Syntrophins Regulate α1D-Adrenergic Receptors through a PDZ Domain-mediated Interaction*

To find novel cytoplasmic binding partners of the α1D-adrenergic receptor (AR), a yeast two-hybrid screen using the α1D-AR C terminus as bait was performed on a human brain cDNA library. α-Syntrophin, a protein containing one PDZ domain and two pleckstrin homology domains, was isolated in this screen as an α1D-AR-interacting protein. α-Syntrophin specifically recognized the C terminus of α1D- but not α1A- or α1B-ARs. In blot overlay assays, the PDZ domains of syntrophin isoforms α, β1, and β2 but not γ1 or γ2 showed strong selective interactions with the α1D-AR C-tail fusion protein. In transfected human embryonic kidney 293 cells, full-length α1D- but not α1A- or α1B-ARs co-immunoprecipitated with syntrophins, and the importance of the receptor C terminus for the α1D-AR/syntrophin interaction was confirmed using chimeric receptors. Mutation of the PDZ-interacting motif at the α1D-AR C terminus markedly decreased inositol phosphate formation stimulated by norepinephrine but not carbachol in transfected HEK293 cells. This mutation also dramatically decreased α1D-AR binding and protein expression. In addition, stable overexpression of α-syntrophin significantly increased α1D-AR protein expression and binding but did not affect those with a mutated PDZ-interacting motif, suggesting that syntrophin plays an important role in maintaining receptor stability by directly interacting with the receptor PDZ-interacting motif. This direct interaction may provide new information about the regulation of α1D-AR signaling and the role of syntrophins in modulating G protein-coupled receptor function.

␣ 1 -Adrenergic receptors (␣ 1 -ARs) 2 are G protein-coupled receptors that mediate various important physiological functions of norepinephrine (NE) and epinephrine, particularly in the cardiovascular system where they are responsible for regulating vascular tone and peripheral resistance. Three ␣ 1 -AR subtypes have been cloned (␣ 1A -, ␣ 1B -, and ␣ 1D -AR) and display differences in sequence homology and affinities for subtype selective ligands (1). Upon agonist stimulation, all three ␣ 1 -AR subtypes signal through G␣ q/11 to increase phospholipase C activity and intracellular Ca 2ϩ mobilization (1,2). In addition, recent studies using ␣ 1 -AR transgenic and knock-out mice have now revealed that all three ␣ 1 -AR subtypes are important for the regulation of blood pressure (2,3). Therefore, it remains unclear if specific functional differences exist between the ␣ 1 -AR subtypes.
Increasing evidence now suggests that ␣ 1 -AR subtypes display differences in their ability to interact with specific protein binding partners. The first ␣ 1 -AR subtype-selective binding partner identified was tissue transglutaminase II, which selectively associates with ␣ 1B -and ␣ 1D -ARs (4,5). Since then, other proteins found to selectively associate with ␣ 1 -ARs include gC1qR (6), adaptor protein complex-2 (7), regulators of G protein signaling-2 (8), and spinophilin (9). These interactions were shown to be important for the signaling (8,9), trafficking (6), and internalization (7) properties of the ␣ 1 -ARs. Therefore, these differential interactions may contribute to subtype-specific differences between the members of the ␣ 1 -AR family.
Previously, the ␣ 1D -AR was the least studied of the ␣ 1 -AR subtypes due to difficulties in obtaining significant cell surface expression and poor signaling in heterologous systems. Recent studies have reported that this is due to the primary intracellular localization of this receptor (10,11). Subsequent findings indicated that N-terminal truncation (11,12) or heterodimerization with ␣ 1B -ARs (13,14) or ␤ 2 -ARs (15) promotes ␣ 1D -AR cell surface expression and increases coupling to functional responses. However, it remains unknown if there are any ␣ 1D -AR accessory proteins involved in ␣ 1D -AR function and/or expression.
In this study, we have identified several closely related syntrophin family isoforms (␣, ␤1, and ␤2) as novel ␣ 1D -AR binding partners using a combination of yeast two-hybrid screening and biochemical techniques. We report that syntrophins directly interact with ␣ 1D -ARs through a PDZ domain-mediated interaction. The specificity of this association and its potential role in regulating ␣ 1D -AR expression and signaling were examined.
