The α_1B/D-adrenoceptor knockout mouse permits isolation of the vascular α_1A-adrenoceptor and elucidates its relationship to the other subtypes

Abstract

Background and purpose:

Mesenteric and carotid arteries from the α_1B/D-adrenoceptor knockout (α_1B/D-KO) were employed to isolate α_1A-adrenoceptor pharmacology and location and to reveal these features in the wild-type (WT) mouse.

Experimental approach:

Functional pharmacology by wire myography and receptor localization by confocal microscopy, using the fluorescent α₁-adrenoceptor ligand BODIPY FL-Prazosin (QAPB), on mesenteric (an ‘α_1A-adrenoceptor’ tissue) and carotid (an ‘α_1D-adrenoceptor’ tissue) arteries.

Key results:

α_1B/D-KO mesenteric arteries showed straightforward α_1A-adrenoceptor agonist/antagonist pharmacology. WT had complex pharmacology with α_1A- and α_1D-adrenoceptor components. α_1B/D-KO had a larger α_1A-adrenoceptor response suggesting compensatory up-regulation: no increase in fluorescent ligand binding suggests up-regulation of signalling. α_1B/D-KO carotid arteries had low efficacy α_1A-adrenoceptor responses. WT had complex pharmacology consistent with co-activation of all three subtypes. Fluorescent binding had straightforward α_1A-adrenoceptor characteristics in both arteries of α_1B/D-KO. Fluorescent binding varied between cells in relative intracellular and surface distribution. Total fluorescence was reduced in the α_1B/D-KO due to fewer smooth muscle cells showing fluorescent binding. WT binding was greater and sensitive to α_1A- and α_1D-adrenoceptor antagonists.

Conclusions and implications:

The straightforward pharmacology and fluorescent binding in the α_1B/D-KO was used to interpret the properties of the α_1A-adrenoceptor in the WT. Reduced total fluorescence in α_1B/D-KO arteries, despite a clear difference in the functionally dominant subtype, indicates that measurement of receptor protein is unlikely to correlate with function. Fewer cells bound QAPB in the α_1B/D-KO suggesting different cellular phenotypes of α_1A-adrenoceptor exist. The α_1B/D-KO provides robust assays for the α_1A-adrenoceptor and takes us closer to understanding multi-receptor subtype interactions.

Introduction

Three genetically defined α₁-adrenoceptor subtypes (α_1A, α_1B, α_1D; nomenclature follows Alexander et al., 2008) can produce similar cellular responses, including contraction of smooth muscle (Guimaraes and Moura, 2001). Previous pharmacological work on α₁-adrenoceptor subtypes suggests that multiple subtypes are present in arterial preparations, causing the analysis of one subtype to be contaminated by the presence of the others (Daly et al., 2002; Deighan et al., 2005).

Little is known of the functional pharmacology of the three subtypes due to the similarity of their signalling, their widespread co-expression and the incomplete availability of selective pharmacological agents (Alexander et al., 2008). This explains the conflicting pA₂ or pK_B estimates calculated by different groups using various assays for what is ostensibly the same subtype.

Little is known also of the location of the subtypes in native tissues. In cell culture, using epitope-tagged recombinant receptors, it has been proposed that through heterodimerization, the α₁-adrenoceptor subtypes might influence each other's cellular location and functional capabilities (Uberti et al., 2003; Stanasila et al., 2003; Hague et al., 2004a,b;). Through such approaches it has also been proposed that these receptor–receptor interactions might result in a different cellular distribution for each subtype. This relates to the concept that the α_1B-adrenoceptor is located mainly on the cell surface, while the α_1A-adrenoceptor and α_1D-adrenoceptor are mainly located at intracellular sites (Hirasawa et al., 1997; Hrometz et al., 1999; McCune et al., 2000; Sugawara et al., 2002). This is not a universal finding as, even in cultured cells, all three subtypes have been shown to be located both on the surface and at intracellular locations (Daly et al., 1998; Hrometz et al., 1999; Pediani et al., 2000; Morris et al., 2004).

There is, however, an alternative approach to studying α₁-adrenoceptor location, using the fluorescent ligand BODIPY FL-Prazosin (QAPB). This ligand has affinity for all three α₁-adrenoceptor subtypes (Daly et al., 1998; Mackenzie et al., 2000; Daly and McGrath, 2003; McGrath et al., 2005). Because the α_1A-adrenoceptor spontaneously internalizes, when QAPB binds it is taken inside the cell and labels the receptor population in intracellular organelles; the receptors continuously recycle so that equilibrium is established between extracellular free ligand and the receptor population (Pediani et al., 2005). This process can be blocked by subtype-selective competitor antagonists to identify subtypes in particular locations (Mackenzie et al., 2000) and enables experiments in native tissue that are normally possible only in cell culture.

Normally in native tissues the presence of more than one receptor subtype complicates both pharmacological analysis and localization of receptors. However, mouse strains are available in which genetic knockouts (KO) of each α₁-adrenoceptor are present [α_1A-KO (Rokosh and Simpson, 2002); α_1B-KO (Cavalli et al., 1997); α_1D-KO (Tanoue et al., 2002b; Turnbull et al., 2003)]. We have therefore pursued a strategy of isolating the α_1A-adrenoceptor in the double KO of the other subtypes (α_1B/D-adrenoceptor knockout; α_1B/D-KO). We employed a small artery, mesenteric first order, in which the α_1A-adrenoceptor is the major mediator of α₁-adrenoceptor-mediated vasoconstriction (Daly et al., 2002; Martinez-Salas et al., 2007) and where the other subtypes are expected to be modulatory (Daly et al., 2002). A conductance artery, the carotid, was also examined to study a system in which the α_1D-adrenoceptor is the major mediator of α₁-adrenoceptor contractility but where an additional functional α_1A-adrenoceptor may exist (Deighan et al., 2005).

The aim of this study was to use the α_1B/D-KO to allow the study in native tissue of the remaining subtype, the α_1A-adrenoceptor, in respect of its pharmacological properties and cellular distribution and whether these change when it is isolated from the other α₁-adrenoceptor subtypes. We subsequently used this knowledge to try to understand the greater complexity that occurs when the other subtypes are present in the wild-type (WT) and to elucidate the possible interactions between subtypes.

Methods

Animals used and vessel preparation

All animal care and this investigation conformed to the provisions of the UK Animals (Scientific procedures) Act 1986. WT mice of C57BL6J/129Sv mixed background were used as controls for single knockouts of the α_1B-adrenoceptor (α_1B-KO), α_1D-adrenoceptor (α_1D-KO) and the α_1B/D-KO. Mice were bred at the University of Glasgow from breeding pairs kindly provided by Professor Susanna Cotecchia (University of Lausanne, Lausanne, Switzerland; α_1B-KO) and Professor Gozoh Tsujimoto (National Children's Medical Research Center, Tokyo, Japan; α_1D-KO). α_1B/D-KO were generated by cross-breeding the homozygous single KO mice at the University of Glasgow. The generation and background of the KO have previously been described in detail [α_1B-KO (Cavalli et al., 1997); α_1D-KO (Tanoue et al., 2002b); α_1B/D-KO (Hosoda et al., 2005)].

