PDFlib PLOP: PDF Linearization, Optimization, Protection

Page inserted by evaluation versionwww.pdflib.com – sales@pdflib.com

[3H]Cirazoline as a Tool for the Characterization of

Imidazoline Sites I. ANGEL,a M. LE ROUZIC, C. PIMOULE,

D. GRAHAM, AND S. ARBILLA Synthilabo Recherche 10, rue des Carrikres

BP 248 92504 Rueil-Malmaison Cedex, France

Compounds chemically related to the imidazoline structure such as idazoxan, cloni- dine, or bromoxidine were reported to possess high affinity for az-adrenoceptors. Recent studies have demonstrated that such compounds also bind with high affinity to nonadrenergic sites. These studies were carried out in a variety of tissues and by using [3H]idazoxan, [3H]clonidine, or ['H]p-aminoclonidine (["]PAC) as radioli- gands.lS2

The heterogeneity of a2-adrenergic receptors (a2-AR) was confirmed following the identification of three structurally distinct human a2-AR subtypes, so-called a2C2-, a2C4-, and azC lo-AR for their chromosomal lo~alization.~ From the different pharmacological behavior of guanidinium-imidazoline derivatives on a2-AR com- pared to imidazoline sites and from differences in their distribution and biochemical properties it was suggested that these sites represent distinct physical and pharmaco- logical

Wikberg and mi in ' demonstrated that cirazoline, an al-adrenoceptor agonist, is among the ligands with the highest affinity for ['Hlidazoxan sites, its affinity for nonadrenergic sites being 500-fold higher than that of a I-adrenoceptors and between 50- and 200-fold higher than that of [3H]-adrenoceptors. In a previous preliminary study, we identified [3H]cirazoline as a selective ligand that labels imidazoline sites in rat and kidney membranes.8

In the present study we further characterize the binding of [3H]cirazoline to rat brain and kidney as well as to beta cell HIT T15 membrane homogenates. Furthermore, to confirm the differences between the a2-AR and the imidazoline recognition site on a molecular level, we compared the specific binding of [3H]cirazoline to these tissues to q - A R binding, using [3H]rauwolscine in Chinese hamster ovary (CHO) cell lines stably transfected with either genomic or cDNA encoding the human ~ z - A R .

'To whom correspondence should be addressed.

112

ANGEL et ul.: [3HJCIRAZOLKNJ3 113

MATERIALS AND METHODS

Materials

Amiloride, benextramine, clonidine, (-)adrenaline, phentolamine, prazosin, pro- pranolol, and yohimbine were purchased from Sigma (Paris, France) and rauwolscine from Laboratoires Sarget (Mirignac, France). [3H]rauwolscine (75 -80 Ci.mmo1-') was from NEN, France. Bromoxidine, idazoxan, cirazoline, and [3H]cirazoline (25 Ci.mmo1-') were synthesized in the chemistry department of Synthilabo Recherche (Bagneux, France).

Membrane Homogenate Preparations

Chinese hamster ovary (courtesy of Drs. Lefkowitz and Caron, Duke University, Durham, North Carolina) and HIT T15 (ATCC) cells were grown in HAM-F12 (L- glutamine) and RPMI 1640 (L-glutamine) medium, respectively, with fetal calf serum lo%, penicillin (50 unitslml), streptomycin (50 pg/ml), 95% air, and 5% CO, at 37°C. Dexamethasone (0.1 pM) was added 18-24 hours before CHO-cell confluency. Membrane preparations from cell lines or whole brain and kidney (from Sprague- Dawley rats, C. River, France, 200-250 g) were performed and stored at -8O"C, respectively, in a hypotonic 2 mM Tris/HCl and isotonic 50 mM Tris/HCl buffer (in mM:EDTA 3, MgCI, 3.5, pH 7.4,4"C). Membranes were obtained using a polytron homogenizer and rinsed twice at 46,500 g (20 min, 4°C). Brain and kidney pellets were resuspended in isotonic buffer (incubation medium) and incubated 30 minutes at 25°C to remove endogenous inhibitors. Before use, thawed CHO and HIT T15 cell membranes were washed and rehomogenized in incubation buffer. Proteins were determined using the BIORAD method.

