Abstract
In this article, we pharmacologically characterized two naturally occurring human histamine H3 receptor (hH3R) isoforms, hH3R(445) and hH3R(365). These abundantly expressed splice variants differ by a deletion of 80 amino acids in the intracellular loop 3. In this report, we show that the hH3R(365) is differentially expressed compared with the hH3R(445) and has a higher affinity and potency for H3R agonists and conversely a lower potency and affinity for H3R inverse agonists. Furthermore, we show a higher constitutive signaling of the hH3R(365) compared with the hH3R(445) in both guanosine-5′-O-(3-[35S]thio) triphosphate binding and cAMP assays, likely explaining the observed differences in hH3R pharmacology of the two isoforms. Because H3R ligands are beneficial in animal models of obesity, epilepsy, and cognitive diseases such as Alzheimer's disease and attention deficit hyperactivity disorder and currently entered clinical trails, these differences in H3R pharmacology of these two isoforms are of great importance for a detailed understanding of the action of H3R ligands.
The histamine H3 receptor (H3R) was originally discovered in the brain on histaminergic neurons as a presynaptic autoreceptor regulating the release of histamine (Arrang et al., 1983). Subsequently, the H3R was found to regulate the release of other neurotransmitters, such as acetylcholine, dopamine, glutamate, noradrenalin, and serotonin (Schlicker et al., 1988, 1989, 1993; Clapham and Kilpatrick, 1992; Brown and Reymann, 1996). The histamine-containing cell bodies, located in the tuberomammillary nucleus of the posterior hypothalamus, project to most cerebral areas in rodent and human brain (Panula et al., 1984; Watanabe et al., 1984). Brain histamine is involved in the regulation of numerous functions of the central nervous system (CNS), including arousal, cognition, locomotor activity, autonomic and vestibular functions, feeding and drinking, sexual behavior, and analgesia (Hough, 1988; Schwartz et al., 1991; Wada et al., 1991). Moreover, H3R-specific ligands show beneficial effects in animal models of obesity, epilepsy, and cognitive diseases such as Alzheimer's disease and attention deficit hyperactivity disorder (Hancock, 2003; Passani et al., 2004; Leurs et al., 2005). Consequently, H3R antagonists are considered as potential new therapeutics and are currently undergoing clinical trails (Celanire et al., 2005).
For a good understanding of the biological effects of H3R ligands, a detailed knowledge of the molecular pharmacology of the human H3R (hH3R) is indispensable. The molecular cloning of the hH3R (Lovenberg et al., 1999) has revealed that the hH3 gene is located on chromosome 20 (20q13.32-20q13.33) and contains three introns (Cogé et al., 2001), which give rise to a large number of hH3R isoforms through alternative splicing (for review, see Hancock et al., 2003; Leurs et al., 2005). At present, there is no detailed knowledge on the pharmacological consequences of the various alternative splicing events. On the basis of the currently published information, one can conclude that the hH3R(365) is one of the most abundantly expressed hH3R isoforms next to the full-length hH3R(445) (Cogé et al., 2001; Wellendorph et al., 2002; Esbenshade et al., 2006). The hH3R(365) isoform lacks 80 amino acids in the third intracellular loop (IL3) and has been described to be nonfunctional in a cAMP, guanosine-5′-O-(3-[35S]thio) triphosphate ([35S]GTPγS) binding, and Ca2+ mobilization assay (Cogé et al., 2001) but to be functional in an R-SAT reporter assay (Wellendorph et al., 2002). For other G-protein-coupled receptors (GPCRs), the IL3 has been shown to dictate G-protein specificity (Burstein et al., 1996; Senogles et al., 2004) and bind β-arrestins (Gelber et al., 1999) and calmodulin (Turner et al., 2004). Furthermore, the carboxyl terminus of the IL3 has been shown to play a role in constitutive activity of GPCRs (Chakir et al., 2003). In view of the relative high abundance of the hH3R(365) isoform and the known importance of the IL3 loop in GPCR-mediated signal transduction, we pharmacologically characterized the hH3R(445) and the hH3R(365) isoforms in full detail in this study.
