A Four Amino Acid Deletion Polymorphism in the Third Intracellular Loop of the Human α2C-Adrenergic Receptor Confers Impaired Coupling to Multiple Effectors*

The α2-adrenergic receptors (α2ARs) play a critical role in modulating neurotransmitter release in the central and peripheral sympathetic nervous systems. A polymorphism of the α2AR subtype localized to human chromosome 4 (the pharmacologic α2CAR subtype) within an intracellular domain has been identified in normal individuals. The polymorphism (denoted Del322–325) is because of an in-frame 12-nucleic acid deletion encoding a receptor lacking Gly-Ala-Gly-Pro in the third intracellular loop. To delineate the functional consequences of this structural alteration, Chinese hamster ovary cells were permanently transfected with constructs encoding wild-type human α2CAR and the polymorphic receptor. The Del322–325 variant had decreased high affinity agonist binding (KH = 7.3 ± 0.95 versus3.7 ± 0.43 nm; %RH = 31 ± 4 versus 49 ± 4) compared with wild-type indicating impaired formation of the agonist-receptor-G protein complex. The polymorphic receptor displayed markedly depressed epinephrine-promoted coupling to Gi, inhibiting adenylyl cyclase by 10 ± 4.3% compared with 73 ± 2.4% for wild-type α2CAR. This also was so for the endogenous ligand norepinephrine and full and partial synthetic agonists. Depressed agonist-promoted coupling to the stimulation of MAP kinase (∼71% impaired) and inositol phosphate production (∼60% impaired) was also found with the polymorphic receptor. The Del322–325 receptor was ∼10 times more frequent in African-Americans compared with Caucasians (allele frequencies 0.381 versus 0.040). Given this significant loss of function phenotype in several signal transduction cascades and the skewed ethnic prevalence, Del322–325 represents a pharmacoethnogenetic locus and may also be the basis for interindividual variation in cardiovascular or central nervous system pathophysiology.

The ␣ 2 -adrenergic receptors (␣ 2 AR) 1 are cell surface receptors for catecholamines, which couple to the G i /G o family of G proteins. ␣ 2 AR are expressed at multiple sites within the central and peripheral sympathetic nervous systems and may also be expressed at noninnervated sites of peripheral tissues as well. In the central nervous system presynaptic ␣ 2 AR act to inhibit the release of neurotransmitters such as norepinephrine, serotonin, and dopamine. As such, a number of responses have been ascribed to activation of these receptors by endogenous catecholamines or exogenously administered agonists. These include modulation of blood pressure, sedation, analgesia, opiate withdrawal, and multiple complex cognitive and behavioral parameters (1)(2)(3)(4)(5).
Three human ␣ 2 AR subtypes have been cloned and characterized and are denoted as the ␣ 2A , ␣ 2B , and ␣ 2C subtypes (6 -8). Based on chromosomal localization, these have previously been denoted as ␣ 2 C10, ␣ 2 C2, and ␣ 2 C4, respectively. Recent studies including those with genetically engineered mice have shown that the ␣ 2C subtype plays specific roles in modulation of the acoustic startle reflex, prepulse inhibition, isolation-induced aggression, spatial working memory, development of behavioral despair, body temperature regulation, dopamine and serotonin metabolism, presynaptic control of neurotransmitter release from cardiac sympathetic nerves and central neurons, and postjunctional regulation of vascular tone (2-5, 9 -11). The therapeutic utility of ␣ 2 AR agonists and antagonists has been limited by the lack of highly subtype-specific compounds as well as marked interindividual variability in efficacy and adverse side effects of available agents.
Given the above, and our recent delineation of functionally significant polymorphisms of the ␤ 1 -and ␤ 2 -adrenergic receptors (12)(13)(14), we have examined the genomic cDNA sequence of the ␣ 2C AR in a cohort of normal individuals. A four-amino acid deletion in the third intracellular loop was found, which was much more common in African-Americans as compared with Caucasians. Recombinant studies revealed that the deletion receptor has a distinct phenotype with a significant loss of signaling to several effector systems.

