The α1D-adrenergic receptor is expressed intracellularly and coupled to increases in intracellular calcium and reactive oxygen species in human aortic smooth muscle cells

Background The cellular localization of the α1D-adrenergic receptor (α1D-AR) is controversial. Studies in heterologous cell systems have shown that this receptor is expressed in intracellular compartments. Other studies show that dimerization with other ARs promotes the cell surface expression of the α1D-AR. To assess the cellular localization in vascular smooth muscle cells, we developed an adenoviral vector for the efficient expression of a GFP labeled α1D-AR. We also measured cellular localization with immunocytochemistry. Intracellular calcium levels, measurement of reactive oxygen species and contraction of the rat aorta were used as measures of functional activity. Results The adenovirally expressed α1D-AR was expressed in intracellular compartments in human aortic smooth muscle cells. The intracellular localization of the α1D-AR was also demonstrated with immunocytochemistry using an α1D-AR specific antibody. RT-PCR analysis detected mRNA transcripts corresponding to the α1A-α1B- and α1D-ARs in these aortic smooth muscle cells. Therefore, the presence of the other α1-ARs, and the potential for dimerization with these receptors, does not alter the intracellular expression of the α1D-AR. Despite the predominant intracellular localization in vascular smooth muscle cells, the α1D-AR remained signaling competent and mediated the phenylephrine-induced increases in intracellular calcium. The α1D-AR also was coupled to the generation of reactive oxygen species in smooth muscle cells. There is evidence from heterologous systems that the α1D-AR heterodimerizes with the β2-AR and that desensitization of the β2-AR results in α1D-AR desensitization. In the rat aorta, desensitization of the β2-AR had no effect on contractile responses mediated by the α1D-AR. Conclusion Our results suggest that the dimerization of the α1D-AR with other ARs does not alter the cellular expression or functional response characteristics of the α1D-AR.


Background
The α 1 -ARs are members of the class I of the G-protein coupled receptors (GPCR) superfamily [1,2]. Three α 1 -ARs, α 1A -AR, α 1B -AR and α 1D -AR have been cloned and characterized [1,3,4]. These receptors mediate responses to epinephrine and norepinephrine thus making a vital contribution to the control of blood flow and systemic arterial blood pressure. Abnormalities in the regulation of the α 1 -ARs may contribute to the development of hypertension and heart failure [5][6][7][8].
It is well known that the localization and trafficking properties of a receptor can modulate its physiological function [9]. Results from heterologous expression systems have demonstrated that the α 1B -AR is localized on the cell surface, as expected for a GPCR, while the α 1A -AR is localized not only on the cell but also in intracellular compartments [10,11]. In contrast, we have shown that the α. In contrast, we have shown that the α 1D -AR is localized intracellularly [11,12]. These results of nonconical cellular locatization are consistent with emerging data that show specific GPCR families can be localized not only to intracellular sites but also on the nuclear membrane [13].
In recent years, the concept of receptor dimerization has brought a new perspective to GPCR function [14][15][16]. Previous studies reported that the α 1D -AR interacts with the α-AR interacts with the α 1B -AR and the β 2 -AR [17] resulting in the cell surface expression of the α 1D -AR. This has lead to the suggestion that these receptors are capable of heterodimerization [17][18][19][20]. These observations have been made in heterologous systems. However, the role of dimerization in the regulation of cells that natively express all three receptors such as vascular smooth muscle cells has not been well studied. This is due in part to the difficulty of tranfecting smooth muscle cells. To overcome this obstacle, we developed a recombinant adenovirus for the efficient expression of the human α 1D -AR. We show that despite the presence of the other α 1 -AR family members, the α 1D -AR is expressed mainly in intracellular compartments. We further show that while receptor dimerization may occur, it does not appear to alter the functional properties of the α 1D -AR.

