HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wieland, K. Articles by Leurs, R. Search for Related Content PubMed PubMed Citation Articles by Wieland, K. Articles by Leurs, R. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 42, 29994-30000, October 15, 1999

Mutational Analysis of the Antagonist-binding Site of the Histamine H₁ Receptor^*

Kerstin Wieland, Anton M. Ter Laak^§, Martine J. Smit, Ronald Kühne^§, Henk Timmerman, and Rob Leurs^¶

From the  Leiden/Amsterdam Center for Drug Research, Department of Pharmacochemistry, Division of Medicinal Chemistry, Faculty of Chemistry, Vrije Universiteit, De Boelelaan 1083, 1081 HV Amsterdam, the Netherlands and the ^§ Institute for Molecular Pharmacology, Biocomputing/Molecular Modelling Group, Alfred Kowalke Strasse 4, D-10315 Berlin, Germany

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We combined in a previously derived three-dimensional model of the histamine H₁ receptor (Ter Laak, A. M., Timmerman, H., Leurs, H., Nederkoorn, P. H. J., Smit, M. J., and Donne-Op den Kelder, G. M. (1995) J. Comp. Aid. Mol. Design. 9, 319-330) a pharmacophore for the H₁ antagonist binding site (Ter Laak, A. M., Venhorst, J., Timmerman, H., and Donné-Op de Kelder, G. M. (1994) J. Med. Chem. 38, 3351-3360) with the known interacting amino acid residue Asp¹¹⁶ (in transmembrane domain III) of the H₁ receptor and verified the predicted receptor-ligand interactions by site-directed mutagenesis. This resulted in the identification of the aromatic amino acids Trp¹⁶⁷, Phe⁴³³, and Phe⁴³⁶ in transmembrane domains IV and VI of the H₁ receptor as probable interaction points for the trans-aromatic ring of the H₁ antagonists. Subsequently, a specific interaction of carboxylate moieties of two therapeutically important, zwitterionic H₁ antagonists with Lys²⁰⁰ in transmembrane domain V was predicted. A Lys²⁰⁰  Ala mutation results in a 50- (acrivastine) to 8-fold (d-cetirizine) loss of affinity of these zwitterionic antagonists. In contrast, the affinities of structural analogs of acrivastine and cetirizine lacking the carboxylate group, triprolidine and meclozine, respectively, are unaffected by the Lys²⁰⁰  Ala mutation. These data strongly suggest that Lys²⁰⁰, unique for the H₁ receptor, acts as a specific anchor point for these "second generation" H₁ antagonists.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Since the initial discovery of the role of histamine in allergic conditions (1) serious efforts have been made to develop drugs that inhibit the actions of histamine. Already in 1933, Fourneau and Bovet (2) reported the first "antihistamine" piperoxan. Following this finding many potent H₁ antagonists that can be considered as variations of diaryl-substituted ethylamines (e.g. diphenhydramine and mepyramine) have been developed (for review see Ref. 3). These "first generation" H₁ antagonists are quite effective in humans in allergic rhinitis and urticaria, but because of central nervous system penetration and central H₁ receptor blockade their clinical use is hampered by sedative side effects (3-5). A "second generation" of nonsedative H₁ antagonists (e.g. astemizole, acrivastine, cetirizine, loratidine, and terfenadine) has recently been developed (for review see Ref. 3). Their altered pharmacokinetics result in good clinical effectiveness combined with a strongly reduced sedative potential (3-5).

The development of H₁ antagonists has so far been directed by traditional medicinal chemistry (3). With the availability of the genetic information of the histamine H₁ receptor (6), the rationalization of drug-protein interaction has become a major challenge for this therapeutically important class of drugs. Like all aminergic G-protein coupled receptors (GPCR),¹ the H₁ receptor contains an aspartate residue (Asp¹¹⁶) in transmembrane domain (TM) III (6), that is involved in the binding of the protonated amine function found in both agonists and antagonists structures (7, 8). Mutagenesis studies have furthermore shown that the imidazole ring of histamine is accommodated by Lys²⁰⁰ and Asn²⁰⁷ in TM V (9, 10).