Mice were male weighing between 26 and 55 g [WT (26–47 g; n= 74); α_1B-KO (27–48 g; n= 12); α_1D-KO (26–42 g; n= 12); α_1BD-KO (26–55 g; n= 80)]. An earlier study showed that the predominant subtype present in hepatocytes could change between the commonly employed ages of 2–3 months (Deighan et al., 2004), therefore the more mature 4–5-month-old mice were used. Mice were maintained on a 12:12 hour light/dark schedule at 22–25°C with 45–65% humidity and fed ad libitum on a standard rodent diet and provided drinking water. Mice were killed by carbon dioxide asphyxiation, and first order mesenteric and carotid arteries were isolated using a dissection microscope.

Experimental protocols for functional pharmacology

Two millimetre sections of artery were mounted in physiological salt solution on a four-chamber wire myograph (Danish Myotechnology, Aarhus, Denmark). Following a 30 min equilibration period, passive tension of 250 mg was applied to both types of arteries, and vessels were allowed to equilibrate for a further 45 min. Data were recorded by using Powerlab and Chart (version 5.0) (ADInstruments, Chalgrove, UK). Vessels were challenged with 10 µM of one of three agonists [the standard α₁-adrenoceptor agonist, phenylephrine; an agonist selective for the α_1A-adrenoceptor, A-61603 (N-[5-(4,5-dihydro-1H-imidazol-2-yl)-2-hydroxy-5,6,7,8-tetrahydronaphthalen-1-yl]methanesulphonamide) (Knepper et al., 1995); or 5-hydroxytryptamine]. Acetylcholine (3 µM) was used to test the viability of the endothelium following preconstriction.

Cumulative concentration–response curves were produced for each agonist in half-log increments from 1 nM to 30 µM or 100 µM. Subtype-selective antagonists were incubated for 30 min: prazosin 1 nM, 10 nM and 100 nM (non-selective α₁-adrenoceptor antagonist); 5-methylurapidil 10 nM and 100 nM [α_1A-adrenoceptor-selective antagonist (Gross et al., 1988)]; RS100-329 (N-[(2-trifluoroethoxy)phenyl],N′-(3-thyminylpropyl) piperazine hydrochloride) 1 nM, 10 nM and 100 nM [α_1A-adrenoceptor-selective antagonist (Williams et al., 1999)]; BMY 7378 (8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione) 1 nM, 10 nM and 100 nM [α_1D-adrenoceptor-selective antagonist (Goetz et al., 1995; Saussy et al., 1996)]; or rauwolscine 10 nM (non-selective α₂-adrenoceptor antagonist).

Experimental protocols for visualization of smooth muscle cell α₁-adrenoceptors

QAPB incubations

3–5 millimetre segments of mesenteric and carotid arteries were incubated with QAPB (100 nM) for 120 min. Live unfixed tissue was used to remain as faithful as possible to the in vitro state used to assess functional pharmacology. All incubations were performed at room temperature (21°C), and solutions were replaced every 30 min to maintain pH conditions.

Additional experiments using non-fluorescent antagonists to compete for QAPB binding sites were performed in segments from the same animals as QAPB controls (see Figure S1). Based on the timescale reported by Pediani et al. (2005), segments were incubated with QAPB for 60 min to establish an equilibrium, then co-incubated for a further 60 min with QAPB and an antagonist at 100 nM: prazosin, rauwolscine, RS100-329 or BMY 7378. The concentration of each antagonist was selected to be capable of competing effectively with QAPB at a specific α₁-adrenoceptor subtype but not at other potential binding sites based on pK_i estimates from the literature at the cloned α₁-adrenoceptor subtypes (Table 1). For instance, at α_1A-adrenoceptors, RS100-329 has higher affinity than QAPB, but at the α_1B-adrenoceptor or α_1D-adrenoceptor, its affinity is lower. Thus, used at the same concentration as QAPB (100 nM), RS100-329 should be capable of displacing QAPB from α_1A-adrenoceptors but not from other binding sites.

Table 1.

Ligand binding at α₁-adrenoceptors (AR) (mean pK_i for antagonists at cloned receptors)

Antagonist		α1A-AR	α1B-AR	α1D-AR	References
Prazosin	Mean	9.6	9.8	9.7	1–3
	Range	9.0–9.9	9.0–10.3	9.0–10.0
5-Methylurapidil	Mean	8.7	7.0	7.4	1, 4–10
	Range	8.3–9.2	6.1–7.4	6.1–7.9
RS100-329	Mean	9.5	7.6	7.9	11–12
	Range	9.3–9.6	7.3–7.8	7.9–7.9
BMY 7378	Mean	6.4	6.6	8.5	1, 4, 8, 12–17
	Range	6.0–7.1	6.2–7.5	8.1–9.3
QAPB		8.7	8.4	8.1	1

1 MacKenzie et al. (2000); 2 Knepper et al. (1995); 3 Michel and Goepel (1998); 4 Saussy et al. (1996); 5 Goetz et al. (1994); 6 Leonardi et al. (1997); 7 Garcia-Sainz et al. (1995); 8 Yoshio et al. (2001); 9 Buckner et al. (1996); 10 Meyer et al. (1996); 11 Williams et al. (1999); 12 Deighan et al. (2004); 13 Goetz et al. (1995); 14 Kenny et al. (1995); 15 Testa et al. (1997); 16 Hieble et al. (1995); 17 Piascik et al. (1997).

BMY 7378, 8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione; QAPB, BODIPY FL-Prazosin; RS100-329, N-[(2-trifluoroethoxy)phenyl],N′-(3-thyminylpropyl) piperazine hydrochloride.

Slide preparation and image capture

Arteries were mounted in the incubation solutions with a coverslip (No. 1.5) on top. The thin-walled mesenteric artery was mounted intact, while the carotid artery was sliced open with a single-edged razor blade and laid flat (Miquel et al., 2005) to optimize visualization through the luminal side of the thick arterial wall.

Optical sections were collected on a Bio-Rad Radiance 2100 Confocal Laser Scanning System at an excitation/emission of 488/515 nm for QAPB. Arteries were visualized by using ×40 oil immersion objective (NA 1.0). Optimal laser, gain and offset (contrast and brightness) were determined and then kept constant for each artery. Photomultiplier sensitivities were set such that all pixels fell within the range 0 (black) to 255 (white). Carotid arteries were imaged at high power (zoom 8) to include only smooth muscle cells and avoid areas containing elastic lamina in the media. Mesenteric artery images were collected at low power (zoom 3), which allows coverage of a wider area of exclusively smooth muscle cells. Arteries were scanned as 0.5 µm slices from the internal elastic lamina through the media producing Z-series in stacks of approximately 20–30 µm in depth. Each procedure was carried out in triplicate on four different mice per strain.