Cellular localization
An adenoviral vector was constructed to drive the efficient expression of a GFP-labeled α 1D -AR. Infection of aortic smooth muscle cells with virus expressing the α 1D -AR/ GFP resulted in approximately 80% receptor transfectional efficiency (not shown) demonstrating that adenovirus can be useful in cells that have been traditionally difficult to transfect with the α 1 -ARs. Following viral infection, the α 1D -AR/GFP was detected in intracellular compartments of aortic smooth muscle cells ( Figure 1A). A similar pattern of vascular smooth muscle intracellular expression was seen with immunocytochemistry studies using an α 1D -AR selective antibody ( Figure 1B). These localization results in smooth muscle cells are similar to our previous findings in HEK 293 cells transfected with an α 1D -AR/GFP expression plasmid or immunohistochemistry studies in fibroblasts stably transfected with the α 1D -AR [12,21]. Recently, it was proposed that the α 1D -AR can dimerize with the α-AR can dimerize with the α 1B -AR promoting its cell surface expression [19]. Our results would argue that the presence of other ARs, particularly the α 1B -AR, does not alter α 1D -AR localization in vascular smooth muscle cells. To further substantiate that the presence of the α 1B -AR does not affect α 1D -AR localization, we infected fibroblasts that stably express the α 1B -AR with the α-AR with the α-AR with the α-AR with the α 1D -AR/GFP adenoviral construct ( Figure 1C). Despite expression in a cell expressing the α 1B -AR at high levels, α 1D -AR was nonetheless expressed in intracellular compartments ( Figure  1C). These data argue that the α 1B -AR does not alter the cellular localization of the α 1D -AR.
It is possible, however, that in these cultured smooth muscle cells the α 1B -AR and/or α 1A -AR are not expressed. To examine this possibility we used RT-PCR to assess message expression. The results of these experiments are shown in figure 2A. A very prominent PCR-product corresponding to the α 1B -AR was detected in aortic smooth muscle cells. We were also able to detect an α 1A -AR transcript. A very faint product corresponding to the α-AR transcript. A very faint product corresponding to the α 1D -AR was also observed. Minute levels of tissue expression are typical for this receptor. A series of antibodies directed against each of the α 1 -ARs was used to determine the cellular localization of these receptors ( Figure 2B). As has been shown by previous work the α). As has been shown by previous work the α 1A -AR is expressed both intracellularly as well as on the cell surface while the α 1B -AR is expressed predominately on the cell membrane. What is also apparent from comparing figures 1A-C and figure 2B is that the expression pattern of the α is that the expression pattern of the α 1D -AR is markedly different from that of either the α-AR is markedly different from that of either the α 1A -or αor α 1B -ARs. Therefore, in a mammalian cell where both the α 1A -and αand α 1B -AR are natively expressed (as opposed to being transfected) the α 1D -AR is nonetheless expressed intracellularly. Further, the data argue that while dimerization may occur in smooth muscle cells, it does not alter the localization of the α 1D -AR.

Effects on intracellular calcium
In aortic smooth muscle cells phenylephrine produced a dose-dependent and statistically significant increase in intracellular calcium (see data for 25 uM presented in Figure 3). This increase was antagonized by 1 nM of the nonselective α 1 -AR blocker prazosin or 30 nM of the highly selective α 1D -AR antagonist BMY 7378. To substantiate the selectivity of this dose of BMY 7378 we measured phenylephrine-induced increases in intracellular calcium levels in fibroblasts stably transfected with either the α 1A -AR, α 1B -AR or the α 1D -AR. Despite pretreatment with 30 nM BMY 7378, phenylephrine maintained the ability to promote increases in intracellular calcium in the α 1A -AR or α 1B -AR expressing lines of fibroblasts ( Figure 4). In contrast, BMY 7378 blocked the phenylephrine-induced increases in intracellular calcium in fibroblasts stably transfected with the α 1D -AR ( Figure 4). Therefore, BMY 7378 at 30 nM, the concentration used in the vascular

Figure 3 Effect of Phenylephrine on Intracellular Calcium Levels in Human Aortic Smooth Muscle Cells.
Human aortic smooth muscle cells were loaded with Calcium Green 1 AM for 1 hr. The ability of 25 uM phenylephrine to increase intracellular calcium was studied alone and following treatment with either 1 nM prazosin or 30 nM BMY 7378. Experiments were carried out as described in Methods. The results of a typical imaging study are presented along with a graphical summary of the statistical analysis of four independent experiments. Data were analyzed by a one-way ANOVA followed by post hoc testing. * Indicates a statistically significant difference from the untreated control. Effect of Phenylephrine on Intracellular Calcium Levels in Stably Transfected Rat 1 Fibroblasts. Rat 1 Fibroblasts, stably transfected with each of the α 1 -ARs, were loaded with Calcium Green 1 AM for 1 hr. The ability of 10 uM phenylephrine to increase intracellular calcium was studied alone and following treatment with either 1 nM prazosin or 30 nM BMY 7378. Experiments were carried out as described in Methods. The results of a typical imaging study are presented along with a graphical summary of the statistical analysis of four independent experiments. Data were analyzed by a one-way ANOVA followed by post hoc testing. * Indicates a statistically significant difference from the untreated control. smooth muscle cells (see above), can selectively block the α 1D -AR. These data support the conclusion that the receptor that mediates increases in intracellular calcium in aortic smooth muscle cells is the α 1D -AR.