In view of the low sequence similarity between GPCRs and bacteriorhodopsin (BR) much controversy exists on the validity of models derived for GPCRs based on the homology with BR (11-13). Nevertheless, despite the speculative nature of BR-derived GPCR models they have been quite helpful in understanding and predicting drug-receptor interactions for a variety of receptors (see e.g. Refs. 14-16). Previously, we also developed a three-dimensional computer model of the histamine H₁ receptor based on the homology with BR, incorporating the results obtained from mutagenesis studies on the agonist binding site (17). In the present study this computer model of the H₁ receptor was combined with a pharmacophoric model of the H₁ antagonistic binding site (18). This ligand-based model for the H₁ antagonistic binding site is based upon an interaction of the protonated amine function of various first generation, semi-rigid H₁ antagonists with an aspartate residue (Asp¹¹⁶ in the guinea pig H₁ receptor) (18) and precisely positions the cis- and trans-aromatic rings of the H₁ antagonists relative to the C and C carbon atoms of this aspartate residue. Combining the three-dimensional receptor model and the ligand-based pharmacophoric model of the H₁ antagonist binding site resulted in the prediction of interactions of aromatic amino acids in TM IV and VI with the H₁ antagonists. Subsequently, we experimentally confirmed the involvement of these predicted amino acids in the binding of the H₁ antagonist [³H]mepyramine by site-directed mutagenesis. Moreover, on the basis of the three-dimensional model of the antagonist-receptor complex, a specific interaction of carboxylate moieties of therapeutically important, second generation zwitterionic H₁ antagonists (acrivastine and cetirizine) with Lys²⁰⁰ in TM V was predicted and experimentally verified.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Bovine serum albumin, DEAE-dextran, polyethyleneimine, and triprolidine hydrochloride were obtained from Sigma. The mouse anti-FLAG M2 monoclonal antibody was obtained from International Biotechnology Inc. The fluorescein isothiocyanate-conjugated rabbit-anti-mouse secondary antibody was supplied by Dakopatts AB (Stockholm, Sweden). [³H]Mepyramine (28 Ci/mmol) was obtained from Amersham Pharmacia Biotech. Gifts of acrivastine (The Wellcome Foundation Ltd., London, United Kingdom), d-cetirizine hydrochloride, meclozine hydrochloride (UCB, Braine-l'Alleud, Belgium), and mianserin hydrochloride (Organon NV, Oss, the Netherlands) are gratefully acknowledged.

Predicition of Ligand-Receptor Interactions-- H₁ antagonists were docked in the previously described three-dimensional receptor model of the guinea pig H₁ receptor (17), using the rigid H₁ antagonist pharmacophoric model of Ter Laak et al. (18). This model describes the three-dimensional topology of the cis- and trans-aromatic rings of cyproheptadine with respect to the positions of the C and C carbon atoms of an putative Asp residue from the receptor (see Fig. 1A). The C and C carbon atoms of the pharmacophore replaced the corresponding atoms of Asp¹¹⁶ in the receptor model. Rotation was carried out along the C-C bond until cyproheptadine was positioned in the receptor in an energetically favorable orientation. The structure of the zwitterionic compounds acrivastine and d-cetirizine were built and optimized with Chem-X and subsequently docked into the H₁ receptor model onto the cyproheptadine template as described previously (18). Subsequently, all freely rotatable bonds in Lys²⁰⁰ and in the side chains of the zwitterionic H₁ antagonist were taken into account in an extensive conformational analysis (MacroModel/AMBER force field (19)).

