Volume 122, Issue 6 p. 1202-1208
Free Access

Species differences in brain adenosine A1 receptor pharmacology revealed by use of xanthine and pyrazolopyridine based antagonists

Takuya Maemoto

Takuya Maemoto

Fujisawa Institute of Neuroscience

Search for more papers by this author
Keith Finlayson

Keith Finlayson

Department of Pharmacology, University of Edinburgh, Edinburgh EH8 9JZ

Search for more papers by this author
Henry J. Olverman

Henry J. Olverman

Department of Pharmacology, University of Edinburgh, Edinburgh EH8 9JZ

Search for more papers by this author
Atsushi Akahane

Atsushi Akahane

New Drug Research Laboratories, Fujisawa Pharmaceutical Co. Ltd., Osaka 532, Japan

Search for more papers by this author
Roger W. Horton

Roger W. Horton

Department of Pharmacology, St. George's Hospital Medical School, London SW17 0RE

Search for more papers by this author
Steven P. Butcher

Corresponding Author

Steven P. Butcher

Fujisawa Institute of Neuroscience

Fujisawa Institute of Neuroscience, *Department of Pharmacology, University of Edinburgh, Edinburgh EH8 9JZSearch for more papers by this author
First published: 03 February 2009
Citations: 52

Abstract

  • 1

    The pharmacological profile of adenosine A1 receptors in human, guinea-pig, rat and mouse brain membranes was characterized in a radioligand binding assay by use of the receptor selective antagonist, [3H]-8-cyclopentyl-1,3-dipropylxanthine ([3H]-DPCPX).

  • 2

    The affinity of [3H]-DPCPX binding sites in rat cortical and hippocampal membranes was similar. Binding site affinity was higher in rat cortical membranes than in membranes prepared from guinea-pig cortex and hippocampus, mouse cortex and human cortex. pKD values (M) were 9.55, 9.44, 8.85, 8.94, 8.67, 9.39 and 8.67, respectively. The binding site density (Bmax) was lower in rat cortical membranes than in guinea-pig or human cortical membranes.

  • 3

    The rank order of potency of seven adenosine receptor agonists was identical in each species. With the exception of 5′-N-ethylcarboxamidoadenosine (NECA), agonist affinity was 3.5–26.2 fold higher in rat cortical membranes than in human and guinea-pig brain membranes; affinity in rat and mouse brain membranes was similar. While NECA exhibited 9.3 fold higher affinity in rat compared to human cortical membranes, affinity in other species was comparable. The stable GTP analogue, Gpp(NH)p (100 μM) reduced 2-chloro-N6-cyclopentyladenosine (CCPA) affinity 7–13.9 fold, whereas the affinity of DPCPX was unaffected.

  • 4

    The affinity of six xanthine-based adenosine receptor antagonists was 2.2–15.9 fold higher in rat cortical membranes compared with human or guinea-pig membranes. The rank order of potency was species-independent. In contrast, three pyrazolopyridine derivatives, (R)-1-[(E)-3-(2-phenylpyrazolo[1,5-a]pyridin-3-yl) acryloyl]-2-piperidine ethanol (FK453), (R)-1-[(E)-3-(2-phenylpyrazolo[1,5-a]pyridin-3-yl) acryloyl]-piperidin-2-yl acetic acid (FK352) and 6-oxo-3-(2-phenylpyrazolo[1,5-a]pyridin-3-yl)-1(6H)-pyridazinebutyric acid (FK838) exhibited similar affinity in human, guinea-pig, rat and mouse brain membranes. pKi values (M) for [3H]-DPCPX binding sites in human cortical membranes were 9.31, 7.52 and 7.92, respectively.

  • 5

    Drug affinity for adenosine A2A receptors was determined in a [3H]-2-[4-(2-carboxyethyl)phenethylamino]-5′-N-ethylcarboxamidoadenosine ([3H]-CGS 21680) binding assay in rat striatal membranes. The pyrazolopyridine derivatives, FK453, FK838 and FK352 exhibited pKi values (M) of 5.90, 5.92 and 4.31, respectively, compared with pKi values of 9.31, 8.18 and 7.57 determined in the [3H]-DPCPX binding assay in rat cortical membranes. These novel pyrazolopyridine derivatives therefore represent high affinity, adenosine A1 receptor selective drugs that, in contrast to xanthine based antagonists, exhibit similar affinity for [3H]-DPCPX binding sites in human, rat, mouse and guinea-pig brain membranes.