Effects on reactive oxygen species
This type of specificity of coupling was also observed using a novel functional response to activation of the α 1 -AR-namely the generation of reaction oxygen species. In human aortic smooth muscle cells, phenylephrine produced a rapid, dose-dependent and statistically significant increase in the level reactive oxygen species ( Figure 5). While we present the data with 10 uM, we could see statistically significant increases in ROS at 1 uM phenylephrine. This increase was blocked by 1 nM prazosin or 30 nM BMY 7378. Therefore, it is the α 1D -AR that mediates increases in reactive oxygen species in these vascular smooth muscle cells.

Effects on smooth muscle contraction
Recent data from heterologous systems have suggested that the interaction between the α 1D -AR and other G-protein coupled receptors alters the pharmacologic properties of the α 1D -AR [17,19]. Our results from vascular smooth muscle cells that naturally express all three receptors indicate that functional responses from a receptor with α 1D -AR characteristics can be detected.
To assess the relevance of the interaction between the α 1D -AR and the other ARs in an intact blood vessel system, we studied contractile responses in the rat aorta. In previous work we have shown that the contractions of the rat aorta are mediated by the α 1D -AR [22]. In addition to potential dimerization among the α 1 -ARs, there is evidence of heterodimerization between the α 1D -AR and β 2 -AR. Studies in heterologous systems also show that desensitization of the β 2 -AR with albuterol promotes the internalization and desensitization of the α 1D -AR [17]. We assessed responses in the rat aorta following a 12 hr exposure to albuterol. After this incubation period, the responses to albuterol were significantly decreased when compared to vehicle treated aorta ( Figure 6). This indicates a desensitization of the β 2 -AR mediated response. The phenylephrine log dose response curves were the same in control and albuterol desensitized aorta. Therefore, desensitization of β 2 -AR does not lead to desensitization of the α 1D -AR.

Discussion
Previous work from our laboratory has shown that in heterologous systems the α 1D -AR localizes intracellularly and does not undergo agonist-mediated internalization or desensitization [11,12,21]. Due in part to problems with transfection, it has been difficult to determine if this type of expression pattern occurs in cells that natively express the α 1D -AR along with the other α 1 -AR family members. To facilitate its efficient expression, we developed an adenoviral vector expressing the human α 1D -AR fused with the GFP. We then used the vector to infect human aortic smooth muscle cells. RT-PCR analysis showed that these cells express all three α 1 -ARs ( Figure 2). The rank order of mRNA expression was α 1B -AR> α 1A -AR>>α 1D -AR. This type of expression pattern is typical for the α 1 -ARs. Following infection of aortic smooth muscle cells we observed that the α 1D -AR was localized to intracellular compartments ( Figure 1A). An intracellular localization pattern was also observed when aortic smooth muscle cells were immunostained with an α 1D -AR antibody (Figure 1B). Therefore, while dimerization between the α 1D -AR and the other α 1 -ARs may occur in aortic smooth muscle cells, this does not alter the cellular localization of the α 1D -AR.
If the α 1D -AR forms heterodimers with the other α 1 -ARs, then it is possible that these complexes exhibit properties different from the α 1D -AR alone [17,19]. We examined this possibility using the selective antagonist BMY 7378. In previous work we calculated that at 30 nM over 90 % of the α 1D -ARs would be occupied by BMY 7378 while less than 10 % of either the α 1A -AR or the α 1B -AR would be