Materials and Methods
Materials. Dulbecco's modified Eagle's medium, trypsin-EDTA, penicillin, l-glutamine, and streptomycin were from Invitrogen (Breda, The Netherlands), and fetal calf serum was from Integro (Zaandam, The Netherlands). Culture dishes were from Costar (Haarlemermeer, The Netherlands). cAMP was obtained from Sigma-Aldrich Chemie B.V. (Zwijndrecht, The Netherlands). All H3R ligands were (re-) synthesized at the Vrije Universiteit Amsterdam or at Abbott Laboratories. G-418 was obtained from Duchefa Biochemie B.V. (Haarlem, The Netherlands). 3-Isobutyl-1-methylxanthine was obtained from Acros (Fischer Scientific, 's Hertogenbosch, The Netherlands). Nα-[methyl-3H]Histamine (83 Ci/mmol) was from PerkinElmer Life Sciences (Zaventem, Belgium). [3H]cAMP (40 Ci/mmol) was from Amersham (Hertogenbosch, The Netherlands). The following compounds were used: FUB322, A-304121, A-317920, A-320436, A-331440, A-349821, A-358239, and A-431404.
Cloning of the Histamine H3 Receptor Isoforms. The human histamine H3 receptor gene was cloned using human thalamus poly-A RNA (Clontech, Palo Alto, CA) with RT-PCR methods using primers designed according to the published human histamine H3 receptor gene sequences (GenBank accession no. AF140538) (Lovenberg et al., 1999). The cDNAs for the human full-length [hH3R(445)] and a shorter histamine H3 receptor isoform [hH3R(365)], with an 80-amino acid deletion from the third intracellular loop, were cloned into the pCIneo expression vector.
Analysis of H3R Isoform Expression. Tissue expression of H3R isoforms was analyzed by RT-PCR. Human mRNA (1 μg; Clontech, Mountain View, CA) was reverse-transcribed using SuperScript III First-Strand Synthesis System (Invitrogen, Carlsbad, CA) and amplified for 35 cycles of 95°C for 45 s, 60°C for 1.5 min, 72°C for 2 min, and a final extension at 72°C for 7 min using 1 unit of Platinum Taq High Fidelity DNA polymerase (Invitrogen). The 5′ forward and the 3′ reverse primers used were F754 (nucleotides 754–778) and R1406 (nucleotides 1383–1406) (Cogé et al., 2001). The PCR products were subcloned into pCR Blunt II-TOPO cloning vector (Invitrogen) and sequenced. PCR products were analyzed using agarose gels stained with ethidium bromide.
Cell Culture. The H3R isoforms were stably expressed in rat C6 glioma cells and were maintained at 37°C in a humidified 5% CO2, 95% air atmosphere in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum, 2 mM l-glutamine, 50 IU/ml penicillin, and 50 μg/ml streptomycin in the presence of 400 μg/ml G-418.
Radioligand Binding. Cells were scraped from their dishes, centrifuged (3 min, 1000 rpm), and the pellets were stored at–20°C until use. Before use, the pellets were dissolved in 50 mM Na2HPO4, pH 7.4 (for Nα-[methyl-3H]histamine and [3H]A-349821) or in 25 mM Tris, 145 mM NaCl, and 5 mM MgCl2 (for [125I]iodophenpropit) and homogenized for 2 s (40 W, Labsonic 1510). The cell homogenates (10–20 μg) were incubated for 60 min at 25°C with or without competing ligands in a total volume of 200 μl. For competition-binding experiments, 0.6 nM Nα-[methyl-3H]histamine (83.0 Ci/mmol) or 2.5 nM [125I]iodophenpropit was used. Saturation experiments were perform with various concentrations of Nα-[methyl-3H]histamine (∼0.1–20 nM), [125I]iodophenpropit (∼0.5–50 nM), or [3H]A-349821 (∼0.01–1.5 nM), and nonspecific binding was defined by 100 μM thioperamide. The incubation was terminated by rapid filtration over polyethylenimine (0.3%)-pretreated Unifilter GF/C filter plates with two subsequent washes with ice-cold 50 mM Tris-HCl, pH 7.4. Radioactivity retained on the filter was determined by liquid scintillation counting on the Microbeta Trilux with 25 μlof Microscint “O.”