MATERIALS AND METHODS
Polymorphism Detection-The sequence encoding the third intracellular loop of the human ␣ 2C receptor (GenBank TM assession no. J03853) was examined for polymorphic variation by performing polymerase chain reactions (PCR) to amplify this portion of the cDNA from genomic DNA derived from blood samples. In this paper the adenine of the initiator ATG codon is designated as nucleotide 1 and amino acid 1 is the encoded methionine. The human receptor consists of 462 amino acids. For initial examination, DNA from 20 normal individuals was utilized. Primers for PCR were: 5Ј-CCACCATCGTCGCCGTGTGGCT-CATCT-3Ј (sense) and 5Ј-AGGCCTCGCGGCAGATGCCGTACA-3Ј (antisense). The PCR consisted of ϳ100 ng of genomic DNA, 5 pmol of each M13 primer, 0.8 mM dNTPs, 10% Me 2 SO, 2.5 units Platinum taq DNA polymerase High Fidelity (Life Technologies, Inc.), 20 l of 5ϫ buffer E (Invitrogen) in a 100-l reaction volume. Reactions were started by an initial incubation at 94°C for four minutes, followed by 35 cycles of 94°C for 30 s, 65°C for 30 s, and 72°C for 1 min, followed by a final extension at 72°C for seven minutes. Attempts to directly sequence this product resulted in ambiguous reads, so the product was ligated into the vector PCR2.1-TOPO (Invitrogen) and TOP 10 cells were transformed. Multiple colonies from each transformation were expanded, and the subsequently isolated DNA was sequenced using an ABI 373A automated sequencer in the forward and reverse directions using dye terminator chemistry with vector T7 and M13 primers. As is discussed below, a 12-bp deletion was found in some individuals beginning at nucleotide 964 (Fig. 1A). This results in the loss of amino acids 322-325 and thus this polymorphic receptor is denoted Del322-325. This deletion results in the loss of a NciI restriction site at nucleotide 974 (forward strand), and thus a rapid detection method was devised. Smaller (384 and 372 base pair) PCR products were produced using 5Ј-AGCCCGACGAGAGCAGCGCA-3Ј as the sense primer and the aforementioned antisense primer (same reaction conditions as above), and genomic DNA derived from blood samples as the template. Within this fragment there are either five or six NciI restriction sites depending on the presence or absence of the deletion, providing for the pattern shown in Fig. 1C. This rapid detection technique was applied to additional DNA samples providing genotypes at this locus from a total of 146 individuals.
Constructs and Cell Transfection-To create the polymorphic ␣ 2C AR construct the larger (723 bp) PCR product described above amplified from a homozygous individual was digested and subcloned into the Bpu1102 I and Eco47 III sites of the wild-type ␣ 2C AR sequence in the expression vector pBC12BI (15). The integrity of the construct was verified by sequencing. Chinese hamster ovary cells (CHO-K1) were permanently transfected by a calcium phosphate precipitation technique as described previously using 30 g of each receptor construct and 0.5 g of pSV 2 neo to provide for G418 resistance (15). Selection of positive clones was carried out in 1.0 mg/ml G418, and expression of the ␣ 2C receptors from individual clonal lines was determined by radioligand binding as described below. Cells were grown in monolayers in Ham's F-12 medium supplemented with 10% fetal calf serum, 100 units/ml penicillin, 100 g/ml streptomycin, and 80 g/ml G418 (to maintain selection pressure) at 37°C in a 5% CO 2 atmosphere.
Adenylyl Cyclase Activities-␣ 2 AR inhibition of adenylyl cyclase was determined in membrane preparation from CHO cells stably expressing the two receptors using methods similar to those previously described (15). Briefly, membranes (ϳ20 g) were incubated with 27 M phosphoenolpyruvate, 0.6 M GTP, 0.1 mM cAMP, 0.12 mM ATP, 50 g/ml myokinase, 0.05 mM ascorbic acid, and 2 Ci of [␣-32 P]ATP in a buffer containing 40 mM HEPES, pH 7.4, 1.6 mM MgCl 2 , and 0.8 mM EDTA for 30 min at 37°C. These conditions minimize the stimulation of adenylyl cyclase, which is observed at high agonist concentrations (16,17). Reactions were terminated by the addition of a stop solution containing excess ATP and cAMP and ϳ100,000 dpm of [ 3 H]cAMP. Labeled cAMP was isolated by gravity chromatography over alumina columns with [ 3 H]cAMP used to quantitate column recovery. Activities were measured in the presence of water (basal), 5 M forskolin, and 5 M forskolin with the indicated concentrations of agonists. Results are expressed as percent inhibition of forskolin-stimulated activity.
