点击显示 收起
Leiden/Amsterdam Center for Drug Research, Faculty of Sciences, Department of Medicinal Chemistry, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands (M.B., A.J., F.v.d.M., R.A.B., R.L.)
UCB S.A. Pharma Sector, Braine-l'Alleud, Belgium (M.G., P.C.)
Abstract
We tested several histamine H1 receptor (H1R) and antagonists for their differences in agonists binding affinities between human and guinea pig H1Rs transiently expressed in African green monkey kidney (COS-7) cells. Especially, the bivalent agonist histaprodifen-histamine dimer (HP-HA) shows a higher affinity for guinea pig than for human H1Rs. Based on the structure of HP-HA, we have further identified VUF 4669 [7-(3-(4-(hydroxydiphenylmethyl)piperidin-1-yl)propoxy)-4-oxochroman-2-carboxylic acid] as a guinea pig-preferring H1R antagonist, demonstrating that the concept of species selectivity is not limited to agonists. To delineate the molecular mechanisms behind the observed species selectivity, we have created mutant human H1Rs in which amino acids were individually replaced by their guinea pig H1R counterparts. Residue Asn84 (2.61) in transmembrane domain (TM) 2 seemed to act as a selectivity switch in the H1R. Molecular modeling and site-directed mutagenesis studies suggest that Asn84 interacts with the conserved Tyr458 (7.43) in TM7. Our data provide the first evidence that for some H1R ligands, the binding pocket is not only limited to TMs 3, 4, 5, and 6 but also comprises an additional pocket formed by TMs 2 and 7.
The biogenic amine histamine exerts its effects through binding and activation of four G protein-coupled receptors (GPCRs), the H1, H2, H3, and H4 receptors. The H1 receptor (H1R) regulates inflammatory and allergic responses and is successfully targeted by various drugs. H1R antagonists have been on the market since 1942 for the treatment of allergies, and newer, nonsedating second generation H1R antagonists are still the medication of choice to relief certain allergic symptoms (Hill et al., 1997).
In contrast to the development of various potent H1R antagonists, the synthesis of selective and potent H1R agonists has not achieved the same success (Hill et al., 1997). Only in 1995, 2-(3-trifluoromethylphenyl)histamine was discovered as the first selective H1R agonist with a potency equal to histamine as determined by the H1R-mediated guinea pig ileum contractions (Leschke et al., 1995; Zingel et al., 1995). Recently, Elz et al. (2000) synthesized a series of compounds constituting a new class of highly active H1R agonists, the histaprodifens. Histaprodifen combines a histamine moiety linked at the 2-position with an ,-diphenylalkyl substituent, a characteristic of the H1R antagonist pharmacophore (ter Laak et al., 1995; Zhang et al., 1997). Based on this new H1R agonist, "dimeric" histaprodifens were subsequently developed, consisting of a histaprodifen moiety linked at the N position to another histamine moiety, for example (histaprodifen-histamine dimer, HP-HA) (Menghin et al., 2003). Compared with histamine, the potency of HP-HA is reported to be 36- to 56-fold and 630-fold higher on guinea pig isolated ileum and trachea, respectively (Christophe et al., 2003; Seifert et al., 2003).
Contrary to the potencies at either the guinea pig ileum and trachea or rat aorta (Elz et al., 2000; Christophe et al., 2003; Seifert et al., 2003), the potencies of various histaprodifen analogs (histaprodifen, MeHP, HP-HA, and HP-HP) at human H1Rs are at best similar to the potency of the endogenous ligand histamine (Seifert et al., 2003; Bruysters et al., 2004), indicating a potential species difference at the level of the H1R recognition of these H1R agonists.
In aminergic GPCRs, including the H1R, the ligand-binding pocket is thought to reside in a hydrophilic cleft formed by the seven transmembrane domains (TMs). Within the third TM (TM3), an aspartate (Asp) residue is a conserved feature among these aminergic GPCRs and is likely to make a direct contact with the protonated amine of aminergic ligands (Shi and Javitch, 2002). Indeed, in the human H1R Asp107 in TM3 (residue 3.32 according to the Ballesteros-Weinstein numbering) is part of the binding pocket of both H1R agonists and antagonists (Ohta et al., 1994; Moguilevsky et al., 1998; Nonaka et al., 1998; Bruysters et al., 2004). Several additional amino acids in TM5 and TM6 are part of the H1R binding pocket of histamine: Lys191 (5.39) (Leurs et al., 1995; Moguilevsky et al., 1998; Wieland et al., 1999; Gillard et al., 2002; Bruysters et al., 2004), Asn198 (5.46) (Leurs et al., 1994; Ohta et al., 1994; Moguilevsky et al., 1995; Bruysters et al., 2004), and Phe435 (6.55) (Bruysters et al., 2004) are considered to accommodate the imidazole ring of histamine. The H1R antagonist binding pocket stretches deeper into the receptor protein and comprises the aromatic amino acids Trp158 (4.56) (Wieland et al., 1999) and Phe432 (6.52) (Wieland et al., 1999; Bruysters et al., 2004).