Site-directed Mutagenesis-- The guinea pig H₁ receptor cDNA was subcloned in the pALTER vector (Promega), and point mutations were introduced according to the manufacturer's protocol. The wild type and mutant receptors were epitope-tagged with an N-terminal FLAG peptide (DYKDDDD) after modification of the cDNA sequence with polymerase chain reaction. In our initial binding studies (see Fig. 2) nontagged receptors were used. The cDNA sequences were verified using the dideoxy chain termination method with the Sequenase kit (U. S. Biochemical Corp.).

Cell Culture and Transfection-- COS-7 and HEK-293 cells were grown at 37 °C in a humidified atmosphere with 5% CO₂ in Dulbecco's modified Eagle's medium, containing 2 mM L-glutamine, 50 IU/ml penicillin, 50 µg/ml streptomycin, and 5 or 10% (v/v) fetal calf serum, respectively. Cells were transiently transfected with pcDNA3, containing the wild type or mutant H₁ receptor cDNA, using DEAE-dextran (COS-7 cells) or calcium phosphate (HEK-293 cells).

Histamine H₁ Receptor Binding Studies-- Transfected cells were harvested after 48 h, homogenized in ice-cold 50 mM Na₂/potassium phosphate buffer (pH 7.4) and used for radioligand binding studies. Cell homogenates (40-50 µg of protein) were incubated for 30 min at 25 °C in 50 mM Na₂/potassium phosphate buffer (pH 7.4) in 400 µl with the indicated concentrations of [³H]mepyramine. The nonspecific binding was defined in the presence of 1 µM mianserin. In displacement studies, cell homogenates were incubated with 1 nM [³H]mepyramine and increasing concentrations of competing ligands. The incubations were stopped by rapid dilution with 3 ml of ice-cold 50 mM Na₂/potassium phosphate buffer (pH 7.4). The bound radioactivity was separated by filtration through Whatman GF/C filters that had been treated with 0.3% polyethyleneimine. Filters were washed twice with 3 ml of buffer, and radioactivity retained on the filters was measured by liquid scintillation counting. The binding data were evaluated by a nonlinear, least squares curve fitting procedure. Protein levels were determined according to Bradford (20), using bovine serum albumin as standard.

[³H]Inositol Phosphate Production-- HEK-293 cells were seeded in 12-well plates and 24 h after transfection labeled overnight in inositol-free culture medium supplemented with 2 µCi/ml myo-[2-³H]inositol. Cells were stimulated for 1 h at 37 °C with Dulbecco's modified Eagle's medium containing 25 mM Hepes (pH 7.4), 20 mM LiCl, and histamine. The incubation was stopped by aspiration of the culture medium and the addition of cold CHCl₃/methanol (1:2 v/v). After extraction with water, [³H]inositol phosphates were isolated by anion exchange chromatography (21).

Immunofluorescence-- Transfected COS-7 cells were grown on glass coverslips and after 48 h fixed with 4% paraformaldehyde/phosphate-buffered saline for 30 min at room temperature and blocked in phosphate-buffered saline/0.1% bovine serum albumin for 1 h at room temperature. Antigen detection was performed as described (22).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Prediction of Ligand-Receptor Interaction Based on Receptor Modelling-- The H₁ antagonist pharmacophoric model of Ter Laak et al. (18) represents low energy conformations of several potent and rigid H₁ antagonists (cyproheptadine, phenindamine, triprolidine, epinastine, mequitazine, and mianserine) and is able to discriminate between the stereochemically different cis- and trans-aromatic rings of these H₁ antagonists (Fig. 1A). To predict amino acid residues involved in the ligand-binding site of these H₁ antagonists the cyproheptadine pharmacophore was docked into the previously derived model of the H₁ receptor. After rotation over the C-C bond of Asp¹¹⁶ and energy optimizations, a single energetically favorable orientation was found for the H₁ antagonist. In this orientation the aromatic rings of the H₁ antagonist were located in the receptor-binding pocket between the TMs III, IV, V, and VI (Fig. 1B). The aromatic rings of cyproheptadine were surrounded by several aromatic amino acids. In the H₁ receptor model, the cis-ring of cyproheptadine is located within 5 Å of a tryptophane in TM IV (Trp¹⁶¹), whereas the trans-aromatic ring is close to two phenylalanines in TM VI (Phe⁴³³ and Phe⁴³⁶) and Trp¹⁶⁷ in TM IV (Fig. 1B). All these predicted residues are conserved in all the reported H₁ receptor sequences (23-29).