MAP Kinase Activation-Activation of p44/42 MAP kinase was determined by quantitative immunoblotting using a phosphospecific antibody. Briefly, confluent cells were incubated overnight at 37°C and 5% CO 2 in serum-free media prior to treatment with media alone (basal), epinephrine (10 M), or thrombin (1 unit/ml) for 5 min. Cells were washed three times with phosphate-buffered saline and then lysed in radioimmune precipitation buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 0.5% deoxycholate, 0.1% SDS, and 5 mM NaF) containing protease inhibitors (10 g/ml benzamidine, 10 g/ml soybean trypsin inhibitor, 10 g/ml aprotinin, and 5 g/ml leupeptin). Western blots of these whole cell lysates were performed essentially as described previously (18) except that polyvinylidene difluoride membranes (Amersham Pharmacia Biotech) were used and incubated with phospho-p44/42 MAP kinase E10 antibody and (after stripping) with the p44/42 MAP kinase monoclonal antibody (both from New England Biolabs, Beverly, MA) at dilutions of 1:2000 for 1 h at room temperature. Washed membranes were subsequently incubated with antimouse fluorescein-linked immunoglobulin followed by incubation with fluorescein alkaline phosphatase (ECF, Amersham Pharmacia Biotech). Fluorescent signals were quantitated by real time acquisition using a Molecular Dynamics STORM imager.
Inositol Phosphate Accumulation-Total inositol phosphate levels in intact cells were determined essentially as described previously (19). Briefly, confluent CHO cells stably expressing each of the ␣ 2C ARs were incubated with [ 3 H]myoinositol (5 Ci/ml) in media lacking fetal calf serum for 16 h at 37°C in 5% CO 2 atmosphere. Subsequently, cells were washed and incubated with phosphate-buffered saline for 30 min followed by a 30-min incubation with 20 mM LiCl in phosphate-buffered saline. Cells were then treated with phosphate-buffered saline alone (basal), 10 M epinephrine, or 5 units/ml thrombin for 5 min, and inositol phosphates were extracted as described by Martin (20). Following separation on Agl-X8 columns, total inositol phosphates were eluted with a solution containing 0.1 M formic acid and 1 M formate.
Radioligand Binding-Expression of mutant and wild-type ␣ 2C AR was determined using saturation binding assays as described (21)  Miscellaneous-Protein determinations were by the copper bicinchoninic acid method (22). Data from adenylyl cyclase and radioligand binding assays were analyzed by iterative least-square techniques using Prizm software (GraphPad, San Diego, CA). Agreement between genotypes was observed, and those predicted by the Hardy-Weinberg equilibrium were assessed by a 2 test with one degree of freedom. Comparisons of results from biochemical studies were paired by t tests, and significance was considered when p Ͻ 0.05. Data are provided as means Ϯ standard errors.

RESULTS AND DISCUSSION
From the initial sequencing of ␣ 2C AR third intracellular loop PCR products from 40 chromosomes, one nonsynonymous sequence variant was identified (Fig. 1). This consisted of an in-frame 12-nucleotide (GGGGCGGGGCCG, sense strand) deletion beginning at nucleotide 964. This results in a loss of Gly-Ala-Gly-Pro at amino acid positions 322-325 within the third intracellular loop of the receptor (Fig. 2). Other than this deletion, the remaining encoded sequence was identical to that shown in Fig. 2. The frequencies of the wild-type and the Del322-325 polymorphic ␣ 2C ARs are shown in Table I. The polymorphism is rare in Caucasians with an allele frequency of 0.040. In contrast, the frequency is ϳ10-fold higher (0.381) in African-Americans. The distribution of homozygous and heterozygous alleles was not different than that predicted from the Hardy-Weinberg equilibrium (p Ͼ 0.8). No other nonsynonymous polymorphisms were found in the third loop sequence. However, five synonymous single nucleotide variations were found at nucleic acids 868, 871, 933, 996, and 1167.