Brain or kidney membranes (500 pg/ml) and HIT T15 or CHO homogenates (400 pg/ml) were incubated in the presence of [3H]cirazoline in 250 p1 of incubation medium (30 min, 25°C). Chinese hamster ovary cell membranes (40 pg/ml) were incubated in the presence of [3H]rauwolscine in 1 ml of incubation buffer (60 min, 25°C). [3H]Rauwolscine and [3H]cirazoline binding assays were stopped by vacuum filtration over Whatman GF/B and GF/C filters, respectively.

Filters were washed with 3 x 5 ml of isotonic buffer (4°C). Nonspecific binding of [3H]rauwolscine was defined by 10 pM phentolamine and of [3H]cirazoline by 10 FM of unlabeled cirazoline or 1 pM of bromoxidine.

Competition studies were carried out under standard incubation conditions with 5 nM of [3H]cirazoline and a range of concentrations of different drugs (1 pM to 1 mM). At 5 nM [3H]cirazoline the nonspecific binding did not exceed 25 and 30% of total binding for kidney and brain homogenates, respectively. Specific binding represented approximately 0.09% of the total radioligand concentration (n = 20).

Association studies were performed by incubating the membranes (500 pg Prod ml) with 1 nM [3H]cirazoline at 25°C for periods ranging from 10 seconds to 120 minutes. Reversibility was assessed by adding an excess of bromoxidine (10 pM) after equilibrium was reached.

114 ANNALS NEW YORK ACADEMY OF SCIENCES

0 Total Binding A Specific Binding I3 Non-specific Binding

0.40

0.35

0.30

0.25

: 0.20 0.15

0.10

0.05

0.00 0.0 0.1 0.2 0.3 0.4 0.5

B Wn)

0 5 10 15 20 25 30 F (nM)

FIGURE 1. (Left) Saturation analysis of [3H]cirazoline binding (1 pM to 50 nM) to brain membrane homogenates: (0) total [3H]cirazoline binding, (A) specific binding, and (0) nonspe- cific binding (in the presence of 10 pM of unlabeled bromoxidine). (Right) Scatchard analysis plot of ['H]cirazoline specific binding, indicating two classes of sites.

Data Analysis

Results are expressed as the mean ? SEM of n experiments performed in triplicate. B,,,, K,, and K, parameters were derived from nonlinear regression analysis using the LIGAND program.' Pseudo Hill coefficients (n,) were deduced from the slope of the Hill plot of inhibition data ( r > 0.92). Appropriate single or two-site model fitting was obtained using the differential F test.

RESULTS

Characterization of [3H]Cirazoline Binding in Brain and Kidney Homogenates

Specific binding studies have shown that ['H]cirazoline, in a range of concentra- tions from 1 pM to 50 nM, binds to brain and kidney membrane homogenates in a reversible manner (to be discussed). Nonspecific binding of [3HJcirazoline was a linear function of the radioligand concentration (FIGS. 1 and 2). Saturation analysis revealed two classes of high affinity binding sites for each hornogenate preparation QJ c 0.001 ; n = 4). Binding parameters of rat brain and kidney homogenate membranes were not significantly different. For higher affinity binding sites, the dissociation constant, Kd, of [3H]cirazoline on brain membranes was 0.38 +. 0.05 nM, and the maximal binding density, B,,,, was 46.15 2 1.69 fmoVmg protein. Likewise, using kidney homogenates, the K, was 0.39 ? 0.06 nM and the B,,, was 30.53 t 1.94 fmoVmg protein. The maximal binding capacity of the low affinity binding sites was 457 2 30 fmoVmg protein and 426 2 29 fmoVmg protein using rat brain and kidney

ANGEL ef al. : [3H]CIRAZOLINE 115

0.4,

0.07 0 Total binding

A Specific binding

0 Nonspecific binding

k 0.03

0.01

0 5 10 15 20 F (nM)

FIGURE 2. (Left) Saturation anal sis of [3H]cirazoline binding (1 pM to 50 nM) to kidney

cific binding (in the presence of 10 pM of unlabeled bromoxidine). (Right) Scatchard analysis plot of [3H]cirazoline specific binding, indicating two classes of sites to kidney membrane.

membrane homogenates: (0) total [ Y Hlcirazoline binding, (A) specific binding, and (0) nonspe-

membranes, respectively. The & of [3H]cirazoline was 5.61 ? 0.67 nM using brain homogenates and 9.14 ? 0.79 nM using kidney membranes (TABLE 1).

