Introduction

Top Abstract Introduction Methods Results Discussion References

Physiologically, alpha-2 adrenoceptors couple primarily to inhibitory responses. However, there has been a growing evidence of coupling to stimulatory responses such as platelet aggregation, smooth muscle contraction and secretion (reviewed in Ruffolo et al., 1993; Hieble et al., 1995). The inhibitory responses are usually manifested at the cellular level both in endogenous and heterologous expression systems as an inhibition of adenylyl cyclase or Ca⁺⁺ channel activity through G_i/o-type G proteins (reviewed in Regan and Cotecchia, 1992). Coupling to both inhibitory G_i and stimulatory G_s proteins has been indicated in recombinant CHO cells expressing all the human alpha-2 adrenoceptor subtypes (alpha-2A, alpha-2B, alpha-2C) (Fraser et al., 1989; Eason et al., 1992), S115 cells expressing alpha-2B (Jansson et al., 1994), JEG-3 cells expressing alpha-2A and alpha-2B (Pepperl and Regan, 1993) and Sf9 cells expressing mouse alpha-2B (Näsman et al., 1997); similar coupling to pertussis toxin-sensitive inhibition and -insensitive stimulation of Ca⁺⁺ currents is implicated in recombinant PC-12 cells expressing rat alpha-2D (analogous to human alpha-2A) and alpha-2B (Soini et al., 1997). In several studies, alpha-2 receptors have also been observed to couple to Ca⁺⁺ elevation. In some instances this has been related to stimulation of cation channels (Aburto et al., 1993; Lepretre and Mironneau, 1994; Musgrave and Seifert, 1995) but also a coupling to Ca⁺⁺ mobilization from the intracellular stores has been shown for alpha-A/D receptors in HEL human erythroleukemia cells (Michel et al., 1989; Dorn et al., 1997), rat cerebral astrocytes (Enkvist et al., 1996) as well as in CHO and COS-7 cells (Dorn et al., 1997). In HEL cells (Åkerman et al., 1996; Dorn et al., 1997) and in rat cerebral astrocytes (Enkvist et al., 1996) also inositol-1,4,5-trisphosphate (IP₃) elevations in response to alpha-2 adrenergic stimulation have been recorded. Pertussis toxin-sensitivity of this response has lead to a suggestion that this response would be mediated by -subunits from G_i/o-type G proteins. Recently, scavenging of -subunits has been shown to inhibit alpha-2A-induced Ca⁺⁺ elevation in COS-7 cells (Dorn et al., 1997).

The current evidence of coupling of alpha-2 adrenoceptors to Ca⁺⁺ elevation is thus based on alpha-2A adrenoceptors in endogenous and heterologous expression systems. Therefore we have in this study aimed to characterize the possible Ca⁺⁺ signals through the other human alpha-2 receptor subtypes. All three subtypes of alpha-2 adrenoceptors were for this purpose expressed separately in CHO cells at similar expression levels. As they were all shown to couple to Ca⁺⁺ elevations, their basic agonist pharmacological profiles were characterized; although the pharmacology of alpha-2 adrenoceptors with respect to cAMP inhibition has been documented in various expression systems, the corresponding pharmacology with respect to Ca⁺⁺ is largely unknown (see above).

	Methods

Top Abstract Introduction Methods Results Discussion References

Cell cultures. CHO cells (American Type Culture Collection, Rockville, MD), transfected as described in (Pohjanoksa et al., 1997), were grown in Minimum Essential Medium (MEM) alpha (Gibco, Paisley, UK) supplemented with 0.22% (w/v) NaHCO₃, 100 U/ml penicillin G (Sigma Chemical Co., St. Louis, MO), 80 U/ml streptomycin (Sigma) and 5% (v/v) fetal calf serum (Gibco) at 37°C in 5% CO₂ in an air ventilated humidified incubator in 260-ml plastic culture flasks (75 cm² bottom area; Nunc A/S, Roskilde, Denmark) or in plastic culture dishes ( 94 mm; Greiner GmbH, Frickenhausen, Germany).