British Journal of Pharmacology (1997) 122, 1202–1208; doi:10.1038/sj.bjp.0701465

Introduction

Although adenosine receptors were originally subdivided into two classes according to functional data (van Calker et al., 1979), recent pharmacological and molecular cloning studies suggest the existence of at least four subclasses, termed A1, A2A, A2B and A3 receptors (Fredholm et al., 1994; Palmer & Stiles, 1995). The adenosine A1 receptor has been particularly well characterized in both pharmacological and functional studies (Linden, 1991; Fredholm et al., 1994; Palmer & Stiles, 1995). Receptor selective agonists, such as N6-cyclopentyladenosine (CPA) and 2-chloro-N6-cyclopentyladenosine (CPA) (Williams et al., 1986; Klotz et al., 1989), and xanthine-based antagonists are available (Schwabe et al., 1985; Bruns et al., 1987; Lohse et al., 1987); [3H]-8-cyclopentyl-1,3-dipropylxanthine ([3H]-DPCPX) is presently the ligand of choice for use in radioligand binding assays (Bruns et al., 1987; Lohse et al., 1987).

Marked species differences in the affinity of adenosine receptor agonists and antagonists have been described (Murphy & Snyder, 1982; Ferkany et al., 1986; Ukena et al., 1986; Klotz et al., 1991; Nonaka et al., 1996). Molecular cloning studies of the adenosine A1 receptor from dog, human, rat, cow, and guinea-pig suggest that these pharmacological discrepancies may be related to the primary amino acid sequence of the receptor (Libert et al., 1991; 1992; Mahan et al., 1991; Reppert et al., 1991; Townsend-Nicholson & Shine, 1992; Tucker et al., 1992; Olah et al., 1992; Meng et al., 1994). Differences in G protein coupling with adenosine A1 receptors purified from rat, human, and bovine brain (Jockers et al., 1994; Nanoff et al., 1995), and immunological differences between the human and rat receptor (Nakata, 1993), have also been demonstrated. In the present study a comparison was obtained of the pharmacological profile of three novel pyrazolopyridine derivatives, FK453 (Terai et al., 1995), FK352 and FK838 (Figure 1), and reference adenosine receptor agonists and antagonists, in a [3H]-DPCPX binding assay by use of brain membranes prepared from human, rat, mouse and guinea-pig. Drug selectivity for adenosine A1; compared with A2A, receptors was also assessed by use of a [3H]-CGS 21680 binding assay with rat striatal membranes.

Details are in the caption following the image

Chemical structures of the pyrazolopyridine derivatives, FK453, FK352 and FK838.

Methods

Human brain tissue

Human brain tissue was obtained at autopsy from 5 male patients with no history of psychiatric or neurological disease. The age of the subjects was 55.0±2.5 years (mean±s.e.mean), and the post mortem delay was 34.6±9.6 h (mean±s.e.mean). Parietal cortex (Brodman area 7) was dissected from fresh tissue, pooled (total weight 9.6 g), and frozen at -80°C until use.

Membrane preparation

Male Sprague-Dawley rats (200–400 g; Charles-River), male Dunkin Hartley guinea-pigs (500–1000 g; Darley-Oak-Farm) and male HPG n/n mice (bred in-house) were killed by cervical dislocation. The brains were removed and immediately placed in ice-cold saline, before the cortex, striatum and hippocampus were dissected as appropriate. Animal and human tissues were homogenized in 15 volumes (vol) of 0.32 m sucrose with a glass/Teflon homogenizer, the homogenate centrifuged at 1,000 g for 10 min, and the resulting supernatant was centrifuged at 17,000 g for 20 min. The synaptosomal/mitochondrial P2 pellet was lyzed in 30 vol of ice-cold water for 30 min, and centrifuged at 48,000 g for 10 min. The membrane pellet was resuspended in 30 vol of 50 mm Tris-HCl buffer (pH 7.4), centrifuged at 48,000 g for 10 min, resuspended in 5 vol of 50 mm Tris-HCl buffer (pH 7.4) and stored at -20°C. Frozen membranes were thawed on the day of use and resuspended in 30 vol of 50 mm Tris-HCl buffer (pH 7.4). The suspension was centrifuged at 48,000 g for 10 min, and the pellet resuspended in 200 vol of 50 mm Tris-HCl buffer (pH 7.4) and kept on ice before use in the binding assay; protein content was determined as described previously (Bradford, 1976).