To further characterize the binding parameters of [3H]cirazoline, kinetic studies of association and dissociation were carried out in rat brain and kidney membrane homogenates. In both tissues, association of [3H]cirazoline was rapid, with equilibrium reached after 5-6 minutes (FIGS. 3 and 4). Analysis of association constants reveals that the association is biphasic. Similar constants of association rates were obtained for brain and kidney membranes (TABLE 3). Upon the addition of bromoxidine (10 FM), a rapid, biphasic dissociation of bound [3H]cirazoline was observed (FIGS. 3 and 4, TABLE 3).

Competition studies of the lower affinity site of [3H]cirazoline binding to kidney and brain homogenates were performed with alkaloid, phenylethylamine, imidazoline, and guanidinium derivatives (TABLE 2). These studies were carried out at 5 nM of

TABLE 1. Binding Parameters of [3H]Cirazoline in Rat Brain and Kidney Homogenates

Brain Kidney

Higher Lower Higher , Lower Binding Parameter Affinity Affinity Affinity Affinity

B,, (fmoYmg protein) 46.15 2 1.69 457 2 30 30.53 +. 1.94 426 2 29 Kd (nM) 0.38 ? 0.05 5.61 ? 0.67 0.39 2 0.06 9.14 ? 0.79

116 ANNALS NEW YORK ACADEMY OF SCIENCES

20

I

c -

H 9 m c -

4

20

0.5

0

-l”--- -2

-3 0 1 2 3 4

Time (minutes)

0 2 4 6 8 10 12 14 16 18 20 Time (minutes)

FIGURE 3. Time course of association and dissociation of [‘Hlcirazoline binding to rat brain homogenates. Association of [3H]cirazoline (1 nM) was performed at 25°C at the indicated time points. Dissociation at equilibrium was initiated with 10 pM of bromoxidine (UK 14304). (0,U) association of total binding, (A, W) association of nonspecific binding, and (0) dissocia- tion of specific binding.

2.0,

o & , 0 2 4 6 8 10 12 14 16 18 20

Time (minutes)

FIGURE 4. Time course of association and dissociation of [3H]cira~oline binding to rat kidney homogenates. Association of [3H]cirazoline ( 1 nM) was performed at 25°C at the indicated time points. Dissociation at equilibrium was initiated with 10 pM of hromoxidine (UK 14304). (0.0) association of total binding, (A, W) association of nonspecific binding, and (0) dissocia- tion of specific binding to rat kidney homogenates. Experimental condition as in FIGIJRE 3.

ANGEL et af. : ['HICIRAZOLINE 117

TABLE 2. Inhibition of [3H]Cirazoline Binding by Drugs"

Y (PM) Ligands Brain Kidney

Cirazoline 0.004 t 0.001 0.002 2 0.004 Clonidine 0.050 2 0.006 0.012 t 0.004 Bromoxidine 0.011 t 0.002 0.012 2 0.001 Idazoxan 0.158 ? 0.044 0.302 t 0.004 Amilo ride 0.214 ? 0.024 0.105 t 0.031

Guanabenz 1.97 ? 0.45 1.97 t 0.88 SL 84.0418 1.97 2 0.71 2.84 t 1.46 SL 86.0715 1.66 t 0.76 6.92 2 0.98 Tolazoline 4.70 ? 1.10 14.21 t 1.98 SL 86.0714 34.70 ? 6.52 13.18 t 7.78 Oxymetazoline 22.06 t 12.18 30.58 2 4.79 PAC 19.25 ? 2.69 33.10 ? 12.3 Rauwolscine 32.50 t 10.50 26.71 t 12.30 Phentolamine 13.98 2 5.36 14.76 t 7.06 Benextramhe 17.10 t 3.53 46.30 ? 12.20 BDF 8933 35.16 2 18.95 39.68 5 2.81 Efaroxan 219 t 55 201 2 72 Yohimbine 232 2 46 293 2 99 Propanolol 395 t 160 418 ? 176 (-)Epinephrine 635 2 213 230 t 92 Prazosine 1772 t 242 1514 t 384 Cimetidine > I mM > I mM Digitoxine > I mM > I mM Dig o x i n e > I mM > I mM Ouabagenine > I mM 1 1 mM