Yeast Two-hybrid Screening-Plasmid pGBKT7/␣ 1D -C-tail (aa 480 -572) was used as bait to screen a human brain pretransformed cDNA library (in pACT2) by using the standard yeast mating protocol stated in the manual. Yeast were plated on high stringency selective medium (SD/ϪLeu/ϪTrp/ϪHis/ϪAde) and incubated for 7 days at 30°C. Positive colonies were restreaked on selective medium (SD/ϪLeu/ϪTrp/ ϪHis/-Ade). Yeast DNA extracts were obtained by using QIAprep Spin Miniprep kit supplemented with 250 l of 425-600 M glass beads in buffer P1 by vortexing for 5 min. Library plasmid DNA was rescued from positive colonies by transforming yeast DNA extracts into Escherichia coli TOP 10 FЈ cells and selected by 30 g/ml ampicillin. Meanwhile, yeast DNA extracts were used as the template for PCR using MATCHMAKER 5ЈAD and 3ЈAD LD-insert screening amplimers as primers and subjected to further sequence analysis. To test the specificity of the interaction, isolated library cDNAs were co-transformed into yeast strain AH109 together with a bait plasmid, either pGBKT7/␣ 1D -C-tail, empty vector pGBKT7, or other plasmids as indicated. Transformed yeast were subjected to growth tests on high stringency selective medium.
Plasmid Construction-Mouse ␣-syntrophin in pBluescript II SK(Ϫ) was kindly given by Dr. Stanley Froehner (University of Washington, Seattle, WA). ␣-Syntrophin in pDT mammalian expression plasmid was constructed by cloning an ␣-syntrophin fragment from pBluescript/␣-syntrophin into HindIII/BamHI sites of pDT vector. The N-terminal FLAGtagged or HA-tagged ␣ 1 -AR constructs were constructed as previously described (12,16). The ␣ 1D -AR construct with the PDZ-interacting motif substituted with alanine residues was generated by PCR using the fulllength ␣ 1D -AR as template and the primers CAACCGCCACCTGCA-GACCGTCACCAACTA (forward) and GCCGACTACAGCAACCT-AGCAGCAGCAGCTGCTTAAACGCGTGCT (reverse), digested with AgeI and MluI, and ligated with HA-Ntr␣ 1D construct (⌬1-79 truncation) (12) to replace the corresponding normal C-terminal domain. The resultant ␣ 1D -AR construct was sequenced to confirm the alanine substitution at the PDZ-interacting motif. The chimeric ␣ 1B -AR construct with the ␣ 1D -AR C terminus was constructed by the following. A silent mutation forming an EcoRI site (GAATTC) was introduced to the conserved EFK motifs at the ␣ 1B -AR (1062G3 A) and the ␣ 1D -AR construct (1224G3 A), respectively. The C-terminal regions of those two mutated constructs were swapped by EcoRI and MluI to generate the C-terminal chimeric receptor constructs.
Blot Overlay Assays-The interaction between the GST-tagged receptor C-terminal fusion proteins and His 6 /S-tagged PDZ domain fusion proteins was assayed via blot overlay assays. 2 g of purified His 6 -tagged syntrophin PDZ domain fusion proteins were run on a 4 -20% Tris-glycine SDS-PAGE gel for 1.5-2 h at 125 V and then transferred to ProTran nitrocellulose. The blot was blocked in Tris-buffered saline containing 0.1% Tween 20 (TBST) consisting of 5% nonfat milk for 1 h at room temperature (RT), subsequently incubated with 25 nM GST-tagged ␣ 1D -C-tail fusion proteins containing 2% nonfat milk for at least 1 h at RT, washed 3 times with TBST, incubated at RT with monoclonal anti-GST antibody, washed 3 times with TBST, and then incubated with a HRP-conjugated anti-mouse IgG secondary antibody. After 6 washes with TBST, bands were visualized with ECL. PSD95-PDZ3 and MAGI2-PDZ1 fusion proteins were used as controls. In other experiments 2 g of purified GST or GST fusion proteins as indicated were run on an SDS-PAGE gel and overlaid with 25 nM His 6 /S-tagged indicated PDZ domain fusion proteins. Interaction was detected by HRP-conjugated S protein and ECL.