Data analysis and statistical procedures

All graphical and statistical analysis was performed by using Graph Pad Prism (version 5.0). Mean data were compared across mouse strains and also in the presence and absence of the selected antagonists using one-way anova and Bonferroni's post test or Student's t-test.

Functional pharmacology

Data were expressed as a percentage of the control curve maximum and represents the mean value ± standard error of the mean. When a maximum response was achieved, pEC₅₀ values were calculated for control curves and agonist curves in the presence of antagonists.

Quantitative confocal microscopy

Image analysis was based on the 2D image analysis as described by Miquel et al. (2005) and performed by using Metamorph software (version 4). In cell cultures fluorescent ligand binding follows similar kinetics to radioligand binding (Mackenzie et al., 2000). However, the relationship between receptor number and fluorescence is not quantitatively established. Therefore, the measurement of ligand binding can be interpreted to identify which receptor subtypes are present but cannot be interpreted in terms of precise numbers of receptors. A single plane at a depth of 4–6 µm from the internal elastic lamina of each Z-series that captured only smooth muscle cells was selected for analysis. Average pixel intensity was measured for the entire plane (mesenteric artery 96 × 96 µm; carotid artery 36 × 36 µm). In addition, individual smooth muscle cells from each image were outlined as a Region of Interest to measure average pixel intensity (see Figure S2). It should be noted that this analysis was performed only on smooth muscle cells that showed fluorescent binding. The average pixel intensity correlated to a greyscale range of 0 (black) to 255 (white), with intermediate intensities being assigned an appropriate grey level.

Materials

Physiological salt solution composition (in mM): 119 NaCl, 4.7 KCl, 1.2 MgSO₄.H₂O, 1.2 KH₂PO₄, 24.9 NaHCO₃, 2.5 CaCl₂ and 11.1 glucose.

All analytical grade drugs were dissolved and diluted to the required concentration in distilled water (unless otherwise stated). Drugs supplied by Sigma Aldrich (Poole, UK) were phenylephrine, acetylcholine, 5-methylurapidil, prazosin, 5-hydroxytryptamine, BMY 7378; by Tocris(Bristol, UK) were A-61603, RS100-329, rauwolscine; and by Invitrogen (Paisley, UK) was QAPB (dissolved in DMSO and diluted in freshly gassed physiological salt solution).

Results

Functional pharmacology in mesenteric arterial rings

In the WT and all three KO strains, the maximum response and sensitivity to 5-hydroxytryptamine was similar (Table 2). This eliminates generic structural or smooth muscle sensitivity changes as a basis for differences subsequently found using drugs acting via α₁-adrenoceptors.

Table 2.

Agonist responses in mouse mesenteric arteries

Mouse	Phenylephrine		A-61603		5-Hydroxytryptamine
	Maximum response (g)	pEC50	Maximum response (g)	pEC50	Maximum response (g)	pEC50
WT	0.52 ± 0.06	6.2 ± 0.08	0.41 ± 0.05	8.3 ± 0.11***	0.18 ± 0.04	6.9 ± 0.18
α1B-KO	0.42 ± 0.07	5.8 ± 0.14	–	–	0.17 ± 0.04	6.9 ± 0.16
α1D-KO	0.48 ± 0.05	5.9 ± 0.09	–	–	0.17 ± 0.02	6.8 ± 0.12
α1B/D-KO	0.49 ± 0.08	6.4 ± 0.17	0.66 ± 0.04**	8.0 ± 0.09***	0.19 ± 0.02	7.0 ± 0.13

Maximum responses and pEC₅₀ values are expressed as mean ± SEM.

A-61603, N-[5-(4,5-dihydro-1H-imidazol-2-yl)-2-hydroxy-5,6,7,8-tetrahydronaphthalen-1-yl]methanesulphonamide; α_1B/D-KO, α_1B/D-adrenoceptor knockout mouse; WT, wild-type.

^**P < 0.01 compared with WT (one-way anova, Bonferroni's post test).

^***P < 0.001 compared with phenylephrine-induced response (one-way anova, Bonferroni's post test).

Cumulative concentration–response curves to the non-selective α₁-adrenoceptor agonist phenylephrine were obtained in all mouse strains [Figure 1A (data not shown for α_1B-KO and α_1D-KO)], with no significant differences between the WT and KO mice (Table 2).

Figure 1.

Contractile responses to (A) phenylephrine (n= 6) and (B) A-61603 (N-[5-(4,5-dihydro-1H-imidazol-2-yl)-2-hydroxy-5,6,7,8-tetrahydronaphthalen-1-yl]methanesulphonamide) (n= 7) in mesenteric arteries of wild-type (WT) and α_1B/D-adrenoceptor knockout (BDKO) mice. Data points are expressed in g tension as mean ± SEM.

The more selective agonist A-61603 was used to study the α_1A-adrenoceptor response. Concentration-dependent contractions were obtained to A-61603 in the WT and α_1B/D-KO (Figure 1B). The maximum response was significantly higher in the α_1B/D-KO than in the WT (Table 2). In both the WT and α_1B/D-KO, sensitivity to A-61603 was significantly higher than to phenylephrine (Table 2) consistent with the α_1A-adrenoceptor being the main functional α₁-adrenoceptor in the mesenteric artery.

In both the WT and α_1B/D-KO tissues, prazosin competitively antagonized the responses to phenylephrine (Figure 2A,B). The high pA₂ of 9.6 calculated for the α_1B/D-KO (Table 3) is consistent with α₁-adrenoceptor pharmacology at recombinant α_1A-adrenoceptor (Table 1). However, in the WT arteries, prazosin showed only moderate potency as indicated by the pK_B of 8.8 (Table 3). Prazosin was also tested against A-61603 in the α_1B/D-KO (Table 3), producing a pK_B indicative of an α₁-adrenoceptor-mediated response.

Figure 2.

Effects of prazosin (A) and (B), and BMY 7378 (8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione) (C) and (D) on contractile response to phenylephrine in mesenteric arteries of the wild-type (WT) (n= 6) and α_1B/D-adrenoceptor knockout (α_1B/D-KO) mice (n= 7). Data points are expressed as percentage of control curve maximum as mean ± SEM.

Table 3.

Potency values of antagonists of responses to phenylephrine and A-61603 in mouse mesenteric arteries

Antagonist	Phenylephrine				A-61603
	WT		α1B/D-KO		WT	α1B/D-KO
	pA2/pKB	Slope (95% CI)	pA2/pKB	Slope (95% CI)	pKB	pKB
Prazosin	8.8	–	9.6	1.0 (0.8–1.2)	–	9.3
RS100-329	–	–	–	–	9.0	9.9
5-Methylurapidil	9.0	–	8.9	–	8.8	8.9
BMY 7378	8.3 (at 10 nM)	–	6.5	1.1 (0.8–1.4)	–	6.5
	7.0 (at 1 µM)	–

A-61603, N-[5-(4,5-dihydro-1H-imidazol-2-yl)-2-hydroxy-5,6,7,8-tetrahydronaphthalen-1-yl]methanesulphonamide; α_1B/D-KO, α_1B/D-adrenoceptor knockout mouse; BMY 7378, 8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione; RS100-329, N-[(2-trifluoroethoxy)phenyl],N′-(3-thyminylpropyl) piperazine hydrochloride; WT, wild-type.