Naphazoline 0.193 5 0.071 0.44 t 0.10

Ouabaine > I mM > I mM

a Inhibition studies of [3H]cirazoline (5 nM) binding to brain and kidney membranes were performed with imidazoline, guanidinium, phenylethylamine, and alkaloid compounds. Nonspe- cific binding was determined using 10 pM of unlabeled bromoxidine. The dissociation constant Ki of each drug is expressed in pM. Data represent the mean 2 SEM of four individual experiments. (See Materials and Methods for details.)

[3H]cirazoline, conditions in which the higher affinity site is saturated, and therefore they represent the pharmacological profile of the low affinity binding sites. Catechola- mines (adrenaline) and non-imidazoline adrenoceptor ligands such as benextramine, prazosin, propranolol, rauwolscine, or yohimbine did not compete with [3H]cirazoline binding (K, > 10 pM). Under our experimental conditions, only guanidinium and imidazoline derivatives except phentolamine (K, > 10 pM) had high affinity for [3H]cirazoline binding sites. Unlabeled cirazoline, clonidine, and bromoxidine had the highest affinity for [3H]cirazoline binding (K, c 0.05 pM) (TABLE 2). None of the compounds tested showed significant differences in their affinity for [3H]cirazoline binding sites in either brain or kidney homogenates.

118 ANNALS NEW YORK ACADEMY OF SCIENCES

TABLE 3. Kinetic Parameters of ['HICirazoline Binding to Rat Brain and Kidney Membrane Homogenates'

Brain Kidney

Higher Lower Higher Lower Kinetic Parameter Affinity Affinity Affinity Affinity

Kobs (min) 0.69 t 0.11 1.05 t 0.05 0.91 t 0.01 1.30 t 0.04 K-, ( m i d , I&-') 0.28 t 0.06 0.10 t 0.02 0.37 t 0.02 0.16 t 0.03 t,,* association

2.65 f 0.50 8.83 +. 2.99 1.85 f 0.10 4.57 Ifr 0.88 K, (min-I) 0.12 t 0.01 0.85 -C 0.01 0.20 t 0.04 0.99 t 0.07 t,,* dissociation

(mid 5.86 f 0.62 0.81 5 0.01 3.83 -C 0.98 0.70 t 0.05 KD apparent (nM) 0.46 t 0.09 10.8 t 3.63 0.56 t 0.16 6.74 t 1.82

Values obtained as described in FIGURE 3.

Characterization of [3H]Cirazoline Binding to HIT T15 Cells

In pancreatic beta cell line HIT T15 membranes, saturation analysis of ['Hlcirazo- line binding revealed two classes of high affinity binding sites (FIG. 5). For the higher affinity site, the dissociation constant, K,, was 0.22 ? 0.05 nM and the maximal binding density, B,,,, 56.5 It 12.7 fmol/mg protein. The K, for the lower affinity site was 11.12 t 1.29 nh4 and the B,,, 207.5 2 18.12 (TABLE 4).

Competition studies were carried out at 1 nM of ['Hlcirazoline using imidazoline, phenylethylamines, and guanidinium derivatives. Except cirazoline, which revealed a biphasic inhibition curve, all other compounds displayed monophasic inhibition curves (FIG. 6). As could be seen, this site was not recognized (K, > 10 pM) by non-imidazoline alpha- or beta-adrenoceptor ligands such as rauwolscine, prazosin, epinephrine, or propranolol.

Furthermore, a good correlation (I = 0.96, p < 0.01) was obtained between drug affinities towards ['Hlcirazoline binding in HIT T15 cells and in rat kidney (FIG. 7) or brain homogenates (I = 0.98, p c 0.01) (data not shown).