[3H]-DPCPX binding assay

[3H]-DPCPX binding was assessed by use of a modification of previous methods (Bruns et al., 1987; Lohse et al., 1987). DMSO 10 μl or test drug was incubated with 100 μl of adenosine deaminase (ADA; 0.1 iu ml−1 final assay concentration), 290 μl of 50 mm Tris-HCl buffer (pH 7.4), 100 μl of [3H]-DPCPX (NEN; 98.1 Ci mmol−1) and 500 μl of membrane suspension. The final assay concentration of [3H]-DPCPX was 0.1 nm for rat and mouse membranes, 0.2 nm for guinea-pig membranes and 0.7 nm for human membranes. Non-specific binding was determined in the presence of 10 μm R(-)N6-(2-phenylisopropyl)adenosine (R-PIA). ADA was included to remove endogenous adenosine at a concentration that did not influence [3H]-DPCPX binding (data not shown). Test compounds were prepared by serial dilution in DMSO; the final assay concentration of 1% DMSO did not affect [3H]-DPCPX binding (data not shown). The incubation times were 20 min, 120 min and 60 min at 25°C for rat, mouse, human and guinea-pig membranes respectively; preliminary experiments confirmed that binding was at equilibrium under these conditions (data not shown). The binding assay was terminated by filtration onto glass filters (GF/B, Whatman) by use of a Brandel Cell Harvester, followed by three washes (3 ml) with 50 mm Tris-HCl buffer (pH 7.4). Filter disks were transferred to scintillation vials, 100 μl of formic acid was added, followed 10 min later by 4 ml Emulsifier SAFE scintillation fluid. Vials were left overnight before radioactivity was determined in a Packard 2500TR liquid scintillation analyser with automatic quench correction.

[3H]-CGS 21680 binding assay

[3H]-CGS 21680 binding was determined by use of identical conditions to the [3H]-DPCPX binding assay except that 500 μl of rat striatal membrane suspension was added to 290 μl of 50 mm Tris-HCl buffer (pH 7.4) containing 10 mm MgCl2 (final assay concentration), 100 μl of [3H]-CGS 21680 (NEN; 40.5 Ci mmol−1; 2 nm final assay concentration), 100 μl of ADA (0.1 iu ml−1; final assay concentration) and 10 μl of DMSO or test drug. Addition of ADA and DMSO (1% final assay concentration) did not affect [3H]-CGS 21680 binding (data not shown). The incubation was continued for 60 min at 25°C before termination by filtration with a Brandel Cell Harvester. Filter disks were washed and radioactivity determined as described previously. Non-specific binding was determined in the presence of 10 μm NECA.

Data analysis

Data were analysed by use of an iterative, non-linear least square curve fitting programme (SigmaPlot; Jandel, U.S.A.) to the logistic expression; Y = M. IC50P/(IP + IC50P) + B, where P is the Hill coefficient and Y is bound ligand in the presence of inhibitor concentration, I; M and B are specific binding in the absence of inhibitor and non-specific binding, respectively. Estimates of M and B were within 10% of experimentally determined values. If the inhibitor was the unlabelled form of the radioligand, KD and Bmax were calculated by use of the equations; KD = IC50-[ligand] and Bmax = (M. IC50)/[ligand]. For other test compounds, Ki values were calculated from the equation Ki = IC50/(1 + ([ligand]/KD)). Statistical comparisons were performed by use of commercially available software (JMP 3.2; SAS Institute Inc.). Before statistical analysis by ANOVA, box plots were inspected to ensure a normal symmetrical distribution of data, and the homogeneity of variance was confirmed to be within acceptable limits. Thereafter, ANOVA was used to demonstrate a significant interaction between the receptor affinity of each drug and the tissue source (data not shown), with further post hoc analysis by use of Dunnett's method.