GTPγS Binding Assay. Human embryonic kidney cell membranes expressing the human H3R were prepared by homogenization in cold buffer containing 50 mM Tris-HCl, pH 7.4, 5 mM EDTA, 10 mM MgCl2, and protease inhibitors. The homogenate was centrifuged two times at 40,000g for 20 min at 4°C, and the resulting pellet was resuspended in buffer containing 50 mM Tris HCl, pH 7.4, 5 mM EDTA, and 10 mM MgCl2. Glycerol and bovine serum albumin were added to a final concentration of 10% glycerol and 1% bovine serum albumin before freezing the membranes. The inverse agonist activity of H3R antagonists was determined as described previously (Krueger et al., 2005). In brief, membranes were diluted in GTPγS assay buffer (25 mM HEPES, 2.5 mM MgCl2, and 75 mM NaCl, pH 7.4), and 10 μg of membrane protein was incubated in a 96-well deep-well block in the presence of 5.0 μM unlabeled GDP, approximately 0.5 nM [35S]GTPγS, and various concentrations of H3R antagonists. Samples were subsequently incubated at 37°C for 20 min. For assays to determine antagonist activity, (R)-α-methylhistamine (30 nM) was added in addition to the assay components described above, and the samples were incubated at 37°C for 5 min. The assays were terminated by the addition of cold buffer (50 mM Tris-HCl, 75 mM NaCl, and 2.5 mM MgCl2, pH 7.6) and subsequent harvesting by vacuum filtration onto a Packard Unifilter 96-well GF/B plate (PerkinElmer Life Sciences). After extensive washing, the plates were dried, Microscint 20 was added to the samples, and the amount of bound [35S]GTPγS was determined utilizing the Topcount (PerkinElmer Life Sciences, Boston, MA). The percentage of [35S]GTPγS bound in each sample was calculated as a percentage of that bound to control samples incubated in the absence of histamine H3R ligands. Triplicate determinations were obtained at each concentration, and the data were analyzed using GraphPad Prism (GraphPad Software, San Diego, CA) to obtain EC50 or IC50 values and Hill slopes. pKb values were calculated using the generalized Cheng-Prusoff equation (Cheng and Prusoff, 1973; Leff and Dougall, 1993). The mean ± S.E.M. of at least three independent experiments is reported.
Measurement of cAMP. C6 cells stably expressing the H3R were washed once with DMEM supplemented with 25 mM HEPES (pH 7.4 at 37°C) and preincubated in the same medium for 30 min at 37°C. Thereafter, 5 × 103 cells/50 μl was added per well to a 96-well plate containing the respective ligands in 50 μl of DMEM supplemented with 0.3 mM 3-isobutyl-1-methylxanthine and 1 μM forskolin (final concentration, 0.5 μM). After 10 min, incubations were terminated by the addition of 20 μl of 0.3 mg/ml saponin to each well to lyse the cells. Subsequently, cAMP levels were determined with a competitive protein kinase A (PKA) binding assay, as described. Protein kinase A-containing fraction was isolated from bovine adrenal glands, as described previously (Wieland et al., 2001). In brief, a PKA-containing fraction was isolated from bovine adrenal glands, which were homogenized in 10 volumes of 100 mM Tris-HCl, 250 mM NaCl, 10 mM EDTA, 0.25 M sucrose, and 0.1% 2-mercaptoethanol (pH 7.4 at 4°C) and centrifuged for 60 min at 30.000g at 4°C. The supernatant, containing PKA, was carefully recovered and frozen in 1-ml aliquots at–80°C. Before use, the PKA was diluted 12-fold in ice-cold phosphate-buffered saline (PBS) and kept on ice. To each well, 20 μl of 0.6 nM [3H]cAMP in PBS (48.0 Ci/mmol) and 60 μlof PKA in PBS was added. After 30 min, the reaction was terminated by filtration over Unifilter GF/B filter plates with two subsequent washes with 200 μl of ice-cold 50 mM Tris-HCl, pH 7.4. Retained radioactivity was determined by liquid scintillation counting on the Microbeta Trilux with 50 μl of Microscint “O.” The amount of cAMP in each sample was calculated with GraphPad Prism version 4.01 for Windows (GraphPad Software), using a generated standard cAMP curve (0.1 mM–10 pM).
Statistical Analysis. Statistical analyses were performed using GraphPad Prism version 4.01 for Windows (GraphPad Software). Differences among means were evaluated by analysis of variance, followed by the Dunnett's post test. For all analyses, the null hypothesis was rejected at the 0.05 level.
Results
Differential Expression in the Central Nervous System. The relative expression of the hH3R(445) and hH3R(365) isoforms was assessed by RT-PCR in several areas of the human CNS. For the hH3R(445), high expression was found in the cerebellum and the caudate, and moderate expression was found in the hypothalamus, cerebrum, and thalamus (Fig. 1). Low expression was found in all other tested regions, and very low expression was found in the spinal cord. In most of the tested regions, the hH3R(365) was found to be approximately 1.4-fold higher expressed than the hH3R(445). The biggest differences in expression for the two isoforms was found in the regions where the hH3R(445) was higher expressed than the hH3R(365), such as the caudate (3.5-fold), corpus callosum (2.8-fold), and spinal cord (2.2-fold).