Comparison between [3H]Cirazoline Binding and az-Adrenoceptors

Specific binding of 1 nh4 ['Hlcirazoline represented 75.8 5 1.5%, 74.2 ? 2.5%, and 65.4 2 3.1% of total binding using brain, kidney, and HIT T15 membrane homogenates, respectively (FIG. 8). Specific binding to a2C2-, a2C4-, and a2C,,-AR CHO cell line preparations, res ectively, represented 1.5 ? 1.0%, 3.7 2 0.6%, and 2.5

represented 86.7 t 0.3%, 88.0 5 1.5%, and 83.2 t 0.7% (FIG. 8). These data indicate that the transfected cells contain large proportions of their

respective cloned a,-adrenoceptor subtype, but they do not possess measurable imida- zoline binding sites.

2 1.5% of total binding using [ P Hlcirazoline. However, using 1 nM [3H]rauwolscine it

ANGEL et al. : [3HlCIRAZOLINE 119

0.,2j 0,

0.026 O . O 3 O t

0.006 ~j 0 0.02 0.04 0.06

B (nM)

FIGURE. 5. (Top) Saturation analysis of [3H]cirazoline binding (1 pM to 50 nM) to HIT T1S cell membrane homogenates: (0) total [3H]cirazoline binding, (0) specific binding, and (0) nonspecific binding (in the presence of 10 pM of unlabeled bromoxidine). (Bottom) Scatchard analysis plot of [3HJcirazoline specific binding, indicating two classes of sites to HIT TI 5 membranes.

DISCUSSION

Previous studies have demonstrated that cirazoline has the highest affinity for the nonadrenergic imidazoline recognition site, with an inhibition constant of 0.5- 1 nM, compared to other imidazoline and guanidinium derivative^.^^' Studies to characterize the nonadrenergic imidazoline binding site are usually performed with radioligands such as [3H]clonidine, [3H]PAC, [3H]bromoxidine, or [3H]idazoxan

120 ANNALS NEW YORK ACADEMY OF SCIENCES

TABLE 4. Binding Parameters of [3H1Cirazoline in HIT T15 Cell Homogenates

Binding Parameter Higher Affinity

Lower Affinity

B,,, (fmoVmg protein) 56.5 -C 12.7 207.5 2 18.12 k (nM) 0.22 -C 0.05 11.12 2 1.29

which are less specific for imidazoline binding sites than is c i r a ~ o l i n e . ' * ~ ~ ~ ~ ' ~ Therefore, such studies were carried out in the presence of adrenaline, noradrenaline, or phentol- amine in order to occupy a,-adrenoceptor sites. In most of these binding studies, relative affinity for a adrenoceptors and imidazoline density of the sites depend on the radioligand ~ s e d ? ; . ~ , ~ . ~ In the present study, with the highly selective ligand ['H]cirazoline, we characterized nonadrenergic imidazoline binding sites in rat brain and kidney and in HIT T15 cell membrane homogenates in the absence of any additional drugs in the incubation medium to occupy the a,-adrenoceptors.

Using ['H]cirazoline, an a,-AR agonist and highly selective drug on imidazoline sites? we characterized two classes of high affinity binding sites on rat brain and kidney membranes and on HIT T15 cell homogenates. We previously described that the pharmacological profile of ['H]cirazoline binding sites using rat brain and kidney membranes corresponds to imidazoline sites.' Furthermore, because idazoxan and clonidine have similar affinities to the ['Hlcirazoline binding site, it was suggested that this site does not discriminate between the so-called I, and I, imidazoline sites. In the present study, no ['Hlcirazoline binding was detected using the a,-AR transfected CHO cell lines, indicating that they are devoid of imidazoline sites. However, our studies show high levels of specific binding of [3H]rauwolscine, a selective a,-AR antagonist, to the three CHO cell membranes.

These data confirm that imidazoline sites, as labeled by ['H]cirazoline, are neither a subunit nor a subtype of a,-adrenoceptors. They thereby confirm and extend previous observations in a,-adrenoceptor transfected COS-7 cells using eH]idazoxan.