Materials

CCPA, CPA, R-PIA, 2-[4-(2-carboxyethyl)phenylethylamino]-5′-N-ethylcarboxamido-adenosine (CGS 21680), 5′-N-ethyl-carboxamidoadenosine (NECA), DPCPX, 2-chloroadenosine (CADO), N6-cyclohexyladenosine (CHA), 1,3-diethyl-8-phenylxanthine (DPX), 8-cyclopentyl-1,3-dimethylxanthine (CPT), 8-phenyltheophylline (8-PT) were purchased from Research Biochemicals International (Natick, U.S.A.). 8-(Noradamantan-3-yl)-1,3-dipropylxanthine (KW 3902; Nonaka et al., 1996), KFM-19 (Linden, 1991), (R)-8-(1-phenylpropyl)-1,3-dipropylxanthine (MDL 102234; Dudley et al., 1992), (R)-1-[(E)-3-(2-phenylpyrazolo[1,5-a]pyridin-3-yl)acryloyl]-2-pipendine ethanol (FK453), (R)-1-[(E)-3-(2-phenylpyrazolo[1,5-a]pyridin-3-yl)acryloyl]pipendin-2-yl acetic acid (FK352) and 6-oxo-3-(2-phenylpyrazolo[1,5-a]pyridin-3-yl)-](6H0-pyridazinebutyric acid (FK838) were synthesized by Fujisawa Pharmaceutical Co. Ltd. (Japan). Adenosine deaminase Type III (ADA), 5′-guanylimidodiphosphate (Gpp(NH)p) and other chemicals were from Sigma (Poole, U.K.).

Results

Species differences in [3H]-DPCPX binding site affinity (pKD) and density (Bmax)

Competition experiments with membranes prepared from human, rat, mouse and guinea-pig brain tissue revealed marked species differences in [3H]-DPCPX binding site affinity. pKD values (m) ranged from 9.55 in rat cortical membranes to 8.67 in human cortical membranes (Figure 2; Table 1). Affinity in rat cortical membranes was significantly higher than in mouse, guinea-pig or human brain membranes (P<0.001). The rank order of affinity was rat cortex≅rat hippocampus ≥ mouse cortex > guinea-pig hippocampus≅uinea-pig cortex > human cortex. [3H]-DPCPX binding site affinity in cortical and hippocampal membranes prepared from the same species was similar (Table 1). Binding site density (Bmax) was lower in rat cortical membranes than in guinea-pig and human brain membranes (P<0.05; Table 1). The affinity of DPCPX was unaltered in the presence of Gpp(NH)p, a stable GTP analogue (100 μm; Table 2). Hill coefficients in the presence and absence of Gpp(NH)p were not significantly different from unity (data not shown).

Details are in the caption following the image

Inhibition of [3H]-DPCPX binding to human, rat, mouse and guinea-pig cortical membranes by DPCPX. P2 membranes were incubated with [3H]-DPCPX (0.1 nm, rat and mouse; 0.2 nm, guinea-pig; 0.7 nm, human) in 50 mm Tris-HCl buffer (pH 7.4) containing unlabelled DPCPX and 0.1 iu ml−1 ADA. Binding was terminated after varying incubation times (20 min, rat and mouse; 60 min, guinea-pig; 120 min, human) by filtration with a Brandel Cell Harvester. Data shown are representative competition curves from a single experiment; pKD and Bmax values were determined from at least three independent experiments.