The consequences of this polymorphism on receptor function were evaluated by permanently expressing the human wildtype ␣ 2C AR and the Del322-325 receptor in CHO cells and examining multiple signaling pathways. As indicated, multiple clones with similar expression levels were utilized for these studies. Saturation radioligand binding studies using the ␣ 2 AR antagonist [ 3 H]yohimbine revealed that Del322-325 had a slightly, but statistically significant, lower affinity for the radioligand compared with wild-type ␣ 2C AR (K d ϭ 3.8 Ϯ 0.55 versus 2.0 Ϯ 0.14 nM, n ϭ 5, p ϭ 0.03, Table II). In competition studies with the antagonist phentolamine, however, no differences in the K i values were found between the Del322-325 polymorphism and the wild-type ␣ 2C AR (11.1 Ϯ 1. Indeed, the location of the deletion in the third intracellular loop of the receptor is within 15 residues of the sequence RRGGRR. This is a motif (BBXB or BBXXB) that has been identified in a number of receptors as a G i coupling domain (23,24). We considered that the deletion of the two glycines or the proline in the Del322-325 receptor may induce conformational changes affecting this region or other G protein coupling domains. Functional studies examining agonist-promoted inhibition of forskolin-stimulated adenylyl cyclase activities were carried out in lines with the wild-type ␣ 2C AR and the Del322-325 receptor at expression levels of 1375 Ϯ 141 versus 1081 Ϯ 157 fmol/mg (n ϭ 5, p Ͼ 0.05) and a second set of lines with lower expressions of 565 Ϯ 69 versus 519 Ϯ 51 fmol/mg (n ϭ 5, p Ͼ 0.05), respectively. The results of these studies are shown in Table II and Fig. 3. As can be seen, there is a marked functional difference between the two receptors. In the higher expressing lines (Fig. 3A), wild-type ␣ 2C AR exhibited a maximal inhibitory response of 60 Ϯ 3%. In contrast, the Del322-325 polymorphic receptor achieved a maximal inhibition of 31 Ϯ 2% (n ϭ 5, p Ͻ 0.001), which represents an ϳ50% impairment of function. Of note, the EC 50 values for these responses (2.6 Ϯ 0.74 versus 1.2 Ϯ 0.37 nM, respectively) were not different. Results from studies with the lower expressing lines re-vealed an even more striking phenotypic difference between the two receptors. As is shown in Fig. 3B, at these more physiologic levels of expression, agonist-promoted inhibition of adenylyl cyclase with wild-type ␣ 2C AR was 73 Ϯ 2.4%. In marked contrast, the Del322-325 receptor exhibited very little inhibition (10 Ϯ 4.3%, n ϭ 5, p Ͻ 0.001). With the low expressing Del322-325 line, the EC 50 in some experiments could not be calculated because of the minimal response. Analysis of the composite curve of the mean data from all experiments with this line revealed an EC 50 of 29.6 nM. This is in contrast to 4.3 nM calculated in a like manner for the low expressing wild-type line. A similar degree of impairment was also observed with the endogenous agonist norepinephrine (Table III). Agonist-promoted functional activities of the two higher expressing receptors were also explored with full and partial synthetic ␣ 2 AR agonists with diverse structures. (Because some of these agents were weak partial agonists, only the high expressing lines could be used.) As is shown in Table III, the Del322-325 receptor has depressed agonist-promoted coupling to inhibition of adenylyl cyclase with all the agonists tested. Similar to the responses observed with epinephrine and norepinephrine, the Del322-325 receptor showed ϳ50% impairment in the maximal inhibition of adenylyl cyclase compared with the wild-type ␣ 2C AR for UK14304 (full agonist) as well as BHT-933, guanabenz, clonidine, and oxymetazoline (partial agonists), with no significant differences observed in the EC 50 values for these responses.