The effect of the selective α_1D-adrenoceptor antagonist BMY 7378 was studied to examine the contractile role of the α_1D-adrenoceptor in the mesenteric artery. In WT arteries BMY 7378 caused parallel shifts of the phenylephrine concentration–response curve but the relationship to antagonist concentration was not clearly dose-related (Figure 2C). Thus, at the lowest BMY 7378 concentration (10 nM), the concentration–response curve was shifted significantly to the right (P < 0.05). An intermediate concentration (100 nM) had little further effect but the highest concentration (1 µM) produced a further significant rightward shift (P < 0.01). pK_B values were calculated for BMY 7378 at the lowest and highest concentrations (Table 3), and these were comparable with α_1D-adrenoceptor and α_1A-adrenoceptor values respectively (Table 1). In the α_1B/D-KO, BMY 7378, at high concentrations (300 nM, 1 µM and 3 µM), caused concentration-related, rightward shifts of the phenylephrine response curve (Figure 2D). In the tissues from this mouse strain, the pA₂ value (6.5) (Table 3) was identical to that obtained against the A-61603 response and indicates the potency of BMY 7378 at the α_1A-adrenoceptor.

5-Methylurapidil was employed as much of the existing literature uses this antagonist as being ‘α_1A-adrenoceptor-selective’. In WT arteries, 5-methylurapidil caused a parallel rightward shift of the phenylephrine response at 10 nM (Figure 3A), which produced a pK_B (Table 3) indicative of an α_1A-adrenoceptor-mediated response (Table 1). However, 100 nM shifted only the upper part of the curve, creating a biphasic concentration–response curve, suggestive of a component of the response being attributable to either α_1D-adrenoceptor or α_1B-adrenoceptor. Consequently, 5-methylurapidil (100 nM) was studied in the α_1B-KO, to reduce the possibilities to the α_1D-adrenoceptor and α_1A-adrenoceptor subtypes. In this strain, 5-methylurapidil (100 nM) also revealed a biphasic response, indicating that the additional response was α_1D-adrenoceptor-mediated (Figure 3B). The position was further clarified by using 5-methylurapidil in combination with BMY 7378: BMY 7378 eliminated the minor component. This was quantified by showing that, while the pEC₅₀ for phenylephrine in the presence of 5-methylurapidil (100 nM) was unaffected by BMY 7378 treatment (4.5 ± 0.1), the pEC₂₅, which was more representative of the first ‘phase’, was shifted significantly rightward by BMY 7378 from 6.4 ± 0.3 to 4.8 ± 0.1 (P < 0.001). The response after 5-methylurapidil plus BMY 7378 was thus monophasic (lacking the first phase) (Figure 3B). In the α_1B/D-KO, a monophasic response to phenylephrine was observed in the presence of 5-methylurapidil (100 nM) (Figure 3C). The phenylephrine concentration–response curve was shifted rightward in parallel by 5-methylurapidil, producing a pK_B value (Table 3) comparable to published values for α_1A-adrenoceptors (Table 1).

Figure 3.

Effects of 5-methylurapidil (5MeU) on contractile response to phenylephrine in mouse mesenteric arteries. (A) Antagonizing effect of 5-methylurapidil in wild-type (WT) (n= 6). (B) Antagonizing effect of 5-methylurapidil and elimination of biphasic response by BMY 7378 (8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione) in α_1B-adrenoceptor knockout (α_1B-KO) (n= 6). (C) Elimination of biphasic response in α_1B/D-KO (n= 6). Data points are expressed as percentage of control curve maximum as mean ± SEM.

As phenylephrine was found to be acting on more than one subtype, hence complicating assessment, antagonist potency for the α_1A-adrenoceptor was tested against a selective α_1A-adrenoceptor agonist, A-61603. Both 5-methylurapidil and a second α_1A-adrenoceptor antagonist, RS100-329, caused a concentration-related, parallel, rightward shift of the A-61603 response curve in both the WT and α_1B/D-KO arteries (Table 3). This verifies that A-61603 activated only the α_1A-adrenoceptor in mesenteric arteries from both the WT and α_1B/D-KO.

Functional pharmacology in carotid arterial rings

In the WT and all three KO strains the maximum response and sensitivity to 5-hydroxytryptamine was similar (Table 4). This eliminates generic structural or smooth muscle sensitivity changes as a basis for differences subsequently found using drugs acting via α₁-adrenoceptors.

Table 4.

Agonist responses in mouse carotid arteries

Mouse	Phenylephrine		A-61603		5-Hydroxytryptamine
	Maximum response (g)	pEC50	Maximum response (g)	pEC50	Maximum response (g)	pEC50
WT	0.27 ± 0.01	6.6 ± 0.05	0.28 ± 0.02	6.0 ± 0.06	0.27 ± 0.04	6.9 ± 0.08
α1B-KO	0.32 ± 0.01	6.9 ± 0.10	0.26 ± 0.02	6.3 ± 0.04*	0.27 ± 0.05	6.9 ± 0.08
α1D-KO	0.17 ± 0.01***	5.6 ± 0.06***	0.14 ± 0.01***	6.7 ± 0.08***	0.26 ± 0.06	6.8 ± 0.04
α1B/D-KO	0.06 ± 0.01***	4.9 ± 0.07***	0.13 ± 0.01***	6.6 ± 0.09***	0.26 ± 0.03	6.8 ± 0.07

Maximum responses and pEC₅₀ values are expressed as mean ± SEM.

A-61603, N-[5-(4,5-dihydro-1H-imidazol-2-yl)-2-hydroxy-5,6,7,8-tetrahydronaphthalen-1-yl]methanesulphonamide; α_IB-KO, α_1B-adrenoceptor knockout; α_ID-KO, α_1D-adrenoreceptor knockout; α_1B/D-KO, α_1B/D-adrenoceptor knockout; WT, wild-type.

^*P < 0.05;

^***P < 0.001 compared with WT (one-way anova, Bonferroni's post test).

Concentration-dependent contractions to phenylephrine were produced in all mouse strains (Figure 4A; Table 4). Maximum response and sensitivity to phenylephrine were significantly reduced in the α_1D-KO compared with the WT. This suggests that the WT response can be attributed predominantly to α_1D-adrenoceptors and the small response in the α_1B/D-KO suggests a minor contribution from α_1A-adrenoceptors. The effect of knocking out the α_1B-adrenoceptor differed according to the other receptors present. Its elimination from the WT (α_1B-KO) resulted in an increase in maximum response and sensitivity to phenylephrine indicating either an up-regulation of the α_1D- or α_1A-adrenoceptor-mediated response or loss of an inhibitory response. In contrast, elimination of the α_1B-adrenoceptors from the α_1D-adrenoceptor strain (α_1B/D-KO) resulted in a reduction in response. This suggests the existence of an α_1B-adrenoceptor-mediated contraction or a down-regulation of an α_1A-adrenoceptor-mediated response.