Table 1. Comparison of the affinity (pKD) and binding site density (Bmax) of [3H]-DPCPX binding sites in membranes prepared from the indicated species and brain regions
Rat Mouse Guinea-pig Human
Cortex Hippocampus Mouse Cortex Hippocampus Cortex
Control PKD 9.55±0.04 9.44±0.05 9.39±0.06*** 8.85±0.09*** 8.94±0.10*** 8.67±0.28***
Bmax 2.38±0.49 3.86±1.23 3.41±0.73 5.26±1.68*** 4.50±1.75** 4.96±4.22*
+ 100mm pKD 9.56±0.07 9.56±0.02 9.33±0.09*** 8.84±0.09*** 8.82±0.17*** n.d.
Gpp(NH)p Bmax 2.45±1.77 3.60±0.96 3.54±0.29 4.27±0.69* 5.78±2.86*** n.d.
  • pKD (m) and Bmax (pmol/mg−1 protein) values are means±95% confidence limits determined from at least three independent experiments. Gpp(NH)p was included in the assay buffer at 100μm where indicated. Statistical analyses compared data with values obtained in rat cortex; *P<0.05, **P<0.01, ***P<0.001 (Dunnett's method). ND = not determined
Table 2. Affinity (pKi) of adenosine receptor agonists for [3H]-DPCPX binding sites in membranes prepared from the indicated species and brain regions
Rat Guinea-pig
Cortex Hippocampus Mouse Cortex Cortex Hippocampus Human Cortex
CCPA 8.48±0.08 8.50±0.48 8.33±0.29 7.60±0.19*** 7.78±0.23*** 7.33±1.22***
+ Gpp(NH)p 7.43±0.24 7.61±0.34 7.14±0.06 6.69±0.26*** 6.71±0.19*** 6.43±0.18***
CPA 8.40±0.16 8.55±0.47 8.48±0.58 7.54±0.21*** 7.49±0.34*** 7.22±0.43***
R-PIA 8.29±0.08 8.20±0.21 8.21±0.17 7.30±0.29*** 7.43±0.33*** 7.08±0.48***
CHA 8.17±0.15 8.23±0.40 8.19±0.35 7.18±0.24*** 7.28±0.39*** 6.75±0.30***
NECA 7.52±0.15 7.75±0.17 7.83±0.17 7.29±0.40 7.52±0.42 6.53±0.47***
CADO 7.45±0.11 7.42±0.66 7.34±0.94 6.94±0.23** 6.86±0.29** 6.51±2.16***
CGS21680 4.44±0.26 n.d. 4.85±0.30* 4.66±0.52 4.65±0.25 3.95±0.08*
  • pKi values (m) are means±95% confidence limits determined from at least three independent experiments. Gpp(NH)p was included in the assay buffer at 100 μm where indicated. Statistical analyses compared data with values obtained in rat cortex; *P<0.05, **P<0.01, ***P<0.001 (Dunnett's method). ND = not determined.

Affinity of adenosine receptor antagonists for brain [3H]-DPCPX binding sites

Although species differences in the affinity of adenosine receptor agonists were apparent, the rank order of potency was CCPA≅CPA≥R-PIA≥CHA>NECA≅CADO>>CGS 21680 in each membrane preparation (Table 2). With the exceptions of NECA and CGS 21680, adenosine based receptor agonists exhibited 3.5–26.2 fold higher affinity in rat cortical membranes than in brain membranes prepared from human or guinea-pig tissue (P<0.001; Table 2); agonist affinity in rat and mouse cortical membranes was similar. Regional differences between cortical and hippocampal membranes prepared from the same species were not noted (Table 2). In the case of CCPA, pKi values (m) ranged from 8.50 in rat hippocampal membranes to 7.33 in human cortical membranes (Figure 3; Table 2). While NECA exhibited 9.3 fold higher affinity in rat compared to human cortical membranes (P<0.001), affinity in other species was comparable (Table 2). The receptor selectivity of the [3H]-DPCPX binding assay was confirmed by the finding that CGS 21680, a specific adenosine A2A agonist, exhibited negligible affinity with pKi values (m) ranging from 4.44 in rat cortical membranes to 3.95 in human cortical membranes. With the exception of CGS 21680 (Hill coefficient = 1), Hill coefficients of 0.6–0.8 were noted in each membrane preparation, suggesting the presence of more than one agonist affinity state (data not shown). The effect of Gpp(NH)p on the affinity of CCPA for [3H]-DPCPX binding sites was also determined. In contrast to DPCPX, addition of Gpp(NH)p (100 μm) reduced CCPA affinity 7–13.9 fold (P<0.05), with no obvious inter-species differences (Table 2).

Details are in the caption following the image

Inhibition of [3H]-DPCPX binding to human, rat, mouse and guinea-pig cortical membranes by CCPA. P2 membranes were incubated with [3H]-DPCPX (0.1 nm, rat and mouse; 0.2 nm, guinea-pig; 0.7 nm, human) in 50 mm Tris-HCl buffer (pH 7.4) containing competing drug and 0.1 iu ml−1 ADA. Binding was terminated after varying incubation times (20 min, rat and mouse; 60 min, guinea-pig; 120 min, human) by filtration with a Brandel Cell Harvester. Data shown are representative competition curves from a single experiment; pKi values were determined from at least three independent experiments.