Drugs. Clonidine (2-[2,6-dichloroaniline]-2-imidazoline HCl), EGTA, noradrenaline, oxymetazoline, pertussis toxin and probenecid were purchased from Sigma. Digitonin was purchased from Merck AG (Darmstadt, Germany) and fura-2 acetoxymethyl ester from Molecular Probes Inc. (Eugene, OR). UK14,304 was from RBI (Natick, MA), D-medetomidine from Orion-Corporation Orion-Pharma (Turku, Finland) and U-73122 and U-73343 from Calbiochem (La Jolla, CA).

Media. The TBM consisted of 137 mM NaCl, 5 mM KCl, 1 mM CaCl₂, 10 mM glucose, 1.2 mM MgCl₂, 0.44 mM KH₂PO₄, 4.2 mM NaHCO₃ and 20 mM TES adjusted to pH 7.4 with NaOH.

Ca⁺⁺ measurements. The fluorescent Ca⁺⁺-indicator fura-2 was used to monitor changes in intracellular [Ca⁺⁺] (Grynkiewicz et al., 1985). The cells were harvested using phosphate buffered saline containing 0.2 g/liter EDTA, spun down, and loaded at 37°C in MEM alpha supplemented with 0.02 g/liter bovine serum albumin, 1 mM probenecid and 4 µM fura-2 acetoxymethyl ester for 20 min. The cells were washed once with Ca⁺⁺-free TBM and stored on ice as pellets (medium removed). The measurement of intracellular free calcium was carried out as follows: one pellet was resuspended in TBM supplemented with 1 mM probenecid at 37°C and placed in a stirred quartz microcuvette in a thermostated cell-holder within a fluorescence spectrophotometer. Fluorescence was monitored either with a Hitachi F-2000 or a Hitachi F-4000 fluorescence spectrophotometer at the wavelengths 340 nm (excitation), 505 nm (emission) or with a PTI QuantaMaster fluorescence spectrophotometer at the wavelengths 340/360/380 nm (excitation), 505 nm (emission). The experiments were calibrated using 60 µg/ml digitonin, which gives the maximum value of fluorescence (F_max) and 10 mM EGTA, which gives the minimum value of fluorescence (F_min). The free Ca⁺⁺-concentration was calculated from the fluorescence at 340 nm, 505 nm (F) using the equation in which the extracellular fura-2 fluorescence is subtracted from the F values.

Data analysis. Student's two-tailed t test was used in all the calculations of the significance. For the results, mean ± S.E.M. is given unless specifically indicated. The nonlinear least square curve fitting was performed using SigmaPlot for Windows 4.00 (Jandel Scientific, Corte Madera, CA).

	Results

Top Abstract Introduction Methods Results Discussion References

Human alpha-2 adrenoceptor subtypes coupled to elevation of [Ca⁺⁺]_i when heterologously expressed in CHO cells (fig. 1). This signal consisted of both a rapid and transient elevation and a more sustained elevation as also has been shown in HEL cells that express endogenous alpha-2A adrenoceptors (Kukkonen et al., 1997). In wild-type (untransfected) CHO cells no Ca⁺⁺ elevations were observed in response to adrenergic stimulation (fig. 2). To be able to perform pharmacological comparison of the different subtypes, clones alpha-2A-E47, alpha-2B-6 and alpha-2C-L3 (Pohjanoksa et al., 1997) were chosen as they expressed the receptors at approximately similar levels (1.88 ± 0.40, 2.40 ± 0.65 and 2.04 ± 0.42 pmol/mg, respectively, Pohjanoksa et al., 1997). No significant differences were seen between different subtypes with respect to the basal Ca⁺⁺ level or the maximum elevation induced by noradrenaline, although the maximum Ca⁺⁺ elevation may be somewhat larger for alpha-2B (fig. 2).

View larger version (10K): [in this window] [in a new window]   Fig. 1.   A typical Ca++ elevation in response to noradrenaline stimulation of alpha-2B-expressing CHO cells (alpha-2B-6). Similar responses were obtained in other alpha-2 adrenoceptor subtype-expressing cells.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   The [Ca++]i level under unstimulated conditions (basal) and after stimulation with 10 µM noradrenaline. There is no significant difference between any basal value or between any stimulated value for the recombinant adrenoceptor-expressing CHO cells (alpha-2A, alpha-2B, alpha-2C). In the case of the wild-type (untransfected) CHO cells, noradrenaline did not induce any Ca++ elevation. Number of batches of cells where the determinations have been done is 6 to 15.