Saturation Analysis Reveals Distinct Populationsfor the hH3R Isoforms. The hH3R isoforms (445 and 365) stably transfected into rat glioma C6 cells were characterized by saturation analysis with the agonist Nα-[methyl-3H]histamine and with the inverse agonists [125I]iodophenpropit (Jansen et al., 1994) and [3H]A-349821 (Witte et al., 2006). Saturation binding experiments with Nα-[methyl-3H]histamine revealed a single high-affinity binding site for both isoforms with a similar maximal number of binding sites [hH3R(445), Bmax = 630 ± 100 fmol/mg protein, n = 7; and hH3R(365), Bmax = 670 ± 40 fmol/mg protein, n = 4]. Yet, for the radioligand Nα-[methyl-3H]histamine, a significantly lower affinity (p < 0.001) was measured for the hH3R(445) compared with the hH3R(365) [Kd values of 0.81 ± 0.07 nM (n = 7) and 0.33 ± 0.04 nM (n = 5) were obtained for hH3R(445) and hH3R(365), respectively; see Fig. 2, A and D]. Both hH3R isoforms exhibited similar affinities for [125I]iodophenpropit [hH3R(445), Kd = 3.2 ± 2 nM; and hH3R(365), Kd = 3.1 ± 2 nM], but a significant difference (p < 0.05) was found for the maximal number of binding sites [hH3R(445), Bmax = 2000 ± 500 fmol/mg protein, n = 3; and hH3R(365), Bmax = 830 ± 20 fmol/mg protein, n = 3] (Fig. 2, A and D). As observed for [125I]iodophenpropit, the affinity values for the H3R inverse agonist [3H]A-349821 did not differ between the two isoforms [hH3R(445), Kd = 0.081 ± 0.01 nM, n = 4; and hH3R(365), Kd = 0.10 ± 0.02 nM, n = 4], but a significant difference was again found for the maximal number of binding sites [hH3R(445), Bmax = 1800 ± 100 fmol/mg protein, n = 4; and hH3R(365), Bmax = 780 ± 100 fmol/mg protein, n = 4] (Fig. 2, C and D).
G-Protein Uncoupling AffectsNα-[methyl-3H]Histamine Binding to hH3R(445). To study the effect of G-proteins on the binding characteristics of the two hH3R isoforms, saturation binding experiments were performed under conditions that would prevent G-protein coupling. Coincubation of 0.1 mM GDP in the saturation binding experiment resulted in a significant 2.7-fold inhibition (p < 0.05) of the maximal number of binding sites of Nα-[methyl-3H]histamine to hH3R(445)-expressing membranes but had no effect on the binding to hH3R(365)-expressing membranes. The affinity of both isoforms for Nα-[methyl-3H]histamine was unaltered in the presence of 0.1 mM GDP [hH3R(445), Kd = 1.2 ± 0.05 nM, n = 3; and hH3R(365), Kd = 0.4 ± 0.09 nM, n = 3]. In hH3R(445)- or hH3R(365)-expressing membranes, neither the maximal number of binding sites nor the affinity for [3H]A-349821 was affected by coincubation of 0.1 mM GDP (Fig. 2E). In homologous competition experiments with Nα-[methyl-3H]histamine at hH3R(445)-expressing membranes, coincubation with GDP, GTPγS, or pretreatment with 200 ng/ml pertussis toxin for 16 h all had a similar effect on the maximal specific binding of Nα-[methyl-3H]histamine. None of these treatments had an effect on the maximal specific binding of Nα-[methyl-3H]histamine to hH3R(365)-expressing membranes (Fig. 2F).