View larger version (71K): [in this window] [in a new window]   Fig. 1.   A, the H1 antagonist pharmacophore of Ter Laak et al. (18) describing the position of the cis- and trans-aromatic rings of H1 antagonists with respect to the C and C carbon atoms of a putative aspartate residue. B, the H1 antagonist pharmacophore was docked into the H1 receptor model based on BR (17). A view from the extracellular side shows the orientation of cyproheptadine between the TMs (yellow) and the ionic hydrogen bond interaction with Asp116 in TM III. The trans-aromatic ring of cyproheptadine is in the proximity of Phe433, Phe 436 (TM VI), and Trp167 (TM IV), and the cis-aromatic ring is near Trp161 (TM IV). Two other aromatic residues that were mutated in this study (Phe435 and Trp174) are not in the proximity of the H1 antagonist in this model. Phe435 points toward the membrane, and Trp174 is not shown because in this alignment the residue lies outside the TM region.

Verification of the H₁ Antagonist-binding Site by Site-directed Mutagenesis-- To verify the interaction of H₁ antagonists with the predicted amino acids these residues were initially mutated to alanines. Moreover, two related amino acids (Trp¹⁷⁴ in TM IV and Phe⁴³⁵ in TM VI) that are also conserved in all the reported H₁ receptor sequences (23-29) but not predicted by the GPCR model were also mutated. In the derived model Phe⁴³⁵ points into the phospholipid bilayer (Fig. 1B), and Trp¹⁷⁴ is located just outside TM IV.

The mutant receptors were expressed transiently in COS-7 cells and analyzed by [³H]mepyramine saturation binding studies. Expression of the wild type H₁ receptor in COS-7 cells resulted in a high affinity binding site for [³H]mepyramine (K_D = 0.7 ± 0.1 nM, mean ± S.E., n = 3) (Fig. 2). Mutation of the tryptophane residues Trp¹⁶¹, Trp¹⁶⁷, and Trp¹⁷⁴ in TM IV to alanine residues resulted in distinct effects on the [³H]mepyramine binding to the mutant receptors (Fig. 2). The introduction of the Trp¹⁷⁴  Ala mutation did not reduce the affinity of [³H]mepyramine (Fig. 2; K_D = 3.6 ± 0.6 nM, mean ± S.E., n = 3) dramatically. In contrast, for the Trp¹⁶⁷Ala receptor the affinity for the H₁ antagonist was reduced more than 10-fold (Fig. 2; K_D > 15 nM, n = 3), whereas cells expressing the Trp¹⁶¹  Ala receptor did not show binding of [³H]mepyramine (Fig. 2) significantly higher than binding to mock-transfected COS-7 cells (15-50 fmol/mg protein; data not shown). Similar findings were obtained with the three Phe-Ala mutations in TM VI. The Phe⁴³⁵Ala receptor mutant still bound [³H]mepyramine with high affinity (Fig. 2; K_D = 1.3 ± 0.2 nM, mean ± S.E., n = 3), whereas for the other two mutants a reduced (Phe⁴³⁶  Ala receptor, K_D > 15 nM, n = 3) or totally impaired (Phe⁴³³  Ala receptor) [³H]mepyramine binding was observed (Fig. 2).

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effects of the mutation of aromatic amino acids in TM IV and VI on the binding of the H1 antagonist [3H]mepyramine to wild type, Phe435  Ala, Trp174  Ala, Trp167  Ala, Phe436  Ala, Trp161  Ala, and Phe433  Ala H1 receptor after transient expression in COS-7 cells. The data shown are from a representative example out of at least three independent experiments.