Affinity of xanthine-based adenosine receptor antagonists for brain [3H]-DPCPX binding sites

The affinity of xanthine-based adenosine receptor antagonists for [3H]-DPCPX binding sites was species-dependent, despite a common rank order of potency of KW 3902>DPCPX>KFM-19>CPT≅MDL 102234>DPX≥8-PT (Tables 1 and 3). Regional differences between cortical and hippocampal membranes prepared from rat and guinea-pig were not noted (Table 3). Antagonist Hill coefficients were close to unity in all species, indicating the presence of a single antagonist affinity state (data not shown). The affinity of CPT, DPX and 8-PT was 4.8–15.9 fold higher in rat cortical membranes than in human or guinea-pig brain membranes (P<0.001); affinity in rat and mouse membranes was similar (Table 3). The affinity of KW 3902, KFM-19 and MDL 102234 was 2.2–8.3 fold higher in rat cortical membranes compared with human and guinea-pig brain membranes (P<0.05), whereas pKi values in rat and mouse membranes were similar (Table 3).

Table 3. Affinity (pKi) of adenosine receptor antagonists for [3H]-DPCPX binding sites in membranes prepared from the indicated species and brain regions
Rat Guinea-pig
Cortex Hippocampus Mouse Cortex Cortex Hippocampus Human Cortex
KW 3902 9.90±0.13 10.2±1.09 9.84±0.65 9.43±0.05*** 9.49±0.41** 9.53±0.24**
KFM-19 8.87±0.10 8.80±0.19 8.67±0.09 8.04±0.22*** 8.09±0.14*** 7.93±0.10***
CPT 8.26±0.15 8.23±0.15 8.14±0.07 7.53±0.05*** 7.55±0.04*** 7.31±0.12***
MDL 102234 8.15±0.06 8.20±0.30 8.05±0.72 7.60±0.37*** 7.59±0.20*** 7.44±0.09***
DPX 7.50±0.0 7.40±0.09 7.37±0.06 6.51±0.07*** 6.52±0.08*** 6.36±0.45***
8-PT 7.35±0.07 7.32±0.16 7.28±0.16 6.44±0.05*** 6.45±0.14*** 6.15±0.37***
FK453 9.31±0.09 9.37±0.13 9.26±0.33 9.19±0.09 9.21±0.07 9.31±0.76
FK838 8.18±0.09 7.96±0.13 7.99±0.34 8.00±0.09 8.12±0.07 7.92±0.76*
FK352 7.57±0.09 7.55±0.09 7.50±0.42 7.65±0.08 7.72±0.07 7.52±0.43
  • pKi values (m) are means±95% confidence limits determined from at least three independent experiments. Statistical analyses compared data with values obtained in rat cortex; *P<0.05, **P<0.01, ***P<0.001 (Dunnett's method).

Affinity of pyrazolopyridine derivatives for brain [3H]-DPCPX binding sites

In contrast to data obtained with xanthine-based adenosine receptor antagonists, three novel pyrazolopyridine derivatives, FK453, FK838 and FK352 did not exhibit species differences in affinity for [3H]-DPCPX binding sites (Table 3). For example, pKi values (m) for FK453 and FK352 in rat cortical membranes were 9.31 and 7.57, respectively, compared with 9.31 and 7.52, respectively, in human cortical membranes (Figure 4; Table 3). The affinity of FK838 was two fold higher in rat cortical membranes than in human cortical membranes (Table 3), whereas pKi values in rat hippocampal and human cortical membranes were similar. Hill coefficients of close to unity were noted in each species (data not shown).

Details are in the caption following the image

Inhibition of [3H]-DPCPX binding to human, rat, mouse and guinea-pig cortical membranes by FK453. P2 membranes were incubated with [3H]-DPCPX (0.1 nm, rat and mouse; 0.2 nm, guinea-pig; 0.7 nm, human) in 50 mm Tris-HCl buffer (pH 7.4) containing competing drug and 0.1 iu ml−1 ADA. Binding was terminated after varying incubation times (20 min, rat and mouse; 60 min, guinea-pig; 120 min, human) by filtration with a Brandel Cell Harvester. Data shown are representative competition curves from a single experiment; pKi values were determined from at least three independent experiments.