Figure 5 Effect of Phenylephrine on the Levels of Reactive Oxygen Species Levels in Human Aortic Smooth
Muscle Cells. Human aortic smooth muscle cells were loaded with Mitotracker ROS for 20 min. The ability of 10 uM phenylephrine to increase the levels of reactive oxygen species was studied alone and following treatment with either 1 nM prazosin or 30 nM BMY 7378. Experiments were carried out as described in Methods. The results of a typical imaging study are presented along with a graphical summary of the statistical analysis of four independent experiments. Data were analyzed by a one-way ANOVA followed by post hoc testing. * Indicates a statistically significant difference from the untreated control.
occupied by this antagonist [22]. Therefore, at this concentration BMY 7378 would be anticipated to be highly selective for the α 1D -AR. This was substantiated in fibroblasts stably transfected with each of the α 1 -ARs. In these fibroblast cell lines phenylephrine treatment produced a significant increase in intracellular calcium. However, only in fibroblasts expressing the α 1D -AR was BMY 7378 capable of antagonizing the calcium response to phenylephrine, indicating that BMY 7378 is selective for the α 1D -AR (Figure 4). The phenylephrine-mediated increases in intracellular calcium in aortic smooth muscle cells were also blocked by this dose of BMY 7378 (Figure 3). In a similar fashion, we demonstrated that the generation of reactive oxygen species in aortic smooth muscle cells was antagonized by 30 nM BMY 7378 ( Figure 5). There is no evidence that BMY 7378 at this concentration can block either the α 1A -or the α 1B -AR. Therefore the antagonism seen with BMY 7378 indicates that the observed increases in intracellular calcium and reactive oxygen species are mediated by a receptor of α 1D -AR character. In aggregate, the data suggest that while heterodimerization may occur, it does not appear to alter the pharmacologic properties of the α 1D -AR. We know that both increases in intracellular calcium and elevations in ROS are mediated by the α 1D -AR. What we do not know is if the intracellularly expressed α 1D -AR is signaling competent and responsible for these effects or whether it is a small population of cell surface expressed receptors. If signaling does indeed emanate from the intracellular α 1D -AR, then there has to be a pathway that would allow agonist access to these receptors.
In addition to dimerization within the α 1 -AR family, there is also evidence that the α 1D -AR can form dimers with the β 2 -AR. Studies in expression systems have shown that not only does the β 2 -AR promote the cell surface expression of the α 1D -AR but that desensitization of the β 2 -AR also desensitized the α 1D -AR response. We wished to determine if this type of activity could be obtained in a functional system that natively expresses these receptors without resorting to overexpression of cloned receptors in a model cell system. The contractile responses of phenylephrine in the rat aorta are due to interactions at the α 1D -AR (see for example, Piascik et al. [22]). Therefore, we assessed potential β 2 -AR/α 1D -AR interactions using this blood vessel. Overnight treatment of blood vessels with albuterol caused desensitization of the β 2 -AR as shown by diminished vasodilatory responses to this agent ( Figure  6A). However, desensitization of the β 2 -AR did not cause a rightward shift of the phenylephrine dose response curve (see Figure 6B). Therefore, desensitization of the β 2 -AR does not alter contractile responses to the α 1D -AR. These results show that cross desensitization between β 2 -AR and the α 1D -AR does not occur in an intact, responding segment of vascular smooth muscle.

Conclusion
In summary, adenoviral vectors expressing the α 1 -ARs are a novel and efficient tool to investigate properties of these receptors in native cells. These vectors were used to show that in human smooth muscle cells expressing all three α 1 -ARs, the α 1D -AR is localized in intracellular compartments. Therefore, despite recent reports of heterodimerization between the α 1 -ARs, in a human vascular smooth muscle cell line, the α 1D -AR is still expressed intracellularly. Indeed none of the data we obtained in this work support the idea of α 1D -AR heterodimerization. Using three independent measures (calcium levels, generation Effect of the Albuterol Pretreatment of Phenylephrine-Induced Contraction of the Rat Aorta of reactive oxygen or vascular smooth muscle contraction) we also could not detect any evidence of altered pharmacologic properties of the α 1D -AR.