Cell Culture and Transfection-HEK293 cells were propagated in Dulbecco's modified Eagle's medium (4.5 g/liter glucose) plus 10% heatinactivated fetal bovine serum, 100 mg/liter streptomycin, and 10 5 units/liter penicillin at 37°C in a humidified atmosphere with 5% CO 2 . For transient transfection, 8 g of indicated plasmid DNA was mixed with Lipofectamine2000 and serum-free medium at RT for 20 min and added to HEK293 cells growing in a 150-mm tissue culture plate. Cells were harvested 48 -72 h after transfection for further experimentation. For stable transfection cells were selected in the presence of 400 -800 g/ml Geneticin.
Membrane Preparation-For radioligand binding, cells grown on 150-mm tissue culture plates were harvested in phosphate-buffered saline (10 mM phosphate buffer, 2.7 mM KCl, 137 mM NaCl, pH 7.4), and membrane preparations were prepared as previously described (18). For immunoprecipitation, cells were collected by centrifugation at 30,000 ϫ g for 20 min, and membrane preparations were prepared as previously described (13).
Radioligand Binding-Receptor density was determined by saturation binding assays as previously described (18). 3

H-Labeled Inositol Phosphate (InsP) Formation-Receptor function in transfected HEK293 cells was measured by [ 3 H]inositol phosphate
(InsP) formation upon NE stimulation as previously described (18). 10 Ϫ4 M carbachol was used as a control.
Data Analysis-Data were expressed as the mean Ϯ S.E. of results obtained from the indicated number of observations. For saturation binding assays, K D and B max were calculated by nonlinear regression using Prism (GraphPad).

RESULTS
␣-Syntrophin Is a Specific Binding Partner for the ␣ 1D -AR C Terminus-To identify novel ␣ 1D -AR-associated proteins, a human ␣ 1D -AR partial C terminus (␣ 1D -C-tail) was used as bait to screen a human brain pretransformed cDNA library. From a total of 3 ϫ 10 5 -independent diploids screened, 9 positive clones were obtained. One positive clone was identified as a gene fragment of human ␣-syntrophin (h␣-syn). This fragment contained an intact PDZ domain and is displayed in schematic form in Fig. 1. The ␣ 1D -AR is known to contain a putative PDZ-interacting motif at the distal end of its C-tail ( 568 RETDI 572 ) that is highly homologous to the conserved PDZ-interacting motif ((K/Q/R)E(S/ T)X(V/I)) previously demonstrated to be recognized by syntrophins (20 -22). The specificity of the interaction between h␣-syn and ␣ 1D -Ctail was confirmed by further yeast two-hybrid analysis. The h␣-syn library construct was co-transformed into yeast strain AH109 with individual baits ␣ 1A -, ␣ 1B -, or ␣ 1D -C-tails. Transformed yeast containing the indicated bait and h␣-syn ( Fig. 2A) were subjected to growth tests on selective medium. As shown in Fig. 2B, only yeast co-transformed with h␣-syn and ␣ 1D -C-tail were able to grow on high stringency selective medium. These findings suggest that h␣-syn specifically interacts with the C terminus of ␣ 1D -but not ␣ 1A -or ␣ 1B -ARs.
Syntrophin Isoforms Co-immunoprecipitate with ␣ 1D -ARs in HEK293 Cells-Because the ␣-syntrophin PDZ domain was found to specifically interact with the ␣ 1D -C-tail in yeast (Fig. 2B) and in blot overlay assays (Figs. 3 and 4), we next determined whether full-length ␣ 1D -AR and syntrophins might associate in intact cells. HEK293 cells were co-transfected with ␣-syntrophin and a single N-terminal FLAG-tagged ␣ 1 -AR subtype. Cells were harvested, and membrane preparations were solu-FIGURE 1. Structure of the ␣-syntrophin clone identified from a yeast two-hybrid screen using the human ␣ 1D -AR C terminus as bait. A fragment of ␣-syntrophin with an intact PDZ domain was identified as a novel binding partner for the ␣ 1D -AR. The comparison between the consensus sequence preferred for binding by syntrophins and the putative PDZ-interacting motif at the ␣ 1D -AR C terminus is shown at the bottom.  bilized and immunoprecipitated using an anti-FLAG affinity matrix. Immunoprecipitation of receptors was resolved on a SDS-PAGE gel, and syntrophins were detected by the pan-specific syntrophin antibody that is able to recognize ␣, ␤1, and ␤2 syntrophins (24). As shown in Fig.  5, untransfected HEK293 cells exhibit endogenous syntrophin immunoreactivity that is somewhat enhanced by ␣-syntrophin transfection. Syntrophin co-immunoprecipitated with full-length ␣ 1D -ARs but not with the other two ␣ 1 -AR subtypes. These data confirm that ␣ 1D -ARs and syntrophins can interact in a cellular context.