Saturation experiments, performed with a range of [3Hlcirazoline concentrations from 1 pM to 50 nM did not reveal significant differences between brain and kidney preparations or between these tissues and HIT T15 cells. In all tissues, the dissociation constants of the high and low affinity binding sites are about 0.2-0.4 nM and 5-12 nM, respectively. The receptor density of the high affinity sites is -50 fmoYmg protein, whereas the B,,, of the low affinity sites Is -470 fmoYmg protein.

The two categories of ['HHJcirazoline binding sites just characterized have similar affinities to the two classes of imidazoline-preferring binding sites described by Greney and coworkers" using ['Hlidazoxan in the human nucleus reticularis lateralis. In addition, the dissociation constants of the low affinity ['Hlcirazoline binding sites are in the same range as are the Kd values observed using [3H]clonidine, ["]PAC, or [3Hlidazoxan.'v23'o However, in both tissues the receptor density of [3H]cirazoline binding sites is different from the results reported by other authors.1,2*10-12 These differences may be due to the radioligand used to characterize imidazoline-preferring sites and to the different experimental conditions.

ANGEL et al.: ['HICIRAZOLINE 121

log [DRUG] . Clr..olina A VR 1 4 304 I Clonldine 0 1d.ror.n A OhontoluIna . hnllorld. 0 5 1 84 0418 v Rauuolrolno . P*.%O.i".

- A - Cime.rIdm. 1- Epin.phrine

--(r Propanalol

FIGURE 6. Inhibition curves of [3H]cirazoline binding to HIT T15 cell homogenates. The indicated concentrations of competing drugs were added to the incubation medium (250 el final volume) containing homogenate preparation (400 pglml) and 1 nM of [3H]cirazoline for 30 minutes at 25°C.

Competition experiments were performed using 5 nM of [3H]cirazoline, a concen- tration at which near saturation of the high affinity binding site is obtained. Under these conditions two groups of compounds could be distinguished. The alkaloid and phenylethylamine derivatives show very low affinity (>lo FM). This included adrenoceptor ligands of different chemical classes and pharmacological activities such as (-)adrenaline (a- and P-agonist), prazosin (a,-antagonist), benextramine, rauwolscine, and yohimbine (a,-antagonists) as well as propranolol, a P-adrenoceptor antagonist (TABLE 1). The second group included the guanidinium and imidazoline derivatives. Phentolamine did not compete with [3H]cirazoline on these binding sites despite its imidazoline moiety (K, > 10 FM).

Similar low affinity of phentolamine for imidazoline sites was previously observed by other authors using [3H]bromoxidine or [3H]idazoxan as radi~ligands.~. '~ Idazoxan and amiloride possess similar intermediate affinity (K, - 0.15 FM). and the highest affinities were observed with bromoxidine, clonidine, and cirazoline (K, - 10-50 nM). The decreasing order of inhibitor potencies in brain membranes was cirazoline > bromoxidine > clonidine > idazoxan > amiloride, whereas in membranes from kidney preparations it was cirazoline > clonidine 2 bromoxidine > amiloride > idazoxan. It is now evident that imidazoline-preferring sites can be classified into at least two groups.14 The I, types, so-called clonidine-preferring sites, bind [3H]PAC and [3H]clonidine, preferentially showing low affinity for amiloride. The I, types or idazoxan-preferring sites are labeled by [3H]idazoxan and can be subdivided into two subtypes such as I,, and I,,, possessing respectively high and low affinity for amiloride. The high potencies of clonidine and bromoxidine to inhibit [3H]cirazoline

122 ANNALS NEW YORK ACADEMY OF SCIENCES

1000

100

10

s 1 .a u;

Y t c I

0.1

0.01

0.00’ C

Coefficient of correlation :

1 1 0.01 0 1 1 0 10 100 1000

KIDNEY (K,, #M)

FIGURE 7. Correlation between drug affinities to [3H]cirazoline binding in HIT T15 cells and rat kidney homogenates. Inhibition constants for HIT T15 cells and brain homogenates (Ki, pM) were used as described in FIGURE 6 and TABLE 1 , respectively.

binding would support the terminology of I, sites. However, the [3H]cirazoline binding sites are well recognized by guanidinium compounds such as amiloride, and therefore they appear to be different from the sites labeled by [3H]PAC.15 On the other hand, the high affinity of unlabeled cirazoline and the potencies of idazoxan and amiloride rather indicate that we are dealing with I, sites. These results suggest that [3H]cirazo- line may not distinguish between the I, and the I, sites or that it represents another, as yet uncharacterized subtype of imidazoline-preferring site.