Ca⁺⁺ elevations were sensitive to the phospholipase C inhibitor (Bleasdale et al., 1990) U-73122 (12 µM, 10 min) whereas the less active analogue U-73343 (12 µM, 10 min) was without an effect (fig. 3). The noncomplete block may reflect the affinity of U-73122 that could not be used in higher concentration due to its low solubility. Almost complete inhibition was also obtained by pertussis toxin pretreatment (100 ng/ml, 24 hr) for all the subtypes although a somewhat larger remaining signal could still be resolved in alpha-2B expressing cells (fig. 3). These results together suggest that the Ca⁺⁺ response is due to phospholipase C- activation by G_i/o. The minor signal remaining after the pertussis toxin-pretreatment may reflect a coupling of alpha-2 receptors to pertussis toxin-insensitive G proteins.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   The effect of the phospholipase C inhibitor U-73122 (12 µM; 10 min), the inactive analogue U-73343 (12 µM; 10 min) and pertussis toxin-pretreatment (100 ng/ml; 24 hr) on the noradrenaline-induced Ca++ elevation. The significancies are calculated with respect to the control noradrenaline signal (ctrl). Number of measurements per bar is three to nine. ns, Not significant  (P > .05); ** P < .01; *** P < .001.

The agonist pharmacology of different alpha-2 subtypes was further characterized. Noradrenaline always appeared to be a strong agonist with a similar potency (EC₅₀) for all the subtypes, whereas the response to the other agonists varied largely among the subtypes (table 1; fig. 4). UK14,304 was a full agonist (as compared to noradrenaline) both for alpha-2A and alpha-2C receptors although it had a much lower potency for alpha-2C, whereas oxymetazoline was a full agonist (as compared to noradrenaline) only for alpha-2B (table 1; fig. 4). D-Medetomidine had a similar potency for all the subtypes and it was an essentially full agonist (as compared to noradrenaline) for alpha-2B and alpha-2C (table 1; fig. 4). Clonidine was never a full agonist and it displayed a particularly small response in alpha-2B-expressing cells (table 1; fig. 4). To control the recombinant alpha-2 adrenoceptor-specificity of the Ca⁺⁺ responses, the agonist responses were also tested in the wild-type CHO cells. Noradrenaline, UK 14,304, D-medetomidine and clonidine were without an effect, whereas oxymetazoline slightly elevated [Ca⁺⁺]_i ([Ca⁺⁺]_max = 7.4 ± 2.3 nM, number of batches of cells = 7). This Ca⁺⁺ elevation is too small to cause any error in the agonist potency measurements in the recombinant CHO cells.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   The dose-response curves for different agonists in all the alpha-2 adrenoceptor subtypes. The results are from at least three to four different batches of cells where the experiments were performed at least in triplicate.

	Discussion

Top Abstract Introduction Methods Results Discussion References

Our results show that all the human alpha-2 receptor subtypes couple to Ca⁺⁺ elevation when heterologously expressed in CHO cells at comparable levels. Both the maximum response and the EC₅₀ of noradrenaline, the physiological ligand, were similar in all the subtypes suggesting that there should not be any major physiological differences in the coupling of the different alpha-2 subtypes to Ca⁺⁺ elevations. The Ca⁺⁺ response is suggested to be mainly due to phospholipase C- activation by G_i/o with a minor contribution from pertussis toxin-insensitive G proteins.

In the case of agonists there were marked difference both with respect to maximum Ca⁺⁺ elevations and EC₅₀ values. UK14,304 was a full agonist for both alpha-2A and alpha-2C but it had a 10-fold higher potency for alpha-2A. This is clearly a reflection of the 10-fold higher binding affinity of UK14,304 for alpha-2A [Marjamäki et al., 1993 (S115 cells)]. D-Medetomidine was a full agonist for alpha-2C but it also gave high responses with both alpha-2A and alpha-2B. Interestingly, it had the same potency for all the subtypes. Its EC₅₀, which was around 3 nM, was the lowest among all the tested agonists for any subtype and the same as its determined binding affinity for alpha-2 adrenoceptors in CHO cells (Pohjanoksa et al., 1997). Oxymetazoline was a full agonist only for alpha-2B, but it had a low potency for this subtype. On the contrary, it was a weak partial agonist for alpha-2A but had a markedly higher potency. This difference in potency seems to be a reflection of its 100-fold higher binding affinity for alpha-2A than for alpha-2B [Bylund et al., 1988 (HT-29 cells and human platelets); Lomasney et al., 1990 (COS-7 cells); Marjamäki et al., 1993 (S115 cells); Uhlen et al., 1994 (CHO cells)]. Clonidine was most active for alpha-2C but had a higher potency for alpha-2A; the latter may relate to its higher binding affinity for alpha-2A than for alpha-2C [Jansson et al., 1994 (S115 cells); Pohjanoksa et al., 1997 (CHO cells)]. Barely any signal could be observed in alpha-2B-expressing cells.