To verify protein expression of the two receptor mutants that did not show detectable [³H]mepyramine binding (Trp¹⁶¹  Ala and Phe⁴³³  Ala), a FLAG epitope was introduced at the N terminus of the wild type and mutant H₁ receptor proteins. Using confocal laser microscopy we identified specific, anti-FLAG immunofluorescence in the plasma membrane of COS-7 cells expressing the epitope-tagged wild type and the Phe⁴³³  Ala H₁ receptor (Fig. 3). For the Trp¹⁶¹  Ala H₁ receptor, the anti-FLAG immunofluorescence was mainly found inside the cell, indicating perturbed receptor expression.

View larger version (33K): [in this window] [in a new window]   Fig. 3.   Localization of epitope-tagged wild type (WT) and mutant H1 receptors, transiently expressed in COS-7 cells. Transfected COS-7 cells were grown on glass coverslips and after 48 h fixed with 4% paraformaldehyde. Immunofluorescence was detected with the mouse anti-FLAG M2 antibody and a fluorescein isothiocyanate-conjugated secondary rabbit anti-mouse antibody.

Based on the observed loss of antagonist affinity upon mutation of Trp¹⁶⁷, Phe⁴³³, and Phe⁴³⁶, we considered these amino acids as likely candidates for the hypothezised interaction with the aromatic rings of the H₁ antagonist. To investigate the role of Trp¹⁶¹, Trp¹⁶⁷, Phe⁴³³, and Phe⁴³⁶ in more detail, we changed the tryptophane residues in TM IV to methionine and phenylalanine (Trp¹⁶¹  Met, Trp¹⁶¹  Phe, Trp¹⁶⁷  Met, and Trp¹⁶⁷  Phe) and the phenylalanine residues in TM 6 to methionine (Phe⁴³³  Met and Phe⁴³⁶  Met). The mutant receptors were epitope-tagged, expressed in COS-7 cells, and evaluated for [³H]mepyramine binding. In contrast to the Trp¹⁶¹  Ala receptor mutant, the Trp¹⁶¹  Met and Trp¹⁶¹  Phe receptor mutants bound [³H]mepyramine with high affinity (Table I). Yet the number of binding sites for the Trp¹⁶¹  Phe receptor mutant was considerably lower compared with the wild type and the Trp¹⁶¹  Met receptor mutant (Table I). As found for the Trp¹⁶⁷  Ala receptor mutant, mutation of Trp¹⁶⁷ to Met or Phe resulted in strongly reduced affinity for [³H]mepyramine (Table I). In contrast, mutating Phe⁴³³ and Phe⁴³⁶ to Met allowed [³H]mepyramine binding with high affinity (Table I).

The affinity of the Trp¹⁶¹  Met and Trp¹⁶¹  Phe mutant receptors for histamine was not changed (Table I), whereas a small to major reduction of the agonist affinity was observed for the Phe⁴³³  Met and Phe⁴³⁶  Met receptor mutants (Table I). Because of the lack of saturable [³H]mepyramine binding, the agonist affinity could not be determined for the other mutants.