Pharmacological selectivity studied in a [3H]-CGS 21680 binding assay

Adenosine receptor selectivity was evaluated by use of a [3H]-CGS 21680 binding assay with rat striatal membranes (Jarvis et al., 1989; Table 4). The pKD of rat striatal [3H]-CGS 21680 binding sites was 8.44, compared with a pKi value in the [3H]-DPCPX binding assay of 4.44, confirming the receptor selectivity of the assay (Tables 2 and 4). On the basis of pKi values for [3H]-DPCPX and [3H]-CGS 21680 binding sites in rat cortical and striatal membranes, respectively (Tables 2 and 4), the adenosine receptor agonists CCPA, CPA, R-PIA, CHA, CADO and NECA showed 93, 46, 13, 43, 0.55 and 0.13 fold selectivity for adenosine A1 compared with adenosine A2A receptors. The xanthine based adenosine receptor antagonists, DPCPX, CPT, KFM-19, KW 3902, MDL 102234 exhibited 546, 129, 1464, 1352 and 100 fold selectivity for the rat cortical adenosine A1 receptor (Tables 3 and 4). Similarly, the pyrazolopyridine derivatives, FK453, FK352 and FK838 showed 2514, 1855 and 181 fold selectivity for the rat cortical adenosine A1 receptor (Tables 3 and 4).

Table 4. Affinity (pKi of adenosine receptor agonist and antagonists for [3H]-CGS 21680 binding sites in rat striatal membranes
Agonist pKi Binding site selectivity (A1:A2A) Antagonist pKi Binding site selectivity (A1:A2A)
CGS 21680 8.44±0.46 0.0014 DPCPX 6.82±0.05 546
NECA 8.38±0.15 0.13 KW 3902 6.74±0.10 1352
CADO 7.70±0.15 0.55 DPX 6.58±0.10 8.1
R-PIA 7.18±0.17 13 8-PT 6.45±0.06 7.8
CPA 6.83±0.55 46 MDL 102234 6.15±0.28 100
CHA 6.54±0.21 43 CPT 6.11±0.06 129
CCPA 6.53±0.25 93 FK838 5.92±0.03 181
FK453 5.90±0.16 2514
KFM-19 5.68±0.19 1464
FK352 4.31±0.41 1855
  • pKi values (m) are means±95% confidence limits determined from at least three independent experiments. Binding site selectivity determined from drug affinities in [3H]-CGS 21680 and [3H]-DPCPX binding assays is also shown (values>1 indicate selectivity for [3H]-DPCPX binding sites).

Discussion

The novel pyrazolopyridine derivatives, FK453, FK352 and FK838 exhibit high affinity for [3H]-DPCPX binding sites in brain membranes prepared from human, rat, mouse and guinea-pig. The adenosine A1:A2A receptor selectivity of these compounds was comparable to xanthine based antagonists; data concerning drug affinity for the adenosine A3 receptor is not presently available. These data compliment functional studies demonstrating that FK453, FK352 and FK838 act as competitive adenosine A1 receptor antagonists (Terai et al., 1995; Ito et al., unpublished observations). In contrast to xanthine based antagonists, the pyrazolopyridine derivatives exhibited similar affinity for [3H]-DPCPX binding sites in each species tested. The use of well established radioligands (Bruns et al., 1987; Lohse et al., 1987; Jarvis et al., 1989), and the systematic nature of the present study, ensured the consistency and reliability of this finding.