Recently, we studied the binding pocket of several histaprodifen analogs in the human H1R (Bruysters et al., 2004). We demonstrated that histamine and the histamine moiety of histaprodifens bind to the human H1R in a similar orientation. Although the diphenylalkyl system of histaprodifen interacts with the H1R in an "antagonistic binding mode", i.e., interacting with Phe432 (6.52) in TM6 (Bruysters et al., 2004), no interactions with Lys191 (5.39) and Phe435 (6.55) were found. Again, the interaction with both Asp107 (3.32) proved crucial. Although Asn198 (5.46) did not affect histaprodifen affinity, it seemed pivotal for agonist-induced activation of the hH1R. An interaction between Asn198 (5.46) and histaprodifen was therefore suggested (Bruysters et al., 2004).
We explored in this study the molecular basis of the observed species differences between human and guinea pig H1Rs by a combined approach of molecular modeling and site-directed mutagenesis. We reevaluated several H1R agonists and antagonists for their differences in affinity between human and guinea pig H1Rs by [3H]mepyramine displacement studies. Based on our knowledge of the H1R binding site of the histaprodifens and the high (93%) level of sequence homology within the TM domains of the human and guinea pig H1Rs, we extended our approach to mutant human H1Rs in which selected amino acids were individually replaced by their guinea pig H1R counterparts. Using this strategy, we identified Asn84 (2.61) in TM2 as the molecular basis for the observed species selectivity of certain H1R ligands and discuss the implications of these findings for future drug design.
Materials and Methods
Chemicals. Chloroquine diphosphate, DEAE-dextran (chloride form), histamine dihydrochloride, mepyramine (pyrilamine maleate), astemizole, ketotifen fumarate, 8R-lisuride, and terfenadine were purchased from Sigma-Aldrich (Bornem, Belgium). Oxatomide was obtained from MP Biomedicals (Zoetermeer, The Netherlands). Fexofenadine was purchased from Ultrafine Chemicals (Manchester, UK). VUF 4669 and VUF 8401 were synthesized at the Vrije Universiteit Amsterdam (Amsterdam, The Netherlands). Cetirizine dihydrochloride (Zyrtec) and loratadine were synthesized at UCB S.A. (Braine l'Alleud, Belgium).
Gifts of 2-(3-trifluoromethylphenyl)histamine dihydrogen maleate, histaprodifen (2-[2-(3,3-diphenylpropyl]imidazol-4-yl)ethanamine dihydrogen maleate), methylhistaprodifen (N-methyl-histaprodifen dihydrogen oxalate), histaprodifen-histaprodifen dimer trihydrogenoxalate and histaprodifen-histamine dimer (N-[2-(1H-imidazol-4yl)-ethyl]-histaprodifen) trihydrogenoxalate) (Dr. W. Schunack, Free University, Berlin, Germany), the cDNA encoding the human H1R (Dr. H. Fukui, University of Tokushima, Tokushima, Japan), and the expression vector pcDEF3 (Goldman et al., 1996) (Dr. J. Langer, Robert Wood Johnson Medical School, Piscat-away, NJ) are greatly appreciated.
Cell culture media, penicillin, streptomycin and fetal bovine serum (FBS) were purchased from Cambrex Bio Science Verviers S.p.r.l. (Verviers, Belgium). Cell culture plastics were obtained from Greiner Bio-one (Wemmel, Belgium). [3H]Mepyramine (20 Ci/mmol) was from Amersham Biosciences Inc. (Roosendaal, The Netherlands).
Numbering Scheme of GPCRs. The indexing method introduced by Ballesteros and Weinstein (1995) was used throughout to identify amino acids in the TM regions. Each residue is identified by two numbers: the first number corresponds to the helix (1-7) in which the residue is located, and the second number indicates its position relative to the most conserved amino acid in that helix, arbitrarily assigned to 50. Numbers depicted in superscript correspond to the number of the amino acid in the human H1R.
Site-Directed Mutagenesis. The cDNA encoding the human H1R (Fukui et al., 1994) was subcloned in the pAlter plasmid (Promega, Madison, WI), and point mutations were created according to manufacturer's protocol (Altered Sites II; Promega). cDNAs of mutant and wild-type receptors were subcloned into the expression plasmid pcDEF3 (Goldman et al., 1996). Sequences were verified by DNA sequencing using the dideoxy chain termination method.