Pharmacological Profile of Radioligand Binding to the hH3R(445) and hH3R(365) Isoforms. A series of imidazole (compounds 1–14 and 22–28) and nonimidazole (compounds 15–21) containing H3R ligands were subsequently tested in a heterologous competitive binding assay with Nα-[methyl-3H]histamine and [125I]iodophenpropit with membranes expressing either the hH3R(445) or the hH3R(365) (see Table 1). The competition binding curves for all compounds were best fitted according to a single binding site model and shown to have Hill slopes close to unity for both radioligands (Fig. 3, A–D). Equilibrium dissociation constants of the agonists correlate highly for competitive binding with either Nα-[methyl-3H]histamine (r2 = 0.82) or [125I]-iodophenpropit (r2 = 0.85) (Fig. 3, E and F). Agonists exhibit a higher affinity for the hH3R(365) compared with the hH3R(445), on average 3.4- or 55-fold when determined with Nα-[methyl-3H]histamine or [125I]iodophenpropit, respectively (Fig. 3E). Equilibrium dissociation constants of the inverse agonists correlate highly as well (r2 = 0.95 for Nα-[methyl-3H]histamine and r2 = 0.91 for [125I]iodophenpropit) but display an opposite preference for the hH3R isoforms and have a 2 to 3-fold higher affinity for the hH3(445) isoform (Fig. 3F). In the correlation plots from the heterologous competitive binding assay with Nα-[methyl-3H]histamine or [125I]iodophenpropit, some compounds are outside the 95% confidence interval of the linear regression for both the agonist and inverse agonist and therefore are statistical outliers (Fig. 3, E and F).
Pharmacological Profile of the hH3R(445)- and hH3R(365)-Induced [35S]GTPγS Binding. A series of compounds was also tested in a functional [35S]GTPγS binding assay with membranes expressing either the hH3R(445) or the hH3R(365) isoform (see Table 2). The dose-response curves for all compounds were best fitted according to a single binding site model and showed to have Hill slopes close to unity (Fig. 4, A–E). Potencies of the agonists correlated highly (r2 = 0.96) and were found to be, on average, 4.6-fold more potent at the hH3R(365) (Fig. 4F). However, the maximal increase in [35S]GTPγS binding for full agonists was approximately 80% lower at the hH3(365) (1.2-fold increase of basal) compared with the hH3R(445) (2.2-fold increase of basal) (Fig. 4, A–C).
Potencies of the tested inverse agonist correlate highly as well (r2 = 0.99) but were found to be, on average, 2.6-fold less potent at the hH3R(365) (Fig. 4F). Interestingly, nonimidazole ligands, like A-349821, showed a significantly (p < 0.01) higher inhibition of basal [35S]GTPγS binding for the hH3R(365) compared with the hH3R(445) (Fig. 4, D and E).
Pharmacological Profile of the Forskolin-Induced cAMP Levels by the hH3R(445) and the hH3R(365). Histamine H3R-specific compounds were also tested for their modulation of cAMP levels in forskolin (0.5 μM)-stimulated C6 cells expressing either the hH3R(445) or the hH3R(365) (see Table 3). All tested agonists and inverse agonists potencies correlate highly (r2 = 0.91 and 0.88, respectively). Agonists are approximately 35-fold more potent at the hH3R(365) (Fig. 5F). Similar to the [35S]GTPγS binding assay, full agonists are, however, more efficacious at the hH3R(445) (80 and 44% inhibition of the forskolin-induced cAMP production, respectively; see Fig. 5, A–C). Conversely, inverse agonists are approximately 14-fold less potent at the hH3R(445) but appear more efficacious (Fig. 5, D and E). Especially, nonimidazole compounds, like A-349821, showed increased efficacy at the hH3R(365) (3.0-fold increase over basal) compared with the hH3R(445) (1.5-fold increase over basal) (Fig. 5D).
The hH3R(365) Is More Constitutively Active Than the hH3R(445). The observed differences in agonist/inverse agonist efficacies are good indications of substantial differences in the levels of constitutive activity of the two hH3R isoforms. To investigate this issue, the amount of [35S]GTPγS formed over a time course of 10 min was measured on C6 cell membranes expressing either the hH3R(445) or the hH3R(365). The C6 parental cell line showed a gradual increase in the amount of [35S]GTPγS bound, and no modulation by the specific H3R ligands (R)-α-methylhistamine and A-349821 was observed (Fig. 6A). Basal increase in [35S]GTPγS binding in C6 cells expressing the hH3R(445) was comparable with the parental cell line, whereas for the hH3R(365), the basal increase was significantly higher (Fig. 6, A–D). Stimulation of the hH3R(445) with the H3R agonist (R)-α-methylhistamine led to a steep increase in [35S]GTPγS binding over time, whereas the inverse agonist, A-349821, slightly inhibited the [35S]GTPγS binding (Fig. 6B). On membranes expressing the hH3R(365), the increase in [35S]GTPγS binding induced by (R)-α-methylhistamine was moderate, whereas the effect of A-349821 was more pronounced compared with its effect at the hH3R(445) (Fig. 6, A–D).