Histamine-induced [³H]Inositol Phosphate Accumulation after Stimulation of Wild Type and Mutant H₁ Receptors-- To test the functionality of the mutant receptors, we initially performed [³H]inositol phosphate accumulation experiments in transfected COS-7 cells. However, in mock-transfected COS-7 cells histamine increased basal [³H]inositol phosphate accumulation, suggesting the presence of an endogenously expressed H₁ receptor.² In mock-transfected HEK-293 cells histamine did not stimulate the [³H]inositol phosphate accumulation (Fig. 4), whereas after overexpression of the epitope-tagged wild type H₁ receptor protein (7.1 ± 1.0 pmol/mg protein, mean ± S.E., n = 3) histamine (100 µM) stimulated the [³H]inositol phosphate accumulation 5.9 ± 0.4-fold (mean ± S.E., n = 3) over basal levels (Fig. 4). Evaluation of the various receptor mutants showed that the Trp¹⁶¹  Met, Phe⁴³³  Ala, Phe⁴³³  Met, Phe⁴³⁶  Ala, and Phe⁴³⁶  Met mutant receptors stimulated the [³H]inositol phosphate accumulation as the wild type H₁ receptor (Fig. 4). Similar EC₅₀ values for histamine were observed for the wild type receptor and the Trp¹⁶¹  Met, Trp¹⁶¹  Phe, and Phe⁴³³  Ala receptors, whereas the Phe⁴³³  Met, Phe⁴³⁶  Ala, and Phe⁴³⁶  Met mutant receptors were stimulated less effectively by histamine (Table I). The expression levels of the mutant receptors in HEK-293 cells ranged from 2.4 (Phe⁴³⁶  Ala) to 10.6 pmol/mg protein (Trp¹⁶¹  Met). For the Phe⁴³³  Ala mutant no radioligand binding was found. As expected by the perturbed membrane expression, the Trp¹⁶¹  Ala receptor mutant did not respond to histamine (Fig. 4). As found for the expression in COS-7 cells, the Trp¹⁶¹  Phe was expressed at relatively low levels in HEK-293 cells (0.3 ± 0.1 pmol/mg protein, means ± S.E., n = 3). Despite its low expression level a significant stimulation (Fig. 4) of the mutant receptor by histamine was observed with comparable potency as the wild type receptor (Table I). For the various mutations of Trp¹⁶⁷ only the Trp¹⁶⁷  Phe mutation resulted in a mutant receptor that could be activated by histamine, although with a very low efficacy (Fig. 4 and Table I). Because of the lack of saturable binding in the [³H]mepyramine binding studies, the expression levels of these mutant receptors could not be estimated.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Basal (open bars) and histamine (100 µM) (closed bars) induced production of [3H]inositol phosphates by wild type and mutant H1 receptors expressed in [3H]inositol prelabeled HEK-293 cells. The data shown are the means ± S.E. of at least three independent experiments.

Predicted Interaction of the Zwitterionic H₁ Antagonists Acrivastine and Cetirizine with Lys²⁰⁰-- To investigate whether the acidic moiety of the nonsedative, zwitterionic H₁ antagonist acrivastine specifically interacts with the H₁ receptor protein, this ligand was docked into the H₁ receptor model on top of the template cyproheptadine. Visual inspection of the resulting ligand-receptor complex suggested a possible interaction with Lys²⁰⁰ in TM V. Following this observation a conformational analysis, giving rotational freedom to Lys²⁰⁰ and the carboxylate group of the H₁ antagonist, indeed predicted an interaction between the positively charged Lys²⁰⁰ and the carboxylate group (Fig. 5A). In the case of d-cetirizine, the carboxylate group is attached to the basic nitrogen via a long ether chain. Docking cetirizine in the H₁ receptor model indicated that the carboxylate of d-cetirizine reaches the proximity of Lys²⁰⁰, although the calculated N-O distance of 3.57 Å is somewhat large for a strong (ionic) hydrogen bond interaction (Fig. 5B).

View larger version (53K): [in this window] [in a new window]   Fig. 5.   Zwitterionic H1 antagonists acrivastine (A) and d-cetirizine (B) docked into the H1 receptor model. A, acrivastine, which fits the cyproheptadine pharmacophore (Fig. 1A), makes an additional (ionic) hydrogen bond interaction with Lys200 (TM V). B, d-cetirizine docked into the same H1 receptor model and presenting the carboxylate moiety near Lys200.