Cell Culture, Transfection, and Membrane Preparation. COS-7 African green monkey kidney cells were maintained at 37°C in a humidified 5% CO2, 95% air atmosphere in Dulbecco's modified Eagle's medium containing 50 IU/ml penicillin, 50 e/ml streptomycin, and 10% (v/v) fetal bovine serum. COS-7 cells were transiently transfected using the DEAE-dextran method as described previously (Bakker et al., 2001), by using 5 e of plasmid DNA per million cells. Two days after transfection, cells were detached by scraping and were harvested by centrifugation. Cell pellets were resuspended in ice-cold water, lysed by repetitive freezing/thawing, and frozen in liquid nitrogen. The obtained crude cell homogenates were stored at -80°C until further use.
H1R Binding Studies. The COS-7 cell homogenates (5 e) were incubated for 60 min at 30°C in 500 e of binding buffer (50 mM Na2/K-phosphate buffer, pH 7.4) containing 3 nM [3H]mepyramine. The nonspecific binding was determined in the presence of 10 e cetirizine. The incubations were stopped by rapid dilution with ice-cold binding buffer. The bound radioactivity was separated by filtration through Whatman GF/C filters (Whatman, Vel, Belgium) that had been treated with 0.1% polyethylenimine. Filters were washed four times with binding buffer, and radioactivity retained on the filters was measured by liquid scintillation counting.
Molecular Modeling. Our H1R homology model was obtained using the bovine rhodopsin crystal structure (Protein Data Bank entry 1L9H ; Okada et al., 2002) as the template. Side chains were added using the SCWRL program (Canutescu et al., 2003). Water molecules present in the rhodopsin structure were retained, and their heavy atoms were kept fixed during all minimizations and molecular dynamic runs. The position of TM3 was manually changed with regard to the rhodopsin structure to avoid a clash between the top of TM3 and TM2 (Lopez-Rodriguez et al., 2002). Given the presence of the H1R-specific Trp158 (4.56), TM3 could not be put into the position as found by molecular modeling studies on the 5HT1A receptor. Therefore, we assumed an intermediate position between the location in the crystal structure of rhodopsin and the proposed location in the 5HT1A receptor model. Short minimization runs were performed (5000 iterations using steepest descent) to refine the initial model. All minimizations were carried out while fixing the C atoms to their initial positions.
Ligands were docked in the wild-type receptor using the automated docking procedure GOLD version 2.1 (Jones et al., 1997) applying default parameters. The complex of the A84S mutant receptor with the ligand was obtained by changing the appropriate residue in the WT receptor-ligand model to its guinea pig homolog. The obtained ligand-receptor complexes were used as input structures for further minimization and molecular dynamic studies. First, the position of the ligand is optimized by fixing the receptor except for the residues involved in ligand binding. Restraints were gradually released before final submission of the resulting complex to two simulated annealing runs at 600 K, each followed by cooling to 200 K before final minimization. In the first round of the simulated annealing run (2500-step initialization, 5000-step production, 5000-step cooling), the C atoms of the receptor are fixed to their position as is the ligand. In the second round (15,000-step production, 5000-step cooling), the ligand is released and free to move. All minimizations and molecular dynamics simulations were performed using Discover (Accelrys, San Diego, CA).
Analytical Methods. Protein concentrations were determined according to Bradford (1976), using bovine serum albumin as a standard. Binding data were evaluated by a nonlinear, least-squares curve-fitting procedure using GraphPad Prism 4 (GraphPad Software, Inc., San Diego, CA). Obtained pKi, pEC50, and Kd values are expressed as mean ± S.E.M. of at least three independent experiments. Statistical analyses were carried out by nonpaired Student's t test. P values <0.05 were considered to indicate a significant difference (P < 0.05, P < 0.01, and P < 0.001). Despite significance, differences in pKi values are only considered relevant when the difference is at least 0.3 logunits.