We next explored coupling of these two receptors to the stimulation of inositol phosphate production. In CHO cells this response is ablated by pertussis toxin, indicating coupling via G i and/or G o (25). The activation of phopholipase C is likely because of both G o -and G i -associated G ␤␥ stimulation of the enzyme (25). As shown in Fig. 4, the loss of function phenotype of the Del322-325 receptor as delineated in adenylyl cyclase experiments was also observed in these inositol phosphate accumulation studies. Epinephrine-stimulated accumulation of inositol phosphates was 30 Ϯ 3% over basal with the wild-type ␣ 2C AR, compared with 11 Ϯ 2% for the Del322-325 receptor (n ϭ 4, p Ͻ 0.005), which amounts to an ϳ60% impairment of function for the polymorphic receptor. Expression levels for the two receptors for these experiments were 806 Ϯ 140 and 733 Ϯ 113, respectively.
Finally, agonist-mediated stimulation of MAP kinase was examined. The mechanism of G protein-coupled receptor-mediated stimulation of this pathway is multifactorial and appears to be both receptor and cell-type dependent (26). For the ␤ 2 AR, coupling to G i , internalization of the receptor and interaction with ␤-arrestin is required for this receptor to activate the MAP kinase cascade (27). Less is known about ␣ 2 AR coupling to this pathway; however, it is clear that it is pertussis toxin-sensitive and that receptor internalization is not necessary (28). For the current studies, MAP kinase activation was assessed using quantitative immunoblots with an antibody specific for the activated (phosphorylated) form of extracellular signal-regulated kinase 1/2. The total amount of MAP kinase was not different between the two cell lines utilized (Fig. 5A). Agonistpromoted activation of MAP kinase was significantly different between the two receptors (Fig. 5), with results expressed both as the agonist-promoted fold increase over basal levels of activated MAP kinase and as the percent of the thrombin response. In five such experiments, MAP kinase activity in Del322-325expressing cells in response to 10 M epinephrine was 57.8 Ϯ 7.0% of the WT␣ 2C AR response (p Ͻ 0.005). When normalized to the thrombin response, epinephrine-stimulated MAP kinase activity was 37 Ϯ 5.7% for the polymorphic receptor compared with 128 Ϯ 10.0% for the wild-type ␣ 2C AR (p Ͻ 0.005).
Recent studies have begun to elucidate specific functions for the ␣ 2C AR subtype. In situ mRNA and immunohistochemical analysis of ␣ 2C AR expression has revealed a distinct pattern of expression in rat brain and spinal cord (29,30). ␣ 2C ARs have been localized primarily in the neuronal perikarya and to a lesser extent in the proximal dendrites, with high levels of receptor expression detected in the basal ganglia, olfactory tubercle, hippocampus, and cerebral cortex (29). These data along with studies of genetically engineered mice indicate that the ␣ 2C AR subtype plays explicit roles in cognitive and behavioral functions. Studies of mice that overexpress or that have targeted inactivation of the ␣ 2C AR gene have shown that this receptor is involved in the regulation of spontaneous motor activity as well as agonist-induced regulation of body temperature and dopamine metabolism (2). In addition, results indicating that activation of ␣ 2C AR reduces hyperreactivity and H]yohimbine binding and adenylyl cyclase assays were performed on membranes prepared from CHO cell lines stably expressing wild-type and Del322-325 ␣ 2C AR as described under "Materials and Methods." Epinephrine competition binding studies were analyzed by nonlinear regression for best fit to a two-site binding model. Adenylyl cyclase activities were determined in the presence of 5.0 M forskolin and increasing concentration of epinephrine.  impulsivity have also been reported (3). These studies show that lack of ␣ 2C AR expression is associated with increased startle reactivity, reduced prepulse inhibition of the startle reflex, and isolation-induced attack latency, whereas overexpression of ␣ 2C AR produces the opposite effects. Consistent with these data, in humans, the ␣ 2 AR agonist clonidine and the ␣ 2 AR antagonist idazoxan reduce and facilitate the acoustic startle response, respectively (31,32). The role of ␣ 2C AR in modulating working memory has also been characterized (4). In these studies, ␣ 2C AR knockout mice performed less accurately in a delayed alternation task and displayed slowed motor initiation in the return phase of the task, supporting a role for the ␣ 2C AR in the cognitive aspect of response preparation. ␣ 2C AR knockout mice were also impaired in spatial and nonspatial water maze tests, thus supporting a role for this receptor in modulating cognitive functions (5), and alteration of ␣ 2C AR expression in transgenic mice has also been linked with behavioral despair development and changes in plasma corticosterone levels (11). Recent studies measuring [ 3 H]norepinephrine release from central neurons and cardiac sympathetic nerves have shown that the frequency-release curves for ␣ 2C AR-deficient mice are rightward shifted compared with wild-type mice (9). Furthermore, the residual agonist-stimulated inhibition of [ 3 H]norepinephrine release observed in ␣ 2A AR-deficient mice was not present in mice deficient in both ␣ 2A and ␣ 2C AR. Thus both subtypes are important in inhibiting neurotransmitter release at these sites. ␣ 2C AR mRNA or receptor protein has also been identified in other peripheral sites (10,33,34) with evidence in some cases indicative of postsynaptic functions (10).