In addition, forskolin (0.5 μM)-induced cAMP levels were 2.3-fold (p < 0.001) lower in C6 cells expressing the hH3R(365) compared with C6 cells expressing hH3R(445) (Fig. 6E). The H3R agonist (R)-α-methylhistamine inhibited the forskolin (0.5 μM) induced to a similar absolute level in both hH3R(445)- and hH3R(365)-expressing C6 cells. Likewise, stimulation of C6 cells expressing either the hH3R(445) or the hH3R(365) with the H3R inverse agonist A-349821 increased the cAMP levels to a similar absolute level of cAMP.
Application of the Cubic Ternary Model to Explain the Observed Properties of the hH3R(365) and hH3R(445). We used the cubic ternary complex (CTC) model (Weiss et al., 1996a,b,c) to examine whether the observed differences in affinities and potencies, the maximal number of binding sites of agonist and inverse agonist and the effect of GDP on the maximal number of binding sites could all be explained by an increase in GPCR constitutive activity, represented in the CTC model by the equilibrium constant of the receptor (Kact; see Fig. 7A).
To examine the maximal number of binding site as found with the H3R radioligands (Fig. 2A), the agonist radioligand Nα-[methyl-3H]histamine was assumed to bind the Ra and RaG state of the receptor, to promote receptor activation (α > 1), and to facilitate G-protein coupling only for the receptor in its active form (δ > 1). The inverse agonist radioligand [3H]A-349821 was assumed to bind the Ri and RiG state of the receptor, to inhibit receptor activation (α < 1), and to facilitate G-protein coupling only for the receptor in its inactive form (δ < 1). Furthermore, to examine whether the difference between the two isoforms could be due to a difference in Kact, the hH3R(445) and the hH3R(365) were assumed to have a Kact < 1 and Kact = 1, respectively. Hereby, the hH3R(445) exists predominantly in the Ri conformation, and the hH3R(365) has no preference for either the Ra of Ri conformation. Experimentally, we found an approximately 3-fold difference in maximal number of binding sites of [3H]A-349821 for the two hH3R isoforms (Fig. 2D). To take this observation into account in the CTC model, the Rtot of the hH3R(365) was assumed to be 3-fold lower compared with the hH3R(445).
Subsequently, the maximal number of binding sites for the agonist, Nα-[methyl-3H]histamine (NαMH), and inverse agonist, [125I]iodophenpropit (IPP), were simulated in the presence (Gtot = 1) and absence (Gtot = 0) of G-proteins. This simulation revealed that only the maximal number of binding sites for the agonist, reflected by the receptor states LRa and LRaG, on the hH3R(445) is affected by the removal of G-proteins (Fig. 7B). Furthermore, the maximal number of binding sites for the hH3R(445) recognized by the inverse agonist, reflected by the receptor states LRi and LRiG, was found to be approximately three times higher, whereas for the hH3R(365), agonist and inverse agonists give rise to the same maximal number of binding sites (Fig. 7B). To account for the differences in ligand affinity and potency found experimentally for the hH3R(445) and the hH3R(365), we calculated the apparent affinity constant when G-proteins are not limiting (KAapp) (Weiss et al., 1996b).
From this equation (eq. 1), it follows that for the hH3R(445), the log(KAapp) for agonists is 10.2 and for inverse agonists, 10.0; in the case of the hH3R(365), the log(KAapp) was increased for agonists to 11.0 and decreased for inverse agonists to 9.0. We defined a partial agonist to have an α > 1 but lower than α of a full agonist, e.g., the ligand is able to activate the receptor but to a lesser extent than a full agonist. The difference for partial agonists in the KAapp for the two isoforms decreases, and with an α ≈ 1, the log(KAapp) = 10 for both the hH3R(445) and the hH3R(365).
Furthermore, we simulated radioligand competition and dose-response curves for the hH3R(445) and hH3R(365) in a G-protein-dependent manner (Monczor et al., 2003). For the competition curves, we assumed the radioligands and the competing ligands to have properties as described above. Subsequently, the concentration of a ligand (L) was varied, and the generated curves were plotted as percentage of the Bmax for both hH3R isoforms and radioligands (Fig. 7, C and D).
For simulation of dose-response curves, we assumed that both the LRaG and RaG give rise to a receptor response (as described by Chen et al., 2000). Subsequently, the following equation (eq. 2) was used to simulate the dose-response curves (Fig. 7E).