Results and Discussion
Evaluation of Species Selectivity of H1R Ligands. Using displacement of [3H]mepyramine binding to guinea pig or human H1Rs transiently expressed in COS-7 cells, we determined the affinity of a series of H1R antagonists [cetirizine (Zyrtec), ketotifen (Zaditor), loratadine (Claritin), oxatomide (Tinset), fexofenadine (Allegra), astemizole, terfenadine, and mepyramine]. As shown in Fig. 1, none of the tested H1R antagonists (open symbols) showed any preference, i.e., a difference in pKi exceeding 0.3 log units (dotted lines), for binding to the guinea pig H1R over the human H1R, corroborating recent findings by Seifert et al. (2003). We also determined the binding affinities of several H1R agonists (closed symbols) for both human and guinea pig H1Rs. Again, the general rank order of affinities is shared between human and guinea pig H1Rs, with histamine having the lowest and the recently characterized partial agonist 8R-lisuride (Bakker et al., 2004) having the highest H1R affinity. Considering all tested agonists and antagonists, we observed a linear correlation (r2 = 0.96) between human and guinea pig H1R affinities over an affinity range of almost six decades. No species differences were observed between the human and the guinea pig H1R for the affinities of the endogenous ligand histamine or the synthetic agonists histaprodifen, 2-(3-trifluoromethylphenyl)histamine, and 8R-lisuride. In contrast, MeHP exhibits a 3-fold higher affinity for the guinea pig H1R than for the human H1R. For the dimeric compounds HP-HP and HP-HA, the guinea pig over human H1R-selectivity is even greater (4- and 10-fold, respectively). The higher affinities of these compounds for the guinea pig H1R are in good agreement with the higher potencies of these agonists for guinea pig versus human H1Rs as recently demonstrated using a GTPase assay (Seifert et al., 2003).
The species-dependent pharmacology of several of the histaprodifen analogs is also observed in functional assays. Measuring the effects on the contraction of the guinea pig ileum, HP-HA is up to 50-fold more potent than histamine (Christophe et al., 2003; Seifert et al., 2003), whereas in assays using heterologously expressed hH1Rs, both HP-HA and histamine seem to be equipotent (Seifert et al., 2003; Bruysters et al., 2004).
Generation and Evaluation of Human H1R Mutants. The ligand-binding pocket of aminergic receptors is generally considered to reside within the TM domains (Shi and Javitch, 2002). Within these TM domains, several amino acids have been identified in the human and guinea pig H1R that are important for the interaction of ligands with the H1R: Asp107 (3.32) in TM3 (Ohta et al., 1994; Moguilevsky et al., 1998; Nonaka et al., 1998; Bruysters et al., 2004), Trp158 (4.56) in TM4 (Wieland et al., 1999), Lys191 (5.39) and Asn198 (5.46) in TM5 (Leurs et al., 1994, 1995; Ohta et al., 1994; Moguilevsky et al., 1995; Moguilevsky et al., 1998; Bruysters et al., 2004), and Phe432 (6.52) and Phe435 (6.55) in TM6. None of these amino acids differ between human and guinea pig H1Rs. Actually, the sequence similarity within these TMs is high (93%), and only 12 amino acids differ between the two proteins (Fig. 2A). In the hH1R, of these 12 amino acids, only Ile37 (1.42) and Cys38 (1.43) in TM1, Asn84 (2.61) and Leu89 (2.66) in TM2, and Leu449 (7.34) and Ile459 (7.44) in TM7 are predicted to be located either in proximity to the hydrophilic cleft in the hH1R or on the interface of two TMs (Fig. 2B). We therefore reasoned that especially these amino acids may be directly involved in ligand binding and that one of these residues might be responsible for the observed differences in pharmacology between human and guinea pig H1Rs. To test this hypothesis, we created the following mutant hH1Rs in which the selected amino acids are individually replaced into their guinea pig counterparts: hH1R Ile37Val, hH1R Cys38Ser, hH1R Asn84Ser, hH1R Leu89His, hH1R Leu449Val, and hH1R Ile459Leu. Although in our H1R model Ile433 (6.53) points toward the plasma membrane, we also included the mutant hH1R Ile433Val receptor in our study because Ile433 is located in between the established hH1R-ligand interaction points Phe432 (6.52) and Phe435 (6.55). In general, we noticed that, at the selected positions, the amino acids present in the human H1R are bulkier than their guinea pig H1R counterparts (Table 1).
Most of the generated mutant H1Rs are expressed at comparable levels (Bmax values of 10 pmol/mg) and bind [3H]mepyramine with unchanged affinity (Kd values of 0.5-1.7 nM) compared with wild-type human H1Rs (Table 1). However, the mutant receptor hH1R-Leu89His (2.66), with a point mutation in the top of TM2, did not show any [3H]mepyramine binding at concentrations up to 30 nM and may not be properly expressed at the cell membrane. Displacement of [3H]mepyramine binding indicated that all tested mutant H1Rs bind the endogenous agonist histamine with unchanged affinity (Table 2). Only for mutant hH1R Asn84Ser (2.61) receptors, which harbor a point mutation in TM2, the affinities for HP-HA are increased (pKi = 6.8) compared with the wild-type hH1R (pKi = 6.1), resulting in a gpH1R-like (pKi = 7.1) pharmacology (Fig. 3; Table 2). In addition, for HP-HP, the species difference was reversed by the Asn84Ser mutation (Table 2).