Figure 4.

Contractile responses to (A) phenylephrine (n= 8) and (B) A-61603 (N-[5-(4,5-dihydro-1H-imidazol-2-yl)-2-hydroxy-5,6,7,8-tetrahydronaphthalen-1-yl]methanesulphonamide) (n= 8) in carotid arteries of wild-type (WT), α_1B-adrenoceptor knockout (1B KO), α_1D-adrenoceptor knockout (1D KO) and α_1B/D-adrenoceptor knockout (BD KO) mice. Data points are expressed in g tension as mean ± SEM.

The α_1A-adrenoceptor agonist A-61603 induced concentration-related contractions in all mouse strains (Figure 4B; Table 4). The maximum responses to A-61603 in the α_1D-KO and α_1B/D-KO were significantly lower compared with the WT. pEC₅₀ values for the A-61603-induced response were significantly higher in the α_1B-KO, α_1D-KO and α_1B/D-KO than in the WT, although the pEC₅₀ value to A-61603 in the α_1B-KO was significantly lower than in the α_1D-KO and α_1B/D-KO (P < 0.001 and P < 0.05 respectively). This can be explained by the presence of the α_1D-adrenoceptors: there is an additional component at high concentrations of A-61603 that has the effect of reducing the apparent pEC₅₀ value; this is eliminated after knocking out the α_1D-adrenoceptors, revealing the true, higher, sensitivity of A-61603 at α_1A-adrenoceptors. Also, for this reason, sensitivity, but not maximum response, to A-61603 appears to be significantly lower than for phenylephrine in the WT (Table 4). In the α_1B/D-KO both maximum response and sensitivity to A-61603 were significantly higher than for phenylephrine, reflecting the true situation at α_1A-adrenoceptors, with a difference in pEC₅₀ values of 1.7, similar to the value of 1.6 in the mesenteric artery.

In the carotid arteries from WT mice, prazosin competitively antagonized the phenylephrine concentration–response curve (Figure 5A; Table 5). In the α_1B/D-KO the phenylephrine concentration–response curve was significantly shifted rightward by prazosin 10 nM, although a true maximum response was not achieved within the agonist concentration used (Figure 5B; Table 5).

Figure 5.

Effects of prazosin on contractile response to phenylephrine (A) and (B), and A-61603 (N-[5-(4,5-dihydro-1H-imidazol-2-yl)-2-hydroxy-5,6,7,8-tetrahydronaphthalen-1-yl]methanesulphonamide) (C) and (D) in carotid arteries of wild-type (WT) (n= 7) and α_1B/D-adrenoceptor knockout (α_1B/D-KO) (n= 6) mice. Data points are expressed as percentage of control curve maximum as mean ± SEM.

Table 5.

Potency values of antagonists of responses to phenylephrine and A-61603 in mouse carotid arteries

Antagonist	Phenylephrine				A-61603
	WT		α1B/D-KO		WT		α1B/D-KO
	pA2/pKB	Slope	pA2/pKB	Slope	pA2/pKB	Slope	pA2/pKB	Slope
Prazosin	9.6*	0.9 (0.8–1.1)	9.0 at 10 nM	–	9.6 at 10 nM	–	9.1 at 10 nM	–
BMY 7378	8.3*	–	–	–	8.3	–	No shift at 10 nM and 100 nM	–
5-Methylurapidil	7.5*	1.1 (0.7–1.5)	Response abolished	–	8.3	1.0 (0.6–1.0)	8.3 at 10 nM	–
RS100-329	7.9	–	–	–	8.7	–	Non-competitive block (at 10 nM)

A-61603, N-[5-(4,5-dihydro-1H-imidazol-2-yl)-2-hydroxy-5,6,7,8-tetrahydronaphthalen-1-yl]methanesulphonamide; α_1B/D-KO, α_1B/D-adrenoceptor knockout; BMY 7378, 8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione; RS100-329, N-[(2-trifluoroethoxy)phenyl],N′-(3-thyminylpropyl) piperazine hydrochloride; WT, wild-type.

^*Denotes data from Deighan et al. (2005).

In the WT, only 10 nM prazosin produced a parallel shift in the A-61603 curve, enabling a pK_B to be calculated (Figure 5C; Table 5). Prazosin (10 nM) also produced a parallel shift in the A-61603-induced response in the α_1B/D-KO (Figure 5D) producing a pK_B (9.1), comparable to published values for α_1A-adrenoceptors.

In the α_1B/D-KO, in the presence of the selective α_1A-adrenoceptor antagonist 5-methylurapidil the contractile response to phenylephrine was obliterated (Table 5). No further antagonist studies were performed against phenylephrine in the α_1B/D-KO due to the low potency of the agonist in this artery.

The potency estimates for 5-methylurapidil against A-61603 were similar in the α_1B/D-KO and WT (Figure S3; Table 5). Therefore, another selective α_1A-adrenoceptor antagonist, RS100-329, was also employed to investigate the role of the α_1A-adrenoceptor in the carotid artery. RS100-329 10 nM acted competitively in the WT (Figure 6A; Table 5) but blocked the A-61603-induced response in a non-competitive manner in the α_1B/D-KO (Figure 6B). The pK_B for 5-methylurapidil against A-61603 in the WT (8.3) was higher than that obtained against phenylephrine (7.5) and closer to published values for an α_1A-adrenoceptor-mediated response (average pK_i 8.7, Table 1).

Figure 6.

Effects of RS100-329 (N-[(2-trifluoroethoxy)phenyl],N′-(3-thyminylpropyl) piperazine hydrochloride) (A) and (B), BMY 7378 (8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione) (C) and (D), and rauwolscine (E), on contractile response to A-61603 (N-[5-(4,5-dihydro-1H-imidazol-2-yl)-2-hydroxy-5,6,7,8-tetrahydronaphthalen-1-yl]methanesulphonamide) in carotid arteries of wild-type (WT) (n= 7) and α_1B/D-adrenoceptor knockout (α_1B/D-KO) mice (n= 7). Data points are expressed as percentage of control curve maximum as mean ± SEM.

In the WT, a biphasic response was produced in the presence of BMY 7378 10 nM (Figure 6C), highlighting the involvement of two subtypes mediating the A-61603 response, as noted above. BMY 7378 100 nM produced a rightward displacement of the A-61603 concentration–response curve producing a pK_B (Figure 6C; Table 5) in line with published values for α_1D-adrenoceptors (average pK_i 8.5, Table 1). This indicates α_1D-adrenoceptor agonism by A-61603 when an efficient α_1D-adrenoceptor system is present. To assess its potency versus α_1A-adrenoceptors and, hence, assess its selectivity, BMY 7378 was tested against A-61603 in the α_1B/D-KO carotid arteries: BMY 7378 (10 nM and 100 nM) had no effect on the concentration–response curve to A-61603 (Figure 6D).