The ␣ 1D -AR Interacts with Syntrophins through Its C-tail in HEK293 Cells-Because the ␣ 1D -AR C-tail was shown to interact with ␣-syntrophin in yeast, we further tested whether the receptor C-tail is the major determinant for this interaction in mammalian cells, utilizing a chimeric ␣ 1B -AR with the ␣ 1D -AR C-tail (FLAG-␣ 1B / D -ARs). This C-tail chimeric receptor showed similar pharmacological properties and G␣ q/11 /Ca 2ϩ signaling to the wild type ␣ 1B -AR (data not shown). Because HEK293 cells endogenously express syntrophins (Fig. 5, left lane), cells were only transfected with FLAG-␣ 1B -ARs, FLAG-␣ 1D -ARs, and FLAG-␣ 1B / D -ARs. Immunoprecipitation of FLAG-tagged receptors followed by detection for syntrophins showed that syntrophins were associated with ␣ 1D -ARs and ␣ 1B / D -ARs but not with ␣ 1B -ARs (Fig. 6), suggesting that the ␣ 1D C-tail plays an important role in its interaction with syntrophins, probably through the PDZ-interacting motif.
Mutation of the PDZ-interacting Motif Affects ␣ 1D -AR Receptor Expression and Signaling-To determine whether the ␣ 1D -AR/syntrophin interaction might be involved in regulating ␣ 1D -AR function, the PDZ-interacting motif at the ␣ 1D -AR C-tail was substituted with alanine residues ( 568 RETDI 572 3 568 AAAAA 572 ) to disrupt its interaction with syntrophins, and the function of the receptors without (HA-Ntr␣ 1D ) or with the mutated PDZ-interacting motif (⌬PDZ-HA-Ntr␣ 1D ) was examined by measuring accumulation of InsPs upon NE stimulation. N-terminal-truncated ␣ 1D -ARs (Ntr␣ 1D -AR) were used to ensure sufficient expression, since truncation does not affect receptor pharmacological or signaling properties but dramatically increases functional receptor expression on the cell surface (11,12). In control experiments, HA-Ntr␣ 1D was also found to co-immunoprecipitate with syntrophins when transfected in HEK293 cells (data not shown). As shown in Fig. 7, HEK293 cells stably expressing each receptor construct had similar basal InsP formations. Cells expressing HA-Ntr␣ 1D showed a sequential increase in InsP production with increasing concentrations of NE (10 Ϫ7 and 10 Ϫ4 M). However, those cells expressing ⌬PDZ-HA-Ntr␣ 1D barely showed any increase over the basal even at 10 Ϫ4 M NE compared with HA-Ntr␣ 1D . Both cell lines showed similar InsP formation when challenged with 10 Ϫ4 M carbachol to stimulate endogenously expressed muscarinic cholinergic receptors, suggesting the difference in InsP formation by NE was specific to the transfected receptors.
To test whether this dramatic decrease in signaling is caused by a difference in receptor expression, saturation binding assays with the ␣ 1 -AR specific radioligand 125 I-labeled BE were performed (Fig. 8A). The receptor densities were 1916 Ϯ 41 fmol/mg of protein for cells expressing HA-Ntr␣ 1D and 301 Ϯ 29 fmol/mg for those with ⌬PDZ-   HA-Ntr␣ 1D . Because the heterologously expressed ␣ 1D -AR has been found to show a dramatic discrepancy between protein expression level and receptor density, probably due to its intracellular localization (14,16), protein expression of those two ␣ 1D constructs was examined by immunoprecipitation and Western blotting (Fig. 8B). ⌬PDZ-HA-Ntr␣ 1D was expressed at a much lower level (nearly undetectable) compared with HA-Ntr␣ 1D , suggesting that the poor receptor density was caused by impaired protein expression.