HP-HA is an agonist for the hH1R as measured using a Gq/11-mediated nuclear factor-B reporter gene assay (pEC50 = 6.4 ± 0.1) with a potency comparable with histamine (pEC50 = 6.4 ± 0.2) (Bruysters et al., 2004). For both the gpH1R and mutant hH1R Asn84Ser (2.61), the potency of HP-HA (pEC50 values of 7.2 ± 0.1 and 7.9 ± 0.1, respectively) strongly exceeds that of histamine (pEC50 values of 6.0 ± 0.1 and 6.5 ± 0.1, respectively). These findings confirm that also in a functional assay we observe species-specific H1R pharmacology, and the mutant hH1R Asn84Ser not only displays a guinea pig H1R binding profile but also a guinea pig H1R functional profile.
These data suggest that residue Asn/Ser84 (2.61) is of critical importance for the observed species-dependent agonist pharmacology of the human and guinea pig H1Rs. Moreover, these data also indicate that for some H1R agonists TM2 is part of the H1R ligand binding-pocket. Interestingly, both human and rat H1Rs have an asparagine at position 2.61. Measuring endothelium-dependent relaxation of rat aortic rings, Menghin et al. (2003) have shown that MeHP and HP-HA are equipotent, corroborating our previous findings with human H1Rs expressed in COS-7 cells (Bruysters et al., 2004). However, measuring guinea pig ileum contractions, the potency of HP-HA exceeds that of MeHP 10-fold (Menghin et al., 2003). These observations further strengthen the involvement of Asn/Ser84 (2.61) in the species-dependent H1R pharmacology. Consequently, pharmacological observations with rat H1Rs will have more predictive power for the action of ligands at human H1Rs.
Characterization of a Novel, Species-Selective H1R Antagonist. The H1R species-selective interactions were originally observed for bulky H1R agonists (HP-HA and HP-HP). These compounds seem to interact with the "classical" binding pocket (TMs 3, 4, 5, and 6) and Asn/Ser84 (2.61), hereby defining an additional binding pocket near TM2. To test whether the additional interactions are restricted to agonists alone, or are also possible for antagonists, we screened an in-house library of H1R antagonists. From this selection, VUF 4669 was identified as an H1R antagonist, which differentiates significantly between human and guinea pig H1Rs. VUF 4669 showed a 17-fold increase in binding affinity for the guinea pig H1R (pKi = 9.0 ± 0.1), compared with its affinity for the human H1R (pKi = 7.7 ± 0.1) (Table 2). Apparently, the concept of species-selective binding is not restricted to H1R agonists but can also be observed for certain H1R antagonists. Again, VUF 4669 exhibits an increased affinity for the mutant hH1 Asn84Ser receptor (pKi = 8.9 ± 0.1), confirming the guinea pig-like pharmacological profile of this mutant human H1R. The other human to guinea pig H1R mutants used in this study exhibit an affinity for VUF 4669 that is identical to the affinity for the WT human H1R (Table 2).
Previously, also several arpromidine analogs, which display both H1R antagonistic and H2R agonistic properties, were characterized as guinea pig H1R-preferring compounds (Seifert et al., 2003). Indeed, VUF 8401, a structural analog of arpromidine displays a 9-fold higher affinity for the guinea pig H1R than for the human H1R (Table 2). In addition, VUF 8401 binds with an increased affinity to the mutant hH1R Asn84Ser (2.61) (Table 2), although this mutation did not fully reverse the species difference. None of the other mutant hH1Rs show an increased affinity for VUF 8401 (Table 2). Interaction with Asn/Ser84 (2.61) therefore partially explains the observed species difference. We hypothesize that for arpromidine-like ligands the higher affinity depends on Ser84 (2.61) and additional guinea pig H1R-specific residues. This hypothesis will be the basis of future investigations.
Like HP-HA and VUF 4669, arpromidine analogs are bulky ligands, having aromatic moieties on either side of a protonated moiety, and we hypothesize that these features are mandatory for species selectivity. H1R antagonists such as terfenadine, fexofenadine, and oxatomide, however, also show such features, but they seem not to be species-selective. Clearly, the simple presence of two aromatic domains in a ligand is not the only denominator for species selectivity.