Furthermore, when [L] = 0, the above equation describes the basal signaling of the receptor, and this was found to be 19-fold increased for the hH3R(365) compared with the hH3R(445).
Discussion
The cloning of the hH3R(445) by Lovenberg et al. (1999) has subsequently led to the discovery of a large number of hH3R isoforms (Hancock et al., 2003; Leurs et al., 2005). In this study, we pharmacologically characterize the two most abundantly occurring hH3R isoforms, the hH3R(445) and the hH3R(365) (Cogé et al., 2001; Wellendorph et al., 2002; Esbenshade et al., 2006). Using an RT-PCR approach, we confirmed the abundant expression of the hH3R(445) and the hH3R(365) isoforms and demonstrated differential isoform expression in various human brain areas. In most regions, the hH3R(365) was slightly higher expressed. However, in brain regions with the highest hH3R(445) expression levels, such as caudate, corpus callosum, and the spinal cord, the largest differences between these two isoforms were found.
The hH3R(365) lacks 80 amino acids in the IL3 in comparison with the hH3R(445) and was found to have different functionality. In particular, we show that the hH3R(365) isoform stably expressed in C6 cells gives rise to a higher agonist-independent inhibition of forskolin-induced cAMP levels and to an increased basal [35S]GTPγS formation. These findings are consistent with an increased constitutive receptor activity of the hH3R(365) compared with the hH3R(445). Basal GPCR signaling highly depends on the receptor expression levels but is also regulated by the expression levels of its cellular signaling partners (Milligan, 2003). However, the observed increase in constitutive signaling of the hH3R(365) cannot be explained by a higher level of receptor expression because it was found to have either the same (Nα-[methyl-3H]histamine) or lower ([3H]A-349821 or [125I]iodophenpropit) number of radioligand binding sites as observed for the hH3R(445). In addition, both isoforms were stably expressed using the same parental C6 cell line, and differences in cellular content, like G-protein expression, are therefore unlikely.
Previously, agonist-induced activation of the hH3R(365) was not found in several signal transduction assays, like the Gαi-mediated inhibition of cAMP formation, [35S]GTPγS binding, or Ca2+ mobilization (Cogé et al., 2001), but in another study, the hH3R(365) could be activated by agonists in a reporter assay (Wellendorph et al., 2002). In the present study, we show that under our experimental conditions, H3R agonists can activate the hH3R(365) isoform, resulting in an increase in [35S]GTPγS binding and the expected inhibition of forskolin-induced cAMP. The apparent discrepancy between the results of Coge et al. and the present study might be due to the high constitutive activity of the hH3R(365), which makes it difficult to measure agonist-mediated responses. Interestingly, the study of Coge et al. does not report on the functional effects of H3R inverse agonists on the hH3R(365).
The hH3R(365) displays higher potencies and affinities for agonists and likewise lower potencies and affinities for inverse agonists, consistent with an increase in constitutive activity. These findings are in line with previous observations with the rat H3RB and rat H3RC isoforms, which have a 32- or 48-amino acid deletion in the IL3 loop, respectively, and also display a slightly higher affinity and potency for agonists compared with the full-length rat H3RA isoform (Morisset et al., 2000; Drutel et al., 2001). Interestingly, partial agonists have a higher affinity and potency at the hH3R(365) as well but to a less extent than full agonists. The observed statistical outliers in the binding affinity correlation plots, in fact, appear to be partial agonists with intrinsic activities between 0.0 and 0.4 or 0.0 and 0.7 in the correlation plots for Nα-[methyl-3H]histamine or [125I]iodophenpropit, respectively (Fig. 3, E and F; Table 3). This is in good agreement with the CTC model simulations, where we found that the α of a compound correlates with the difference in apparent affinity (KAapp) for the two isoforms, e.g., a decrease in α leads to a smaller difference in KAapp for the two hH3R isoforms. Interestingly, agents that prevent G-protein coupling were found to affect the maximal number of binding sites of Nα-[methyl-3H]histamine only for the hH3R(445) but not for the hH3R(365). In the present study, the binding of the inverse agonists was unaffected by a treatment that would uncouple the receptor from the G-protein. This seems in contrast with earlier studies (Witte et al., 2006) but is most likely due to increased ionic strength of the incubation buffers, which is known to shift the equilibrium of the receptor to the inactive conformation (Costa et al., 1990). To account for these observations, the CTC model was used to examine whether a higher Kact of a constitutively active receptor might be able to explain these observations. Indeed, the CTC model can account for the observed different maximal number of binding sites for the different radioligands and confirmed that the number of agonist-labeled sites was affected only at the hH3R(445) when [Gtot] = 0. Furthermore, an increase of the Kact resulted in the experimentally observed increase of potency and affinity of agonists and decreased potency and affinity of inverse agonists at the hH3R(365) compared with the hH3R(445). The CTC model also accounted for the intermediate shift in affinity of partial agonists, showing that the preference over the hH3R(365) is dependent on intrinsic activity of the compounds. We used the CTC model because other “simpler” models, such as the extended ternary complex model, could not explain the effect of GDP. In this model, the removal of the G-protein affects predominantly the receptor with the biggest Kact, which is not in accordance with our experimental observations.