Overexpression of ␣-Syntrophin Increases Receptor Protein and Cell Surface Expression-Because our previous data showed that mutation of the PDZ-interacting motif at the receptor C-tail affects receptor protein expression, we wanted to determine whether syntrophins were involved. As shown in Fig. 9A, when ␣-syntrophin was stably overexpressed in HEK293 cells that were then transiently transfected with either the HF-␣ 1D -AR or the ⌬PDZ-HF-Ntr␣ 1D mutant, increased ␣ 1D -AR protein expression (2.3 Ϯ 0.6-fold higher; n ϭ 3) was observed by Western blotting. However, the expression of the ⌬PDZ-HF-Ntr␣ 1D mutant was not significantly altered by stable ␣-syntrophin overexpression.
To test whether this increase in protein expression might result in an increase in functional receptors, saturation binding assays with 125 Ilabeled BE were performed on cell membranes (Fig. 9B). Receptor densities were 1353 Ϯ 271 fmol/mg of protein for cells expressing HF-Ntr␣ 1D alone and 1977 Ϯ 229 fmol/mg (n ϭ 4; p Ͻ 0.02) for cells also overexpressing ␣-syntrophin, indicative of an increase in cell surface expression. However, there was no change in ⌬PDZ-HF-Ntr␣ 1D expression in the presence or absence of overexpressed ␣-syntrophin. This suggests that ␣-syntrophin plays an important role in regulation of ␣ 1D -AR protein expression and/or stability by a direct interaction with the PDZ-interacting motif at the ␣ 1D -AR C-tail.

DISCUSSION
In this study ␣-syntrophin was identified as a novel interacting protein for the ␣ 1D -AR in a yeast two-hybrid screen. Comparison of the C-terminal amino acids (RETDI) of the ␣ 1D -AR with the ideal syntrophin PDZ-interacting motif ((K/Q/R)E(S/T)X(V/I)) (20 -22) showed a remarkable degree of identity. Studies using biochemical assays and co-immunoprecipitation in mammalian cells further determined the interaction specificity between the five syntrophin isoforms (␣, ␤1, ␤2, ␥1 and ␥2) and the three ␣ 1 -AR subtypes (␣ 1A , ␣ 1B and ␣ 1D ). The data indicated that the ␣ 1D -AR strongly interacts with three syntrophin isoforms (␣, ␤1, and ␤2) via its C-terminal domain and that syntrophins specifically associate with ␣ 1D -ARs but not the other ␣ 1 -AR subtypes. Mutation of the PDZ-interacting motif at the ␣ 1D -AR C terminus caused a dramatic decrease in receptor protein expression, binding, and signaling. Overexpression of ␣-syntrophin did not rescue this decrease but did increase both protein expression and binding of the receptor with the intact PDZ interacting motif.
Syntrophins, like many PDZ domain-containing proteins, have been shown to play a key role in anchoring proteins to the cell membrane for properly assembling various signaling complexes, such as neuronal nitric-oxide synthase (25,26), voltage-gated sodium channels (20,27), water channel protein aquaporins-4 (26, 28), stress-activated protein kinase-3 (29), and Grb 2 (30). Five mammalian syntrophin isoforms (␣,   ␤1, ␤2, ␥1, and ␥2) have been cloned, each containing at least one pleckstrin homology (PH) domain, a conserved PDZ domain, and a C-terminal syntrophin-unique domain responsible for interaction with dystrophin (31). PH domains are known to bind to phosphatidylinositol 4,5-bisphosphate, a key component in G␣ q/11 signaling (32). This pathway is the primary signaling mechanism for ␣ 1 -ARs (1). However, the function of PH domains in syntrophins is currently unknown. Although PH domain 1 of ␣-syntrophin has been shown to bind to phosphatidylinositol 4,5-bisphosphate at a biochemical level (33), syntrophins have not yet been directly implicated in G protein-coupled receptor signaling or other phosphatidylinositol 4,5-bisphosphate interactions. Therefore, the finding that syntrophins directly and specifically associate with ␣ 1D -ARs may provide new insights into their functions.
Although previous studies based on protein sequence alignment showed that the five syntrophin isoforms contain PDZ domains with high homologies (23), our findings and other studies suggest that these domains show slightly different specificities in recognizing PDZ-interacting motifs. ␣, ␤2, and ␥1 syntrophins are able to interact with phosphoinositol 3,4-bisphosphate-binding protein TAPP1 (34), whereas only ␣, ␤1, ␤2 syntrophins showed consistent interactions with ␣ 1D -ARs in our experiments, suggesting that ␣ and ␤2 syntrophins may share similarity in terms of recognition of PDZ-interacting motifs. This idea has been supported by a recent study on ␣/␤2-syntrophin null mice, where ␣ and ␤2 syntrophins were found to be able to compensate for the functions of the other isoform (35).