Interaction of H₁ Antagonists with Lys²⁰⁰ in TM V-- To verify the predicted interaction of Lys²⁰⁰ with the carboxylate group of acrivastine, we mutated the basic lysine to alanine and methionine to disrupt the potential ionic interaction with the H₁ antagonist. We also mutated Lys²⁰⁰ to an arginine residue, because this basic amino acid should be able to interact with the carboxylate group of acrivastine. Previously, we showed that Lys²⁰⁰ in TM V specifically interacts with some classes of H₁ agonists (including histamine) but not with the prototypic H₁ antagonists [³H]mepyramine and d- and l-chlorpheniramine (10). Moreover, also the Lys²⁰⁰  Arg and the Lys²⁰⁰  Met mutations allowed high affinity [³H]mepyramine binding after expression in COS-7 cells and similar to the Lys²⁰⁰  Ala mutant (10) showed a slightly altered affinity for histamine (Table II). Functional studies in HEK-293 cells indicated that cells transiently transfected with the mutant receptors all responded to histamine with the accumulation of [³H]inositol phosphates, although the EC₅₀ values were higher than found for the wild type receptor (Table II). Displacement of the binding of [³H]mepyramine to the Lys²⁰⁰  Ala, Lys²⁰⁰  Arg, and Lys²⁰⁰  Met receptor mutants indicated a specific interaction of Lys²⁰⁰ with acrivastine. The K_i value for acrivastine was increased more than 50-fold following the Lys²⁰⁰  Ala and Lys²⁰⁰  Met mutations (Fig. 6A and Table II). In contrast, acrivastine still showed high affinity for the Lys²⁰⁰  Arg receptor mutant (Fig. 6A and Table II). To further substantiate these findings we tested the affinity of triprolidine, a structural analog of acrivastine lacking the carboxylate moiety (Table III), for the Lys²⁰⁰  Ala receptor mutant. As expected, the affinity of this close structural analog was not reduced by the Lys²⁰⁰  Ala mutation (Fig. 6B and Table III). The Lys²⁰⁰ residue is also involved in a specific interaction with the nonsedative, zwitterionic H₁ antagonist d-cetirizine (Fig. 6C and Table III). Again, no effect of the Lys²⁰⁰  Ala mutation was found on the affinity of the analog meclozine, which does not contain a functional group that can interact with the Lys²⁰⁰ residue in TM V (Fig. 6C and Table III).

View this table: [in this window] [in a new window]   Table II Effects of mutation of Lys200 of the histamine H1 receptor on ligand binding and signal transduction Receptors were expressed in COS-7 cells or HEK-293 cells and used 48 h after transfection for radioligand binding studies and the accumulation of [3H]inositol phosphates, respectively. The wild type, Lys200  Ala, Lys200  Met, and Lys200  Arg were expressed in HEK-293 cells at receptor densities of respectively 7.1 ± 1.0, 13.6 ± 1.5, 7.2 ± 1.1, and 7.1 ± 0.4 pmol/mg protein. Data were calculated as the means ± S.E. from at least three independent experiments.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Contribution of Lys200 in TM V to the binding of the zwitterionic H1 antagonists acrivastine and cetirizine and their structural analogs triprolidine and meclozine to the H1 receptor. A, the binding of [3H]mepyramine to the wild type, Lys200  Ala, Lys200  Met, and Lys200  Arg H1 receptor was displaced by acrivastine. The binding of [3H]mepyramine to the wild type and Lys200  Ala H1 receptor was displaced by acrivastine and triprolidine (B) and cetirizine and meclozine (C). The data shown are from a representative example out of at least three independent experiments.

View this table: [in this window] [in a new window]   Table III Structural formulas and affinities of the H1 antagonists d-cetirizine, meclozine, acrivastine, and triprolidine for the wild type and Lys200  Ala H1 receptor The affinities of the H1 antagonists for the wild type and Lys200  Ala receptor mutant were determined in [3H]mepyramine displacement studies. Data shown are the means ± S.E. of at least three independent experiments, each performed in triplicate.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

More than 25 years after the initial hypothesis of Nauta et al. (30) of an interaction of the trans-aromatic ring of the H₁ antagonist diphenhydramine with a Phe residue of an hypothetical -helical structure of the H₁ receptor, we identified the aromatic amino acids Trp¹⁶⁷, Phe⁴³³, and Phe⁴³⁶ in the putative -helical TMs IV and VI of the H₁ receptor as probable interaction points for the trans-aromatic ring of the H₁ antagonists. Moreover, we found Lys²⁰⁰ (TM V) to be a specific anchor point for the carboxylate moiety of the nonsedative, zwitterionic second generation H₁ antagonists acrivastine and cetirizine.