In summary, [3H]cirazoline is a novel high affinity radioligand that specifically labels imidazoline-guanidinium recognition sites without significant a- or p-adreno- ceptor binding on rat brain and kidney membranes and HIT T15 cells. These special properties make [3H]cirazoline a very useful ligand for further studies on imidazoline recognition sites. In addition, cirazoline is a useful tool for in v i m and in vivn characterization of the function associated with imidazoline recognition sites.

SUMMARY

Recent studies have shown that cirazoline, an a,-adrenoceptor agonist, has greater affinity than do other imidazoline or guanidinium compounds at imidazoline recogni-

ANGEL et aL: ['HJCIRAZOLINE 123

100

90

80

m 70

5 60 E

50 1 4o u) ' 30

20

10

0

Brain Kidney HIT T15 calls

(3HjCirazoline [3H]Rauwolscine

FIGURE 8. Specific binding of [3H]cirazoline and [3H]rauwolscine on membrane homoge- nates. Specific bindings are expressed as percentage of total binding and results as the mean t SEM of n experiments performed in triplicate (n = 4 for [3H]cirazoline binding; n = 3-7 for [3H]rauwolscine binding).

tion sites. In this report we used [3H]cirazoline as a probe to characterize imidazoline recognition sites present in membrane homogenates of rat brain and kidney as well as pancreatic beta HIT T15 cells.

Specific binding of [3H]cirazoline to these various homogenates was saturable and reversible and was resolved into two classes of high affinity binding sites. Competition inhibition studies of [3H]cirazoline binding to these different membrane preparations were performed with alkaloid, phenylethylamine, imidazoline, and gua- nidinium compounds. Catecholamines and non-imidazoline adrenoceptor ligands such as epinephrine, benextramine, prazosin, propranolol, rauwolscine, or adrenoceptor ligands such as epinephrine, benextramine, prazosin, propranolol, rauwolscine, or yohimbine did not compete with [3H]cirazoline (K, > 10 kM). Under our experimental conditions, only guanidinium and imidazoline derivatives had high affinities for [3H]cirazoline binding sites. Unlabeled cirazoline, clonidine, bromoxidine, idazoxan, and amiloride had the highest affinities with this respective rank order. These results suggest that [3H]cirazoline is a novel high affinity radioligand that specifically labels nonadrenergic imidazoline-guanidinium sites in the brain, kidney, and beta cells. Furthermore, the obtained rank order of inhibition suggests that [3H]cirazoline binding does not distinguish between I, and I, sites. In addition, we compared the specific binding of ['HH]cirazoline with that of the a,-adrenoceptor antagonist [3H]rauwolscine in Chinese hamster ovary (CHO) cell lines stably expressing human a,C,-, a&-, and a,C,,-adrenoceptor subtypes. Using [3H]rauwolscine as a probe, each of these transfected cell lines expressed high levels for the three different a,-adrenoceptor subtypes (Bmm values were between 2 and 7 pmol.mg-1 protein). In contrast, none of these cell lines displayed measurable imidazoline recognition sites.

In summary, [3H]cirazoline is a novel high affinity radioligand that specifically labels imidazoline recognition sites without significant a- or P-adrenoceptor binding.

124 ANNALS NEW YORK ACADEMY OF SCIENCES

Furthermore, our results using a,-adrenoceptor transfected cells confirm that the imidazoline recognition sites and each of the cloned a,-adrenoceptor subtypes repre- sent distinct macromolecular entities.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

REFERENCES

ERNSBERCER, P., M. P. MEELEY, J. MA" & D. J. REIS. 1987. Clonidine binds to imidazole binding sites as well as a,-adrenoceptors in the ventrolateral medulla. Eur. J. Pharmacol. 134: 1.