The presence of functionally distinct polymorphic ␣ 2C ARs may account for interindividual variability in physiological responses or may be the basis of differences in clinical characteristics of diseases where ␣ 2C AR function is important. In addition, the Del322-325 polymorphism could conceivably predispose individuals to the development of disease. The response to agonist or antagonist therapeutic agents may also vary depending on receptor genotype. In this regard individuals with Del322-325 might be more sensitive to antagonists because they have receptors that are less responsive to endogenous catecholamines. For agonists, the response or sensitivity would be predicted to be less for those with the polymorphic ␣ 2C AR due to its impaired coupling. Given the relatively high frequency of the polymorphism in healthy African-Americans (Table I), modification of a disease or drug response is more likely than predisposition to disease, although all these possibilities need to be explicitly tested. We and others have recently shown that functional polymorphisms of the ␤ 2 AR indeed appear to have one or more of the above effects in asthma, congestive heart failure, and obesity (35)(36)(37). Interestingly, Comings et al. (38) have found that increased levels of plasma norepinephrine levels in children with attention deficit hyperactivity disorder with learning disabilities were associated with polymorphisms near the coding regions of the ␣ 2A , ␣ 2C , and dopamine ␤-hydroxylase genes.  4. Stimulation of inositol phosphate accumulation by wild-type and Del322-325 ␣ 2C ARs. Total inositol phosphate production in intact CHO cells was measured as described under "Materials and Methods" in response to a 5-min exposure to 10 M epinephrine. Receptor expression was 806 Ϯ 140 and 733 Ϯ 113 fmol/mg, respectively, for these experiments. *, p Ͻ 0.005 compared with wild-type response (n ϭ four experiments). IP, inositol phosphate.
FIG. 5. Stimulation of MAP kinase by wild-type and Del322-325 ␣ 2C ARs. Phosphorylation of MAP kinase was determined in CHO cells by quantitative immunoblotting with enhanced chemifluorescence using antibodies specific for phosphorylated extracellular signal-regulated kinase 1/2. The same blots were stripped and reprobed for total kinase expression, which was not significantly different between the two cell lines (A). Cells were studied after incubation with carrier (basal) (NT), 10 M epinephrine (Epi), or 1 unit/ml thrombin (Thr). Results are depicted as the fold stimulation over basal normalized to the wild-type response (B) and the percent of the thrombin response (C). *, p Ͻ 0.005 compared with wild-type response (n ϭ five experiments).
In summary, we have identified a polymorphic ␣ 2C AR that consists of a deletion of four amino acids in the third intracellular loop of the receptor. Such a deletion has a significant impact on agonist-promoted formation of the active receptor-G protein ternary complex resulting in significantly altered functional signaling to inhibition of adenylyl cyclase, stimulation of inositol phosphate accumulation, and activation of MAP kinase. For all three effector pathways, the Del322-325 receptor displays markedly impaired coupling. The polymorphism is rare in Caucasians but is ϳ10-fold more prevalent in African-Americans with an allele frequency of 0.381. To our knowledge, this is the greatest racial difference in a polymorphism of any G protein-coupled receptor reported to date. Given the extreme phenotype, this locus should be considered a basis for interindividual variation in physiologic responses, disease predisposition or modification, and drug responsiveness.