Rationalization of the Role of Asn84 (2.61) in the Species-Selective Binding. To rationalize the potential role of the amino acid at position 2.61 (Asn/Ser) in the species-selective binding of HP-HA, we created a homology model for the human H1R on the basis of the available structural information on bovine rhodopsin (Palczewski et al., 2000; Okada et al., 2002). In the absence of ligand, our H1R homology model suggests hydrogen bonding between Asn84 (2.61) and Tyr458 (7.43), a residue that is conserved between human and guinea pig H1Rs. Using the automated docking procedure GOLD version 2.1 (Jones et al., 1997), we subsequently docked the agonist HP-HA in the receptor model (Fig. 4A). In contrast to H1R antagonists such as cetirizine, the diphenyl moiety of HP-HA is not oriented toward TM6, but it is predicted to point toward TMs 1, 2, and 7, confirming our previous suggestions based on site-directed mutagenesis studies of the histamine binding pocket (Bruysters et al., 2004). Thereafter, we changed Asn84 (2.61) into Ser, thus creating a model of the hH1R Asn84Ser receptor containing HP-HA (Asn84Ser model). Molecular dynamics simulations were subsequently performed to optimize both HP-HA containing WT and Asn84Ser models. During both simulations, hydrogen bonding was maintained between Asn84 (2.61) and Tyr458 (7.43) in the WT model (3.31 ; Fig. 4B) and between Ser84 (2.61) and Tyr458 (7.43) in the Asn84Ser model (2.80 ; Fig. 4C). However, the orientation of Tyr458 differs between both models, probably because of the structural differences between Ser and Asn at position 2.61 (e.g., length and flexibility of the side chain). Because the affinity of HP-HA is higher for the Asn84Ser H1R, the conformation of HP-HA in the Asn84Ser model is considered more favorable. In the WT model, Tyr458 occupies the space that in the Asn84Ser model is occupied by one of the phenyl rings of HP-HA. Our computational studies therefore suggest that Tyr458 might sterically hinder the binding of HP-HA in the hH1R, thereby "forcing" HP-HA to bind deeper within the receptor.
To test the potential involvement of Tyr458 (7.43) in the binding of HP-HA to the human H1R, we mutated Tyr458 (7.43) in the human H1R into an alanine (hH1R Tyr458Ala). Saturation binding analysis using [3H]mepyramine shows that this mutant H1R is expressed at comparable levels (Bmax = 8.2 ± 3.5 pmol/mg protein) and with an unchanged affinity for [3H]mepyramine (Kd = 3.0 ± 0.7) in comparison with the wild-type H1R. The Tyr458Ala mutation did also not affect the affinity for histamine (pKi = 4.4 ± 0.2) (Fig. 5). Because the mutation Tyr458Ala would remove potential steric hindrance between HP-HA and the hH1R, we expected an increased affinity of HP-HA. Indeed, mutation of Tyr458 into an alanine results in a 5-fold increase in affinity for HP-HA (pKi = 6.8 ± 0.1) compared with the wild-type H1R (Fig. 5). This affinity is similar to the affinity of HP-HA for both the gpH1R (pKi = 7.1 ± 0.1) and the hH1R Asn84Ser (pKi = 6.8 ± 0.1) (Table 2).
The results of our computational and mutagenesis studies indicate an important role of Asn84 (2.61) as selectivity switch. Moreover, our results illustrate the first structural features of an additional binding pocket between TM2 and TM7 in the H1R. Residues in both TM2 and TM7 have been implicated in ligand binding for only a few other aminergic receptors (for review, see Shi and Javitch, 2002). For example, bulky H2R agonists were suggested to interact with TM7 in the H2 receptor (Kelley et al., 2001), whereas dopamine D2/D4 receptor subtype selectivity of several classes of antagonists is determined by a hydrophobic microdomain formed by six amino acids in TM2, TM3, and TM7 (including position 2.61) (Javitch et al., 1999). Also for adrenergic receptors, the key to 1/2 agonist selectivity seems to be localized in TMs 2 and 7 (Isogaya et al., 1999). Moreover, amino acids present at position 7.43 (homologous to hH1R Tyr458) are demonstrated to be involved in ligand binding to 5HT2A (Roth et al., 1997) and muscarinic acetylcholine M3 receptors (Wess et al., 1991). The involvement of TMs 2 and 7 in the H1R binding pocket of some H1R ligands is therefore highly likely.
Conclusions
In conclusion, the human and guinea pig H1Rs exhibit significantly different affinities for agonists, such as HP-HA and HP-HP, as well as for several antagonists such as VUF 4669 and VUF 8401. These differences can be explained by the presence of Asn84 (2.61) in the hH1R versus Ser84 (2.61) in the gpH1R. Based on molecular dynamics simulations and site-directed mutagenesis data, we suggest a possible role for Tyr458 (7.43) in the binding of certain H1R ligands. Our data provide the first evidence that for these H1R ligands, TM2 and TM7 are also part of the ligand binding pocket. Exploitation of these additional interaction points within the H1R ligand binding pocket in drug development programs may yield a new generation of antihistamines with increased structural diversity compared with the currently known ligands.