While several studies have noted species differences in the pharmacological profile of native adenosine A1 receptors (Murphy & Snyder, 1982; Ukena et al., 1986; Ferkany et al., 1986; Klotz et al., 1991), this is the first detailed study to employ an antagonist radioligand. Xanthine-based adenosine receptor antagonists exhibited higher affinity for [3H]-DPCPX binding sites in rat cortical membranes, compared with human and guinea-pig brain membranes. Although similar findings have been obtained with agonist radioligands, such as [3H]-R-PIA (Ukena et al., 1986) or [3H]-CHA (Murphy & Snyder, 1982; Ferkany et al., 1986), inter-species differences in antagonist affinity were more pronounced in the present study. The affinity of adenosine receptor agonists for [3H]-DPCPX binding sites was also higher in rat than in human brain membranes, whereas equivalent Ki values were noted in studies with agonist radioligands (Ukena et al., 1986; Ferkany et al., 1986). Indeed, the affinities of NECA and CADO were shown to be higher in human than in rat brain membranes in one study with a [3H]-CHA binding assay (Ferkany et al., 1986), although these data conflicted with earlier findings (Murphy et al., 1982). Discrepancies in agonist affinity between studies employing agonist and antagonist radioligands can be expected because of differential binding to high and low affinity agonist binding states. Thus, agonist radioligands preferentially bind to a high affinity state (Williams et al., 1986; Klotz et al., 1989), whereas [3H]-DPCPX binds to both high and low affinity states, as confirmed by the low agonist Hill coefficients noted in this and earlier studies (Burns et al., 1987; Lohse et al., 1987). However, the conflicting data concerning inter-species affinity ratios between studies in which agonist and antagonist radioligands were used remain unexplained. The effect of Gpp(NH)p on the affinity of CCPA for rat cortical [3H]-DPCPX binding sites also provides evidence for the existence of high and low affinity agonist binding states, as shown previously (Lohse et al., 1987). While the finding that this stable GTP analogue did not affect [3H]-DPCPX affinity and binding site density would appear to conflict with previous data, this may be due to the experimental conditions employed. Consistent with the present data, Gpp(NH)p effects on antagonist binding kinetics were noted in studies which included magnesium in the assay buffer (Ströher et al., 1989; Prater et al., 1992), but not in the absence of this divalent cation (Lohse et al., 1987; Olah & Stiles, 1990; Prater et al., 1992).

Perhaps the most interesting finding of the present study was that the affinity of xanthine based antagonists was species-dependent, whereas the affinity of pyrazolopyridine derivatives was similar in the four species examined. The molecular basis for species differences in the affinity of xanthine and pyrazolopyridine derivatives is intriguing, especially because the human, rat and guinea-pig cloned receptors exhibit approximately 90% sequence homology at the amino acid level (Mahan et al., 1991; Reppert et al., 1991; Libert et al., 1992; Townsend-Nicholson & Shine, 1992; Meng et al., 1994). However, this high degree of homology may not be of great importance because site directed mutagenisis studies on several members of the G protein linked receptor superfamily, including the 5-HT1B, 5-HT2, neurokinin and α2-adrenoceptors, have demonstrated that a single amino acid can modulate ligand affinity and explain species differences in pharmacological profile (Oksenberg et al., 1992; Kao et al., 1992; Link et al., 1992; Jensen et al., 1994). In the case of the adenosine A1; receptor, amino acid residues 270 and/or 277 in transmembrane domain 7 may be of critical importance (Tucker et al., 1994; Townsend-Nicholson & Scholfield, 1994). The former residue is involved in the binding of C8 substituted xanthines and N6 adenosine analogues, and the latter recognizes 5′-ribose derivatives. Interestingly, amino acid residue 270 may be largely responsible for pharmacological differences between canine and bovine adenosine A1 receptors (Tucker et al., 1994). Since the isoleucine at this position of the rat receptor is replaced by a threonine in the human and guinea-pig receptor, this could also explain the inter-species differences in the binding site affinity of C8 substituted xanthines and N6 adenosine derivatives noted in this and earlier studies. While it is tempting to speculate that residue 270 is not crucial for binding of pyrazolopyridine derivatives, thereby explaining the maintenance of affinity across species, confirmation of this hypothesis will require further pharmacological studies at the molecular level. Such studies will also need to address the importance of other amino acid residues, including histidine residues 251 and 278 (Olah et al., 1992), as well as other regions of the receptor. For example, mutagenisis studies suggest that the second extracellular loop and transmembrane domains 1–4 may be involved in ligand recognition (Olah et al., 1994; Rivkess et al., 1995). In conclusion, the present study suggests that different structural determinants may be responsible for binding of xanthine and pyrazolopyridine derivatives to the adenosine A1 receptor.

K.F. was supported by an MRC Fellowship.