In the α_1B/D-KO, the α₂-adrenoceptor antagonist, rauwolscine did not produce a rightward shift of the A-61603 concentration–response curve (Figure 6E; Table 5), suggesting that this agonist-induced response was not α₂-adrenoceptor-mediated.

Visualization of α₁-adrenoceptors using the fluorescent ligand QAPB

Mesenteric artery

In both strains, smooth muscle cells were clearly defined throughout the medial layer although the amount of fluorescence varied between individual cells (Figure 7). In the WT, QAPB appeared to bind all smooth muscle cells, while in the α_1B/D-KO, some smooth muscle cells did not appear to show fluorescent ligand binding. When measured quantitatively, the intensity of fluorescence of the 96 × 96 µm image was significantly reduced in the α_1B/D-KO compared with the level in the WT (Table 6). However, in the α_1B/D-KO tissues, the intensity of the fluorescence of those smooth muscle cells that QAPB did bind was not significantly different to that of WT smooth muscle cells (Table 6). These data suggest that there was a reduction in the number of smooth muscle cells that bound QAPB in the α_1B/D-KO mesenteric artery. Binding was present both on the cell surface and intracellularly in punctate compartments, showing that the α_1A-adrenoceptor is non-uniformly distributed.

Figure 7.

Comparison of QAPB fluorescent ligand binding in mesenteric arteries of WT and α_1B/D-KO mice. Non-fluorescent antagonists (100 nM) were used to compete with the fluorescent ligand QAPB (100 nM) after QAPB equilibrium was established. All images are collected under identical conditions of laser intensity, photomultiplier gain (contrast) and offset (brightness). Imaging was performed at low power (zoom 3). An image size of 512 × 512 pixels produced a field size of 96 × 96 µm. Calibration bar indicates 10 µm. Each image is representative of those generated from four mice. α_1B/D-KO, α_1B/D-adrenoceptor knockout mouse; BMY 7378, 8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione; QAPB, BODIPY FL-Prazosin; RS100-329, N-[(2-trifluoroethoxy)phenyl],N′-(3-thyminylpropyl) piperazine hydrochloride; WT, wild-type.

Table 6.

Comparison of average pixel intensity of entire image and smooth muscle cells (SMC) in wild-type (WT) and α_1B/D-adrenoceptor knockout mouse (α_1B/D-KO)

Mouse strain	Mesenteric arteries		Carotid arteries
	Average pixel intensity of 96×96 µm image	Average pixel intensity of SMC	Average pixel intensity 36×36 µm image	Average pixel intensity of SMC
WT	79.9 ± 4.1	84.8 ± 2.6	41.2 ± 9.2	48.8 ± 2.5
α1B/D-KO	24.3 ± 2.3***	75.7 ± 3.7	20.6 ± 2.9*	50.4 ± 3.6

Data are expressed as mean ± SEM.

Mesenteric arteries imaged at low power (zoom 3); carotid arteries imaged at high power (zoom 8).

Only smooth muscle cells showing fluorescent ligand binding were outlined as a Region of Interest to measure average pixel intensity.

P > 0.05 for smooth muscle cells, α_1B/D-KO compared with WT (Student's t-test).

^*P < 0.05;

^***P < 0.001 compared with WT (Student's t-test).

Effective competition from the non-selective α₁-adrenoceptor antagonist prazosin but not from the non-selective α₂-adrenoceptor antagonist rauwolscine confirmed that the fluorescent ligand was binding to α₁-adrenoceptors and not α₂-adrenoceptors (Figure 7; Table 7).

Table 7.

Effect of subtype-selective antagonists on fluorescent ligand binding in mesenteric and carotid arteries

Antagonist	Average pixel intensity
	WT		α1B/D-KO
	Mesenteric arteries	Carotid arteries	Mesenteric arteries	Carotid arteries
QAPB control	79.9 ± 4.1	41.2 ± 9.2	24.3 ± 2.3	20.6 ± 2.9
Prazosin	13.3 ± 7.7***	5.9 ± 1.4***	7.4 ± 1.2***	4.7 ± 1.9**
Rauwolscine	74.4 ± 10.7	55.3 ± 9.3	23.1 ± 1.3	17.3 ± 2.8
BMY 7378	27.7 ± 3.1***	10.0 ± 0.6**	24.6 ± 1.9	16.8 ± 2.6
RS100-329	24.4 ± 1.2***	13.3 ± 2.9**	9.1 ± 1.3***	5.5 ± 0.6***

Data are expressed as mean ± SEM; mesenteric arteries imaged at low power; carotid arteries imaged at high power (zoom 8).

α_1B/D-KO, α_1B/D-adrenoceptor knockout; BMY 7378, 8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione; QAPB, BODIPY FL-Prazosin; RS100-329, N-[(2-trifluoroethoxy)phenyl],N′-(3-thyminylpropyl) piperazine hydrochloride; WT, wild-type.

^**P < 0.01;

^***P < 0.001 compared with QAPB control (one-way anova, Bonferroni's post test).

BMY 7378 was employed at a concentration that should selectively compete with QAPB for α_1D-adrenoceptor binding sites in the WT and to test its ability to compete with QAPB at the α_1A-adrenoceptor in the α_1B/D-KO. In the presence of BMY 7378 QAPB fluorescent binding was reduced in the WT but not in the α_1B/D-KO (Figure 7; Table 7).

The selective α_1A-adrenoceptor antagonist RS100-329 was used to compete with QAPB at a concentration that should be effective at the α_1A-adrenoceptor but not at other potential binding sites. In the presence of RS100-329, fluorescent ligand binding was reduced in WT and abolished in α_1B/D-KO (Figure 7; Table 7).

Carotid artery

Smooth muscle cells were clearly defined by QAPB binding throughout the media of the WT artery (see Figure S4) but fluorescence varied between individual cells (Figure 8). Quantitative analysis revealed that intensity of fluorescence of the 36 × 36 µm image was significantly lower in the α_1B/D-KO, compared with the WT artery (Table 6). QAPB binding was more heterogeneous in the α_1B/D-KO, with patches of cells virtually devoid of ligand binding. This is reflected by quantitative analysis, which showed that in smooth muscle cells showing fluorescent binding in the α_1B/D-KO, the intensity of the fluorescence was not significantly different to that in the WT artery (Table 6). This suggests that the intensity of fluorescence measured for the 36 × 36 µm image reflects a reduction in the number of smooth muscle cells showing binding in the α_1B/D-KO samples. In both strains, QAPB bound to both the cell surface and intracellular compartments with evidence of both perinuclear binding and punctate binding being observed intracellularly.

Figure 8.