HAMILTON, C. A., M. A. YAKUBU, E. JARDINE & J. L. REID. 1991. Imidazole binding sites in rabbit kidney and forebrain membranes. J. Auton. Pharmacol. 11: 277.

LOMASNEY, J. W., W. LORENZ, L. F. ALLEN, K. KING, J. W. REGAN, T. L. YANG-FENG, M. G. CARON & R. J. LEFKOWITZ. 1990. Expansion of the a,-adrenergic receptor family: Cloning and characterization of a human a,-adrenergic receptor subtype, the gene for which is located on chromosome 2. Proc. Natl. Acad. Sci. USA 87: 5094.

MICHEL, M. D., J. W. RECAN, M. A. GERHARDT, R. R. NENBIG, P. INSEL & H. J. MOTULSKY. 1989. Nonadrenergic [3H]idazoxan binding sites are physically distinct from a,-adrener- gic receptors. Mol. Pharmacol. 35: 324-330.

DE Vos, H., A. COUVENTS, J. DEKEYSER, J. P. DEBAKER, I. J. B. VAN MEGEN, G. EBINGER & G. VAUQUELIN. 1991. Autoradiographic distribution of a adrenoceptors,

Brain Res. 566: 13. MIRALLES, A,, G. OLMOS, M. SASTRE, F. BARUREN, 1. MARTIN & J. A. GARCIA-SEVILLA.

1993. Discrimination and pharmacological characterization of I,-imidazoline sites with [3H]idazoxan and a,-adrenoceptor with [3H]RX 821 002 (2-methoxy idazoxan) in the human and rat brain. J. Pharmacol. Exp. "her. 264: 1187.

WIKWRG, J. E. S. & S. UHLEN. 1990. Further characterization of the guinea pig cerebral cortex idazoxan receptor: Solubilization, distinction from imidazoline site, and demon- stration of cirazoline as an idazoxan receptor selective drug. J. Neurochem. 55: 192.

ANGEL, I., H. SCHOEMAKER, J. ALLEN, M. LE ROUZIC, C. PIMOULE, S. ARBILLA & S. Z. LANCER. 1992. The use of [3H]cirazoline, a novel, selective and high affinity ligand to label the imidazoline recognition site. Fundam. Clin. Pharmacol. Q(supp1. 1): P15, 54s.

MUNSON, P. J. & D. RODBARD. 1980. LIGAND: A versatile computerised approach for characterization of ligand binding systems. Anal. Biochem. 107: 220.

COWRY, I., R. A. PODEVIN, J. P. DAUSSE & A. PARINI. 1987. Evidence for imidazoline binding site in basolateral membranes from rabbit kidney. Biochem. Biophys. Res. Commun. 147: 1055.

ZONNENSCHEIN, R., S. DIAMANT & D. ATLAS. 1990. Imidazoline receptor in rat liver cell: A novel receptor or a subtype of a,-adrenoceptor? Eur. J. Pharmacol. 190: 203.

GRENEY, H., G. BRICCA, M. DONTENWILL, A. BELCOURT & P. BOUSQUET. 1992. [3H]ida- zoxan and [3H]clonidine binding in the human Nucleus Reticularis Lateralis (NRL). Fundam. Clin. Pharmacol. Q(supp1. I): 51s.

LANGIN, D. & M. LAFONTAN. 1989. [3H]idazoxan binding at non-a,adrenoceptors in rabbit adipocyte membranes. Eur. J. Pharmacol. 159: 199.

REIS, D. J., S. REGIINATHAN, H. WANG, D. L. FEINSTEN & M. P. MELLEY. 1992. Imidazoline receptors in the nervous system. Fundam. Clin. Phannacol. 6 (suppl. 1): 23s.

MICHEL, M. C. & P. A. INSEL. 1989. Are there multiple imidazoline binding sites? Trends Pharmacol. Sci. 10: 342.

NAIBS and 5-HT,, receptors in human brain using [3H]idazoxan and [ 3- H]rauwolscine.