Acknowledgements
We thank F. Aelbrecht, C.v.d. Perren, G. J. Sterk, and J. Hulshof for valuable assistance.
This study was supported in part by UCB Pharma.
References
Bakker RA, Schoonus SB, Smit MJ, Timmerman H, and Leurs R (2001) Histamine H1-receptor activation of nuclear factor-B: roles for G- and Gq/11-subunits in constitutive and agonist-mediated signaling. Mol Pharmacol 60: 1133-1142.
Bakker RA, Weiner DM, ter Laak T, Beuming T, Zuiderveld OP, Edelbroek M, Hacksell U, Timmerman H, Brann MR, and Leurs R (2004) 8R-lisuride is a potent stereospecific histamine H1-receptor partial agonist. Mol Pharmacol 65: 538-549.
Ballesteros J and Weinstein H (1995) Intergrated methods for the construction of three-dimensional models of structure-function relations in G protein-coupled receptors. Methods Neurosci 25: 366-428.
Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254.
Bruysters M, Pertz HH, Teunissen A, Bakker RA, Gillard M, Chatelain P, Schunack W, Timmerman H, and Leurs R (2004) Mutational analysis of the histamine H1-receptor binding pocket of histaprodifens. Eur J Pharmacol 487: 55-63.
Canutescu AA, Shelenkov AA, and Dunbrack RL Jr (2003) A graph-theory algorithm for rapid protein side-chain prediction. Protein Sci 12: 2001-2014.
Christophe B, Carlier B, Schunack W, Chatelain P, Peck MJ, and Massingham R (2003) Histamine H1-agonist properties of histaprodifen derivatives on guinea-pig isolated trachea and ileum. Inflamm Res 52: S51-S52.
Elz S, Kramer K, Pertz HH, Detert H, ter Laak AM, Kuhne R, and Schunack W (2000) Histaprodifens: synthesis, pharmacological in vitro evaluation and molecular modeling of a new class of highly active and selective histamine H1-receptor agonists. J Med Chem 43: 1071-1084.
Fukui H, Fujimoto K, Mizuguchi H, Sakamoto K, Horio Y, Takai S, Yamada K, and Ito S (1994) Molecular cloning of the human histamine H1 receptor gene. Biochem Biophys Res Commun 201: 894-901.
Gillard M, Van Der Perren C, Moguilevsky N, Massingham R, and Chatelain P (2002) Binding characteristics of cetirizine and levocetirizine to human H1 histamine receptors: contribution of Lys191 and Thr194. Mol Pharmacol 61: 391-399.
Goldman LA, Cutrone EC, Kotenko SV, Krause CD, and Langer JA (1996) Modifications of vectors pEF-BOS, pcDNA1 and pcDNA3 result in improved convenience and expression. Biotechniques 21: 1013-1015.
Hill SJ, Ganellin CR, Timmerman H, Schwartz JC, Shankley NP, Young JM, Schunack W, Levi R, and Haas HL (1997) International Union of Pharmacology. XIII. Classification of histamine receptors. Pharmacol Rev 49: 253-278.
Isogaya M, Sugimoto Y, Tanimura R, Tanaka R, Kikkawa H, Nagao T, and Kurose H (1999) Binding pockets of the 1- and 2-adrenergic receptors for subtype-selective agonists. Mol Pharmacol 56: 875-885.
Javitch JA, Ballesteros JA, Chen J, Chiappa V, and Simpson MM (1999) Electrostatic and aromatic microdomains within the binding-site crevice of the D2 receptor: contributions of the second membrane-spanning segment. Biochemistry 38: 7961-7968.
Jones G, Willett P, Glen RC, Leach AR, and Taylor R (1997) Development and validation of a genetic algorithm for flexible docking. J Mol Biol 267: 727-748.
Kelley MT, Burckstummer T, Wenzel-Seifert K, Dove S, Buschauer A, and Seifert R (2001) Distinct interaction of human and guinea pig histamine H2-receptor with guanidine-type agonists. Mol Pharmacol 60: 1210-1225.
Leschke C, Elz S, Garbarg M, and Schunack W (1995) Synthesis and histamine H1 receptor agonist activity of a series of 2-phenylhistamines, 2-heteroarylhistamines and analogues. J Med Chem 38: 1287-1294.
Leurs R, Smit MJ, Meeder R, Ter Laak AM, and Timmerman H (1995) Lysine200 located in the fifth transmembrane domain of the histamine H1 receptor interacts with histamine but not with all H1 agonists. Biochem Biophys Res Commun 214: 110-117.