Comparison of QAPB fluorescent ligand binding in carotid arteries of WT and α_1B/D-KO mice. Non-fluorescent antagonists (100 nM) were used to compete with QAPB (100 nM) after QAPB equilibrium was established. All images are collected under identical conditions of laser intensity, photomultiplier gain (contrast) and offset (brightness). Imaging was performed at high power (zoom 8) to include only smooth muscle cell and avoid elastic lamina in the media. An image size of 512 × 512 pixels produced a field size of 36 × 36 µm. Calibration bar indicates 10 µm. Each image is representative of those generated from four mice. α_1B/D-KO, α_1B/D-adrenoceptor knockout; BMY 7378, 8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione; QAPB, BODIPY FL-Prazosin; RS100-329, N-[(2-trifluoroethoxy)phenyl],N′-(3-thyminylpropyl) piperazine hydrochloride; WT, wild-type.

The use of prazosin and rauwolscine in carotid arteries confirmed that the fluorescent ligand was selective for α₁-adrenoceptors and was not binding α₂-adrenoceptors (Figure 8; Table 7).

In the WT the selective α_1D-adrenoceptor antagonist BMY 7378 markedly reduced QAPB binding (Figure 8; Table 7), demonstrating that the α_1D-adrenoceptor is present in the WT. In the α_1B/D-KO, the lack of effect of BMY 7378 on fluorescence showed that BMY 7378 did not compete with QAPB for α_1A-adrenoceptor binding sites (Figure 8; Table 7).

Using the selective α_1A-adrenoceptor antagonist, RS100-329, to compete with QAPB showed reduced binding in the WT indicating that the α_1A-adrenoceptor is present in this strain (Figure 8; Table 7). In the α_1B/D-KO, QAPB binding was quantitatively reduced and could not be detected by visual inspection in the presence of RS100-329 (Figure 8; Table 7).

Discussion

The α_1B/D-KO strain shows ‘pure’α_1A-adrenoceptor pharmacology and elucidates the complex pharmacology of the WT. Potencies of agonists and antagonists in the WT did not correlate with any single subtype as shown in pure assay systems. By working back from the α_1B/D-KO, the subtypes in the WT tissues could be identified. α_1A-adrenoceptor-mediated contraction was identified in both arteries but was greater in the mesenteric artery. Both arteries showed α_1D-adrenoceptor-mediated contraction but this was greater in the carotid. There was indirect evidence for contraction by α_1B-adrenoceptors only in the carotid, but, in both arteries, removal of α_1B-adrenoceptors from the WT enhanced α_1D- or α_1A-adrenoceptor-mediated contractions.

The tissues from the α_1B/D-KO provide the first genuine example of α_1A-adrenoceptor pharmacology in native tissues. In particular, the α_1B/D-KO mesenteric artery represents a more accurate assay for functional pharmacology of the α_1A-adrenoceptor than is otherwise available and can be used to interpret data from other species in which KO are impractical. Deviations from radioligand binding estimates are now shown to be consistent with ‘contamination’ of the assay vessels with other α₁-adrenoceptor subtypes. Similarly, supposed ‘species’ or ‘tissue-specific’ differences may be due to varying proportions of different subtypes.

The only evidence for genetic compensation was enhancement of α_1A-adrenoceptor-mediated contraction in the mesenteric artery: the maximum response of A-61603 increased when moving from the WT to α_1B/D-KO. This must mean some increase in the efficiency of α_1A-adrenoceptors, whether by up-regulation of receptor number or signalling system. This would have been missed by studying only phenylephrine, because loss of its potent α_1D-adrenoceptor response negates the gain in α_1A-adrenoceptor response. Imaging data did not provide evidence for an up-regulation of α_1A-adrenoceptors in the α_1B/D-KO mesenteric artery. This suggests that ‘compensation’ lies in the signalling process. This is specific to the α_1A-adrenoceptor as the 5-hydroxytryptamine response was unaffected by the loss of the α_1B- and α_1D-adrenoceptors.

Pharmacology of α_1A-adrenoceptors

In the α_1B/D-KO, pharmacological analysis provided a clear profile for the α_1A-adrenoceptor. In both arteries, A-61603 had greater efficacy and potency than phenylephrine: the potency ratio was similar (40-fold). This differential is less than that reported by Pediani et al. (2000) for recombinant bovine α_1A-adrenoceptors (70-fold) or for WT tissues by Knepper et al. (1995) (rat vas deferens, 300-fold; canine prostate, 165-fold) but was closer to radioligand binding at bovine α_1A-adrenoceptors [60-fold; (Pediani et al., 2000)]. Both agonists had lower potency in the carotid (approximately 50-fold) indicating a smaller receptor reserve for α_1A-adrenoceptors.

In the carotid, the data from A-61603 clarify the phenylephrine data. Because responses to A-61603 are similar in the α_1D-KO and α_1B/D-KO this confirms that the reduction in the phenylephrine response is caused by the loss of an α_1B-adrenoceptor-mediated contraction to phenylephrine, which A-61603 does not activate. Conversely, the similarity of the increases in response to A-61603 and phenylephrine in the α_1B-KO suggests that this change is not due to the loss of an inhibitory effect of the agonists via α_1B-adrenoceptors (as A-61603 is not a good agonist at α_1B-adrenoceptors) but rather is a compensatory effect on the signalling process or receptor number for α_1D- or α_1A-adrenoceptors.

The functional pharmacology was straightforward in mesenteric arteries where α_1A-adrenoceptor is dominant in the WT strain; the major gain in accuracy was for the α_1D-adrenoceptor antagonist BMY 7378, whose estimated affinity fell to its α_1A-adrenoceptor value in the α_1B/D-KO. In contrast, the selective α_1A-adrenoceptor antagonists, RS100-329 and 5-methylurapidil, retained high potency, as expected in the α_1B/D-KO mice. A complication in the analysis of the α_1B/D-KO mesenteric arteries was that prazosin was less potent in the WT than in the α_1B/D-KO tissues. This was unexpected and, although this cannot be fully explained as yet, it reflects the complexity of analysis in the WT. It is interesting that the ‘α_1L-adrenoceptor’ phenomenon, that is, low affinity for prazosin (Flavahan and Vanhoutte, 1986; Muramatsu et al., 1990) has been reported to be lost in the α_1A-KO (Muramatsu et al., 2008), yet isolation of the α_1A-adrenoceptor (in the present study) does not reveal it. Thus the ‘α_1L-adrenoceptor’ phenomenon may require the presence of both the α_1A-adrenoceptor and another subtype.

In carotid arteries, where the residual α_1A-adrenoceptor response to agonists is small, the response was not sufficiently robust to sustain a comprehensive analysis. Threshold sensitivities to antagonists were as expected but were difficult to quantify in competitive terms. In addition, RS100-329 produced non-competitive blockade. Non-competitive blockade of recombinant α_1A-adrenoceptors has been shown for prazosin and WB4101 and was attributed to the non-equilibrium nature of post-receptor events (Pediani et al., 2000). Similar factors may apply here, where the maximum response achievable is small, relative to the capability of the artery, as shown by 5-hydroxytryptamine.