It has previously been shown that agonists differ in their ability to induce activation of G_i vs. G_s through alpha-2 receptors (Eason et al., 1994). Noradrenaline is a full agonist with respect to both G_i and G_s coupling (Eason et al., 1994), whereas imidazolines, like UK14,304, differ markedly in their ability to activate G_s (Eason et al., 1994). In our study, only a marginal Ca⁺⁺ elevation remained after pertussis toxin-pretreatment. Therefore coupling to G_s is not a significant pathway for Ca⁺⁺ elevation by alpha-2-adrenoceptors in CHO cells, in accordance with previous studies in alpha-2A-expressing HEL and CHO cells (Michel et al., 1989; Dorn et al., 1997) and cerebral astrocytes (Enkvist et al., 1996). Thus, G_i/o proteins are likely to be implicated in the transduction of both cAMP decreasing and Ca⁺⁺ elevating responses in CHO cells. Pertussis toxin-sensitive cAMP decrease is thought to be caused by inhibition of adenylyl cyclase via _i whereas pertussis toxin-sensitive Ca⁺⁺ elevation may occur through stimulation of phospholipase C- by subunits from G_i/o (Lee et al., 1993; Smrcka and Sternweis, 1993; Wu et al., 1993). This also appears to be the mechanism of alpha-2A induced Ca⁺⁺ elevation in COS-7 cells (Dorn et al., 1997).

It is of interest to compare the current Ca⁺⁺ results and the previously published results on the coupling of the alpha-2 adrenoceptor subtypes to cAMP inhibition (Pohjanoksa et al., 1997) in the same CHO cell clones as in the present study (alpha-2A-E47, alpha-2B-6, alpha-2C-L3). When the responses to the three agonists, noradrenaline, D-medetomidine and clonidine are compared with respect to cAMP inhibition and Ca⁺⁺ elevation, the results are largely similar in the alpha-2C expressing cells whereas marked differences are seen in the alpha-2A- and alpha-2B-expressing cells. First, D-medetomidine and clonidine seem to be full agonists with respect to cAMP inhibitionas compared to noradrenalinein alpha-2A-E47 and alpha-2B-6, whereas clonidine is less efficacious (40 and 16% of noradrenaline signal for alpha-2A-E47 and alpha-2B-6, respectively) and even D-medetomidine is less efficacious than noradrenaline (60 ± 3%) with respect to Ca⁺⁺ elevation in alpha-2A-E47. This could suggest there is a receptor reserve with respect to cAMP inhibition in alpha-2A- and alpha-2B-expressing CHO cells. However, what contradicts this is the fact that the EC₅₀ value for noradrenaline is of the same magnitude or even lower in all the clones with respect to Ca⁺⁺ elevation than to cAMP inhibition. When cAMP inhibition is examined in clones expressing lower numbers of alpha-2A (alpha-2A-E30) and alpha-2B receptors (alpha-2B-10), the relative responses to clonidine are with both subtypes significantly higher in the cAMP inhibition assay (72 and 71%, respectively) as compared to Ca⁺⁺ elevation assay (40 and 16%, respectively) in our study (P < .01 for both). Therefore, it seems that the coupling of alpha-2A and alpha-2B adrenoceptors to cAMP inhibition and Ca⁺⁺ elevation displays some agonist specific features at least with respect to the response to clonidine (as compared to noradrenaline).

Altogether, our results show that all three human alpha-2 adrenoceptor subtypes couple to Ca⁺⁺ elevation in CHO cells in a similar manner. This seems to occur through pertussis toxin-sensitive activation of phospholipase C. Different agonists display characteristic subtype-specific differences in potency and activity.