In the present study mutation of the PDZ-interacting motif of the ␣ 1D -AR resulted in greatly impaired receptor protein expression and signaling, suggesting the PDZ-interacting motif may determine receptor expression and/or stability. Besides directing the appropriate cell surface targeting of many proteins via PDZ domain-mediated interactions, syntrophins are also known to regulate the stability of other proteins. Deletion of only three amino acids in the PDZ-interacting motif of aquaporin-4 increased the protein degradation rate from a half-life of 24 to 8 h (28). We found that overexpression of ␣-syntrophin significantly increased ␣ 1D -AR expression but not that with a mutated PDZ-interacting motif, which is consistent with many previous reports demonstrating that syntrophins increase the stability of various cellular proteins (28,36,37). In addition, ␤2-syntrophin was found to stabilize islet cell autoantigen 512 (ICA512) by binding to its C-terminal PDZ-interacting motif, thus masking the nearby PEST sequence (38) and preventing the cleavage of ICA512 by calpain (36). Interestingly, ␣ 1D -ARs also contain a putative PEST sequence between amino acids Arg 446 and Ser 487 (score, ϩ8.22), suggesting that disruption of the interaction between syntrophins and ␣ 1D -ARs by mutating the PDZ-interacting motif may expose the PEST sequence to proteolytic attack, thereby accounting for the dramatic decrease in receptor levels observed in this study.
␣-ARs play an important role in vasoconstriction mediated by NE. Because syntrophins interact with ␣ 1D -ARs, they might modulate receptor expression and may, therefore, affect ␣ 1D -AR-mediated vasoconstriction in syntrophin null mice. However, NE-mediated vasoconstriction in hind limb is not altered in ␣-syntrophin null mice (39). This could be explained by the presence of other compensatory syntrophin isoforms (35), since at least three isoforms associate with ␣ 1D -ARs. Alternatively, this result could be due to the lack of involvement of ␣ 1D -ARs in hindlimb vasoconstriction, since the specific ␣ 1 -AR subtypes involved in this phenomenon are currently unknown. To determine whether syntrophins affect ␣ 1D -AR expression and/or function would require further characterization of ␣ 1D -AR-mediated vasoconstriction in mice lacking multiple syntrophin isoforms.
Another interesting finding in our set of data reported here is that syntrophins seem to interact equally well with ␣ 1D -ARs mainly expressed intracellularly (full-length) (10,11) and ␣ 1D -ARs expressed at the cell surface (N-terminal-truncated receptors) (11,12), suggesting that syntrophins may function in different subcellular compartments. Although syntrophins have been thought to function mainly at the cell surface as a component in the dystrophin-associated glycoprotein complex (31), a recent study shows that ␣, ␤1, and ␤2 syntrophins were found to localize at the cell surface and also in the cytosolic fraction in cardiac muscle (40), suggesting that syntrophins may function independent of association with dystrophins. Because ␣-syntrophin has been found to modulate protein stability, the interaction between syntrophins and ␣ 1D -ARs may partly account for the puzzling observation that in heterologously expressed cells ␣ 1D -ARs accumulate intracellularly at a very high level without being degraded (16).
The data presented in this study clearly show that ␣ 1D -ARs specifically and strongly interact via their C termini with the PDZ domains of syntrophin isoforms (␣, ␤1, and ␤2). This idea is supported by the presence of the well defined syntrophin PDZ-interacting consensus sequence in the last five amino acids of the ␣ 1D -AR C terminus, suggesting that this interaction occurs with a high affinity in vivo. To our knowledge this is the first report of syntrophins interacting with G protein-coupled receptors. In addition, it is interesting that this particular receptor utilizes phosphatidylinositol 4,5-bisphosphate as its major signaling component, providing a potential function for the syntrophin PH domain(s). This novel interaction between syntrophins and ␣ 1D -ARs may play an important role in receptor stability, since mutation of the ␣ 1D -AR PDZ-interacting motif caused a dramatic decrease in receptor protein and function, and overexpression of ␣-syntrophin caused an increase in receptor expression. Because this interaction is unique to the ␣ 1D -AR, further investigation of the functional role of this interaction may broaden our knowledge of the differences between the three ␣ 1 -AR subtypes.