methoxamine acted as full agonists in α_1D-AR/β₂-syntrophin expressing cells.

Table 3.

Effect of syntrophin isoforms on agonist potencies and maximal responses. Maximum responses were normalized to maximal norepinephrine response for each α_1D-AR/syntrophin data set.

	+α-syntrophin		+β1-syntrophin		+β2-syntrophin
Agonist	pEC50 (M)	Max (%)	pEC50 (M)	Max (%)	pEC50 (M)	Max (%)
(+/−) Epinephrine	7.80 ± 0.08	82.5 ± 2.4	7.86 ± 0.11	132.1 ± 5.6***	7.45 ± 0.22	126.4 ± 10.6***
(+/−) Norephinephrine	7.78 ± 0.09	92.6 ± 3.2	7.55 ± 0.13	93.3 ± 4.3	8.00 ± 0.12	98.7 ± 4.4
(−) Phenylephrine	7.17 ± 0.11	73.8 ± 3.0	6.81 ± 0.12	92.5 ± 4.1*	6.90 ± 0.12	109.6 ± 4.8***
Naphazoline	8.00 ± 0.15	32.9 ± 1.6	7.24 ± 0.19	39.4 ± 2.5	7.35 ± 0.19	84.0 ± 5.3***
(+/−) p-Octopamine	4.82 ± 0.31	73.0 ± 10.2	5.17 ± 0.16	104.7 ± 7.0***	5.88 ± 0.23	38.7 ± 3.3***
(+/−) p-Synephrine	5.53 ± 0.23	24.8 ± 2.1	5.38 ± 0.10	70.4 ± 2.8***	5.65 ± 0.26	56.6 ± 5.4***
Methoxamine	4.90 ± 0.11	59.9 ± 3.5	4.89 ± 0.18	66.7 ± 5.3	5.84 ± 0.14	93.6 ± 4.9***

Agonist values obtained in β1 and β2-syntrophin cells were compared to those obtained in α-syntrophin cells using 2 way ANOVA and statistically significant differences are indicated by * (p < 0.05) and *** (p < 0.001). Data shown are the means +/− SEM of 3 experiments performed in triplicate.

Fig. 4.

Effect of syntrophin isoforms on α₁-AR full and partial agonist efficacy. Assays were performed using HEK293 cells stably expressing α_1D-ARs and α-syntrophin (■), β₁-syntrophin (□) or β₂-syntrophin (Δ). Representative concentration–response curves for PI hydrolysis are shown for the full agonists (A) norepinephrine, NE and (B) epinephrine, E and the partial agonists (C) methoxamine, MX and (D) naphazoline, NP. Responses are reported as fold activity over basal. Agonist potencies and intrinsic activities were calculated using GraphPad Prism software and are reported in Table 3. Data points are the mean +/− SEM from 2–3 experiments performed in triplicate.

Taken together, the results of our study suggest syntrophin isoforms play selective roles in the α_1D-AR/DAPC signalosome. α-syntrophin increases α_1D-AR binding site density while β₂-syntrophin enhances α_1D-AR coupling to downstream signaling effectors. These experimental results may explain why α_1D-AR control of blood pressure is abrogated in double α/β₂-syntrophinKO mice [5], as our findings suggest both α- and β₂-syntrophin are required to create a functional α_1D-AR signalosome. However, it remains unknown precisely how α_1D-ARs, syntrophins and associated proteins utrophin, dystrobrevin, α-catulin are organized in the plasma membrane. We previously demonstrated α_1D-ARs form homodimers in vitro and in vivo [5], thus we propose each protomer of the α_1D-AR homodimer associates with a different syntrophin isoform. Protomer 1 may associate with α-syntrophin to increase α_1D-AR homodimer binding site density, while Protomer 2 associates with β₂-syntrophin to facilitate signal transduction. But how does each syntrophin isoform perform a selective function in the signalosome? One potential explanation is different syntrophin isoforms may selectively associate with unique interacting proteins. For example, we previously discovered in cell-based proteomic screens that β₂-syntrophin selectively associates with the peripheral plasma membrane protein CASK (a.k.a. CAMGUK protein 2, calcium/calmodulin-dependent serine protein kinase 3, membrane associate guanylate kinase 2), which is a multidomain scaffolding protein that could potentially recruit in various signaling molecules to the complex [8]. Simultaneously, α-syntrophin could link α_1D-ARs to the cellular cytoskeleton through the DAPC and ensure proper folding and localization within the plasma membrane. Further studies will be required to validate these speculations.

In conclusion, our studies suggest the α_1D-AR/DAPC signalosome is a highly complex and tightly regulated GPCR signaling network, and that syntrophin isoforms play important and selective roles within the signalosome. Interestingly, 25 human GPCRs contain analogous Type-I PDZ binding motifs in their C-terminal domains, so this novel GPCR complex may represent a commonly used apparatus for global GPCR signaling.

Abbreviations

AR

adrenergic receptor

DAPC

dystrophin-associated protein complex

ERK

extracellular signal regulated kinase

GPCR

G-protein coupled receptor

HEK

human embryonic kidney

KO

knock-out

PDZ

PSD95/DlgA/Zo-1