Leurs R, Smit MJ, Tensen CP, Ter Laak AM, and Timmerman H (1994) Site-directed mutagenesis of the histamine H1-receptor reveals a selective interaction of asparagine207 with subclasses of H1-receptor agonists. Biochem Biophys Res Commun 201: 295-301.
Lopez-Rodriguez ML, Vicente B, Deupi X, Barrondo S, Olivella M, Morcillo MJ, Behamu B, Ballesteros JA, Salles J, and Pardo L (2002) Design, synthesis and pharmacological evaluation of 5-hydroxytryptamine(1a) receptor ligands to explore the three-dimensional structure of the receptor. Mol Pharmacol 62: 15-21.
Menghin S, Pertz HH, Kramer K, Seifert R, Schunack W, and Elz S (2003) N-Imidazolylalkyl and pyridylalkyl derivatives of histaprodifen: synthesis and in vitro evaluation of highly potent histamine H1-receptor agonists. J Med Chem 46: 5458-5470.
Moguilevsky N, Differding E, Gillard M, and Bollen A (1998) Rational drug design using mammalian cell lines expressing site-directed mutants of the human H1 histaminic receptor. Anim Cell Technol: Basic Appl Asp 9: 65-69.
Moguilevsky N, Varsalona F, Guillaume JP, Noyer M, Gillard M, Daliers J, Henichart JP, and Bollen A (1995) Pharmacological and functional characterisation of the wild-type and site-directed mutants of the human H1 histamine receptor stably expressed in CHO cells. J Recept Signal Transduct Res 15: 91-102.
Nonaka H, Otaki S, Ohshima E, Kono M, Kase H, Ohta K, Fukui H, and Ichimura M (1998) Unique binding pocket for KW-4679 in the histamine H1 receptor. Eur J Pharmacol 345: 111-117.
Ohta K, Hayashi H, Mizuguchi H, Kagamiyama H, Fujimoto K, and Fukui H (1994) Site-directed mutagenesis of the histamine H1 receptor: roles of aspartic acid107, asparagine198 and threonine194. Biochem Biophys Res Commun 203: 1096-1101.
Okada T, Fujiyoshi Y, Silow M, Navarro J, Landau EM, and Shichida Y (2002) Functional role of internal water molecules in rhodopsin revealed by X-ray crystallography. Proc Natl Acad Sci USA 99: 5982-5987.
Palczewski K, Kumasaka T, Hori T, Behnke CA, Motoshima H, Fox BA, Le Trong I, Teller DC, Okada T, Stenkamp RE, et al. (2000) Crystal structure of rhodopsin: a G protein-coupled receptor. Science (Wash DC) 289: 739-745.
Roth BL, Shoham M, Choudhary MS, and Khan N (1997) Identification of conserved aromatic residues essential for agonist binding and second messenger production at 5-hydroxytryptamine2A receptors. Mol Pharmacol 52: 259-266.
Seifert R, Wenzel-Seifert K, Burckstummer T, Pertz HH, Schunack W, Dove S, Buschauer A, and Elz S (2003) Multiple differences in agonist and antagonist pharmacology between human and guinea pig histamine H1-receptor. J Pharmacol Exp Ther 305: 1104-1115.
Shi L and Javitch JA (2002) The binding site of aminergic G protein-coupled receptors: the transmembrane segments and second extracellular loop. Annu Rev Pharmacol Toxicol 42: 437-467.
ter Laak AM, Venhorst J, Donne-Op den Kelder GM, and Timmerman H (1995) The histamine H1-receptor antagonist binding site. A stereoselective pharmacophoric model based upon (semi-)rigid H1-antagonists and including a known interaction site on the receptor. J Med Chem 38: 3351-3360.
Wess J, Gdula D, and Brann MR (1991) Site-directed mutagenesis of the m3 muscarinic receptor: identification of a series of threonine and tyrosine residues involved in agonist but not antagonist binding. EMBO (Eur Mol Biol Organ) J 10: 3729-3734.
Wieland K, Laak AM, Smit MJ, Kuhne R, Timmerman H, and Leurs R (1999) Mutational analysis of the antagonist-binding site of the histamine H1 receptor. J Biol Chem 274: 29994-30000.
Zhang M-Q, Leurs R, and Timmerman H (1997) Histamine H1-receptor antagonists, in Burger's Medicinal Chemistry and Drug Discovery (Wolf ME ed) pp 495-559, John Wiley & Sons, Inc., New York.
Zingel V, Leschke C, and Schunack W (1995) Developments in histamine H1-receptor agonists. Prog Drug Res 44: 49-85.