Agonist activity at α_1D-adrenoceptors by A-61603 was revealed in the KO. In the carotid, this occurred at high agonist concentrations as introducing the α_1D-KO reduced the responses to A-61603. Furthermore, the A-61603-induced response has an α_1D-adrenoceptor component in the WT, as there was a biphasic response to A-61603 in the presence of BMY 7378 10 nM and the potency estimate for BMY 7378 (100 nM) was comparable to the α_1D-adrenoceptor values (Table 1). This did not occur in the mesenteric artery where the receptor reserve for α_1D-adrenoceptor is smaller, consistent with A-61603 being a partial agonist at α_1D-adrenoceptor. This detailed pharmacological information refines the ability to assess the presence of the subtypes in non-genetically modified species.

Location of α_1A-adrenoceptors

When α_1A-adrenoceptors were isolated in the α_1B/D-KO they showed, in general, a more heterogeneous distribution than when the whole family of subtypes was present in the WT. This was most apparent in the carotid, where the α_1A-adrenoceptor is secondary to the α_1D-adrenoceptor. This suggests a general phenomenon of different phenotypes of vascular smooth muscle in respect of α_1A-adrenoceptors. This may be relevant to function, for example, of pacemaker cells (Peng et al., 2001) and partly explains the reduced maximum response in the α_1B/D-KO carotid, in which there is less likely to be a functional syncytium due to the separation of smooth muscle cells by collagen layers: consequently, the maximum response becomes dependent on the number of cells expressing the receptors, in contrast to a functional syncytium such as in the mesenteric artery where depolarization can spread through the tissue despite a limited number of cells expressing receptors.

In both arteries, receptors were found both on the cell surface and in intracellular compartments, reflecting findings in cell cultures with recombinant α_1D- (Daly et al., 1998) and α_1A-adrenoceptors (Pediani et al., 2005). This seems to be a genuine phenotype and not an artefact. Vascular fibroblasts in intact arteries also have a substantial population of intracellular α₁-adrenoceptors (McGrath et al., 2005), so the phenomenon is a general one for native and recombinant α₁-adrenoceptors, expressed in different cell types. In addition, fixed and permeabilized tissues also show these different cellular phenotypes (Miquel et al., 2005) eliminating differential ligand uptake as an explanation. There is no simple explanation based on different proportions of the subtypes in these different phenotypes or chaperoning by heterodimers. Nevertheless, increased heterogeneity in both types of artery, once the other subtypes were eliminated, shows a strong tendency for the α_1A-adrenoceptor subtype to be heterogeneously distributed.

Potential interaction of α₁-adrenoceptor subtypes

This is the first example of a native tissue in which it has been possible to observe the consequences, for receptor distribution, of moving from the WT strain with multiple subtypes of α₁-adrenoceptor to a double KO expressing only one subtype. There was, however, no evidence that removing the α_1B-adrenoceptor had any influence on the balance of surface to intracellular distribution of the remaining subtypes, as had been inferred from artificial systems (Stanasila et al., 2003; Hague et al., 2004b).

Physiological and therapeutic relevance

Our demonstration of the likely presence of all three subtypes in native vascular smooth muscle cells is extremely interesting for physiological function. The relative preponderance of the subtypes, as shown by fluorescent ligand binding, was similar between the two arteries despite the quantitatively different contribution of the subtypes to contraction. Thus, measuring the presence of the receptor or its RNA will not give a useful guide to function and, perhaps, the relative contribution of each subtype is more effectively regulated by other factors than by the amount of receptor present. This correlates also with functional compensation of the α_1A-adrenoceptor for the loss of the other two subtypes, which showed as an increase in response without evidence for an up-regulation of the receptors present.

As binding and function both show multiple subtypes in these two arteries, it is not beneficial to assign one α₁-adrenoceptor subtype to a specific vessel or to a class of vessel that performs a particular physiological function. α₁-Adrenoceptor heterogeneity is an important consideration when assessing the selectivity of α₁-adrenoceptor compounds. Blockade of either α_1A-adrenoceptor or α_1D-adrenoceptor can affect the response and, thus, can be expected to block adrenergic control of peripheral resistance, as has been demonstrated in vivo (Villalobos-Molina and Ibarra, 1996; Zhou and Vargas, 1996) and in vitro (Castillo et al., 1998; Zacharia et al., 2004; 2005;). This indicates that no one subtype plays a part in maintaining and regulating blood pressure as suggested by the mild hypotension in each single KO model (Rokosh and Simpson, 2002; Tanoue et al., 2002a).

Glossary

Abbreviations:

A-61603

N-[5-(4,5-dihydro-1H-imidazol-2-yl)-2-hydroxy-5,6,7,8-tetrahydronaphthalen-1-yl]methanesulphonamide

α_1B/D-KO

α_1B/D-adrenoceptor knockout

BMY 7378

8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione

KO

Knockouts

QAPB

BODIPY FL-Prazosin

RS100-329

N-[(2-trifluoroethoxy)phenyl],N′-(3-thyminylpropyl) piperazine hydrochloride

SMC

smooth muscle cells

WT

wildtype

Conflict of interest

None.

Supporting information

Additional Supporting Information may be found in the online version of this article:

Figure S1 Example of abolishment of BODIPY FL-Prazosin (QAPB) binding by prazosin (left image) but presence of smooth muscle cells confirmed by the nuclear stain Syto 61 (right image) in carotid artery of wild-type (WT). Imaging was performed at high power (zoom 8) to include only smooth muscle cells and avoid elastic lamina in the media. An image size of 512 × 512 pixels produced a field size of 36 μm × 36 μm. Calibration bar indicates 5 μm.

Figure S2 Illustration of measurement of average pixel intensity from individual smooth muscle cells from wild-type (WT) and α_1B/D-adrenoceptor knockout (α_1B/D-KO) mesenteric and carotid arteries. Smooth muscle cells were outlined and measured as Regions of Interest. Imaging was performed at low power (zoom 3) for mesenteric arteries and high power (zoom 8) for carotid arteries. An image size of 512 × 512 pixels produced a field size of 96 × 96 μm for mesenteric arteries and 36 μm × 36 μm for carotid arteries.

Figure S3 Effect of 5-methylurapidil (5MeU) on contractile response to phenylephrine in mouse carotid arteries of α_1B/D-adrenoceptor knockout (n = 6). Data points are expressed as percentage of control curve maximum as mean ± S.E.M.

Figure S4 Example images of BODIPY FL-Prazosin (QAPB) fluorescent binding on a plane imaged from the endothelial side (left image; average pixel intensity 78.3) and a plane imaged from the adventitial side (right image; average pixel intensity 77.7) from the same z-stack of a mesenteric artery. An image size of 512 × 512 pixels produced a field size of 96 μm × 96 μm.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.