In summary, our experimental and CTC modeling data indicate that the hH3R(365) isoform mainly functions in a constitutive manner. These data indicate that the 80-amino acid stretch in the third intracellular loop either imparts a negative constraint on the GPCR activation directly or is involved in the interaction of intracellular proteins that inhibit H3R activation. Future studies should address the role of the 80-amino acid stretch in the IL3 on H3R receptor function in more detail. Because the H3R is one of the few examples for which GPCR constitutive activity has been shown to be prominent under in vivo conditions (Morisset et al., 2000; Wieland et al., 2001), our present results suggest that this effect might be mainly mediated by the hH3R(365) isoform. The nonimidazole compounds show a higher negative intrinsic activity at the hH3R(365) compared with classical imidazole containing compounds like thioperamide. In combination with the potential differential expression in the CNS, this might lead to specific effects of these nonimidazole inverse agonists in brain areas where the hH3R(365) is higher expressed than the hH3R(445).
Footnotes
-
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
-
doi:10.1124/jpet.107.127639.
-
ABBREVIATIONS: H3R, histamine H3 receptor; CNS, central nervous system; h, human; [35S]GTPγS, guanosine-5′-O-(3-[35S]thio) triphosphate; IL3, intracellular loop 3; GPCR, G-protein-coupled receptor; [3H]A-349821, {4′-[3-((2R, 5R)-2,5-dimethyl-pyrrolidin-1-yl)-propoxy]-biphenyl-4-yl}-morpholin-4-yl-methanone; RT, reverse transcriptase; PCR, polymerase chain reaction; DMEM, Dulbecco's modified Eagle's medium; PKA, protein kinase A; PBS, phosphate-buffered saline; CTC, cubic ternary complex; NαMH, Nα-[methyl-3H]histamine; IPP, [125I]iodophenpropit; FUB322, 3-(1H-imidazol-4-yl) propyl-di(p-fluorophenyl)-methyl ether hydrochloride; A-304121, (R)-2-amino-1-{4-[3-(4-cyclopropanecarbonyl-phenoxy)-propyl]-piperazin-1-yl}-propan-1-one; A-317920, furan-2-carboxylic acid, ((R)-2-{4-[3-(4-cyclopropanecarbonyl-phenoxy)-propyl]-piperazin-1-yl}-1-methyl-2-oxo-ethyl)-amide; A-320436, furan-2-carboxylic acid, ((R)-2-{4-[3-(4′-cyano-biphenyl-4-yloxy)-propyl]-[1,4]diazepan-1-yl}-2-oxo-1-thiazol-4-ylmethyl-ethyl)-amide; A-331440, 4′-[3-((R)-3-dimethylamino-pyrrolidin-1-yl)-propoxy]-biphenyl-4-carbonitrile; A-349821, {4′-[3-((2R, 5R)-2,5-dimethyl-pyrrolidin-1-yl)-propoxy]-biphenyl-4-yl}-morpholin-4-yl-methanone; A-358239, 4-{2-[2-((R)-2-methyl-pyrrolidin-1-yl)-ethyl]-benzofuran-5-yl}-benzonitrile; A-431404, (4-fluoro-phenyl)-{2-[2-((R)-2-methyl-pyrrolidin-1-yl)-ethyl]-benzofuran-5-yl}-methanone.
-
↵1 Current affiliation: Department of Metabolic Diseases, Boehringer Ingelheim Pharma GmbH and Co. KG, Biberach, Germany.
- Received June 22, 2007.
- Accepted July 25, 2007.
- The American Society for Pharmacology and Experimental Therapeutics