Cell culture conditions
Human aortic smooth muscle cells were obtained from Cascade Biologics (Portland, OR) and grown in Medium 231 supplemented with smooth muscle growth supplement until they become confluent (Cascade Biologics, Portland, OR). Stably transfected Rat 1 fibroblast lines expressing each of the α 1 -AR subtypes were maintained in Dulbecco's modified Eagle's medium (Cellgro, Herdon, VA) supplemented with 10% fetal bovine serum and a 1% antibiotic/antimycotic cocktail (Invitrogen, Carlsbad, CA). All cells were grown in T75 flasks in a 37°C cell culture incubator with a humidified atmosphere (95% air and 5% CO 2 ) and were fed every 2 to 3 days. After reaching confluence the cells were plated on plain untreated coverslips in 35 mm tissue culture dishes.

Construction of recombinant adenoviruses expressing the α 1 -ARs
A vector expressing the human α 1D -AR coupled to the green fluorescent protein (α 1D -AR/GFP) was provided by Dr. Gozoh Tsujimoto [23,24]. This vector was digested with EcoRI and XbaI enzymes and cloned into the pCI expression vector (Promega. Madison, WI). pCI was digested with BglII and ClaI. This fragment included the CMV I.E. promoter, the α 1D -AR/GFP and the SV40 late poly (A). The BglII and ClaI fragment was cloned into the adenovirus recombination vector pAdLink. pAdLink and the wild type adenovirus vector, dl327, were linearized with NheI and ClaI respectively. Homologous recombination occurred by co-transfecting linearized pAdLink and the wild type adenovirus vector into HEK 293 cells. Positive plaques appeared 10 to 14 days after recombination and were then amplified. Plaques were purified by serial dilutions of a positive plaque (usually from 10 -3 to 10 -12 ) in 96 well plates using HEK 293 cells. After plaque purification, samples of viral DNA were analyzed for wild type virus contamination by PCR [25]. Once a purified adenovirus was obtained, the plaque was amplified for largescale production. Fifty 150 mm dishes of HEK 293 cells were used for amplification of adenovirus which was then purified using double cesium gradients. Adenovirus was tittered using the Adeno-X™ Rapid Titer Kit from BD Biosciences (Palo Alto, CA).

Infection of cells with recombinant adenovirus
Human aortic smooth muscle cells or Rat 1 fibroblasts were grown on glass coverslips. Two hours prior to infection, cells were placed in serum free medium and infected with adenovirus. Twenty four hours after infection the medium was changed and the virus free incubation was allowed to proceed for an additional 24 hours. Cells expressing the GFP labeled α 1D -AR were fixed with 3.7% Formaldehyde in PBS for 10 mins. Cells were then mounted on slides with Vectashield (Vector Labs, Burlingame, CA). Cells were visualized using a Leica TCS SP 5 AOBS confocal microscope with a Plan-Apo 64X oil immersion objective lens (Leica, Wetzlar, Germany) using Leica TCS NT version 2.5 software. Images were transferred to a computer for reduction with Adobe Photoshop version 6.0 (Adobe Systems, Mountain View, CA).

Immunocytochemistry
Human aortic smooth muscle cells or grown on glass cover slips, were washed in PBS and fixed with 3.7% Formaldehyde in PBS for 10 min. Cells were then washed with .05% BSA in PBS and permeabilized with 0.1% Triton in PBS for 5 min. After permeabilization, the cells were washed and blocked with 10% lamb serum for 1 hour at room temperature. After washing polyclonal antibodies (Affinity Bioreagents, Golden, CO) against each of the α 1 -ARs, diluted 1:100 in 1% BSA in PBS, was added and incubated overnight at 4°C. Following this incubation, the cells were washed with .05% BSA in PBS and a Texas Red secondary antibody (Abcam, Cambridge MA), diluted 1:500 in PBS, was added and incubated in the dark at room temperature for 1 hr. Cells were washed with PBS and mounted on glass slides with Vectashield (Vector Laboratories, Burlingame, CA). Cells were visualized with a confocal microscope as described above.

RT-PCR
Total RNA from HASMCs was isolated and purified with the ChargeSwitch Total RNA kit from Invitrogen (Carlsbad, CA), and 0.5 μg samples were reverse transcribed at 45°C for one hour using Cloned Avian Myeloblastosis Virus (AMV) Reverse Transcriptase and Oligo(dT) 20 primer (Invitrogen, Carlsbad, CA). After heating at 85°C for 5 min. to terminate the reaction, cDNA samples were stored at -20°C until used. Negative controls for the presence of genomic DNA were performed by replacing the reverse transcriptase enzyme with Taq DNA polymerase.