The basis for the identification of these amino acids came from the docking of an H₁ antagonistic pharmacophoric model (18) into a previously derived three-dimensional model of the H₁ receptor (17). As a representative example of the first generation H₁ antagonists, the rigid tricyclic cyproheptadine was allowed to interact with its protonated amine function with the highly conserved Asp¹¹⁶ in TM III (7, 8). Several aromatic amino acid residues were predicted to interact with the aromatic rings of the H₁ antagonist. Mutation of Trp¹⁷⁴ and Phe⁴³⁵, which were predicted to not be involved in ligand binding, had indeed no effect on the [³H]mepyramine binding. In contrast, for the mutant Trp¹⁶⁷  Ala and Phe⁴³⁶  Ala receptors a dramatic loss of affinity of [³H]mepyramine was observed. Moreover, the Phe⁴³³  Ala mutation caused a total loss of H₁ antagonist binding, despite membrane expression of the receptor protein and normal responsiveness toward histamine. Replacing Phe⁴³³ and Phe⁴³⁶ with the hydrophobic but aliphatic Met residue only slightly affected the binding of [³H]mepyramine, indicating that these two Phe residues in TMVI are most likely involved in a hydrophobic interaction with the H₁ antagonist. Furthermore, a large reduction in the affinity and efficacy of histamine was observed for the Phe⁴³⁶ mutants. This observation fits well with our recently developed model for the agonist interaction with the H₁ receptor. In this model an interaction of Phe⁴³⁶ with the imidazole ring of histamine is predicted.³ Also for the Phe⁴³³  Met mutant lower affinity and efficacy are observed. In our model for the agonist-receptor interaction, a direct involvement of Phe⁴³³ with the agonist binding is not predicted,³ as substantiated by the full agonist activity at the Phe⁴³³  Ala mutant. Currently, we cannot explain the reduced agonist responses at the Phe⁴³³  Met mutant, although we can speculate that the flexible Met side chain prevents optimal agonist-receptor interaction by steric hindrance.

Replacing Trp¹⁶⁷ with either the aromatic Phe or the aliphatic hydrophobic Met did not allow high affinity [³H]mepyramine binding. Yet, the Trp¹⁶⁷  Phe receptor mutant was able to functionally interact with histamine. These data suggest that Trp¹⁶⁷ is important for high affinity [³H]mepyramine binding but that hydrophobicity or aromaticity per se is not sufficient for a proper interaction. This could be explained if Trp¹⁶⁷would be properly positioned in the binding crevice of the H₁ receptor by an interaction of its indole nitrogen with another (yet unidentified) amino acid of the H₁ receptor.

Searching the GRAP mutant data base (31) for mutated residues at similar positions in other aminergic GPCRs, revealed considerable support for an important role of Trp¹⁶⁷, Phe⁴³³, and Phe⁴³⁶ in the binding of the trans-aromatic ring of the H₁ antagonists. A polymorphism of the human ₂ receptor (Thr¹⁶⁴  Ile) at the position of Trp¹⁶⁷ in TM IV has been reported to alter ligand binding characteristics (32). Moreover, in the dopamine D_2L (33), the 5HT_2A (34), and the ₂ receptor (35), a Phe residue in homologous position as Phe⁴³³ has been implicated in hydrophobic or - interactions with aromatic rings of receptor ligands. Similarly, in the muscarinic receptors an Asn residue is found at the same position as Phe⁴³³, and this Asn⁵⁰⁷ plays a key role in the binding of various muscarinic antagonists to the m3 receptor (36). Finally, in both the dopamine D_2L (37) and the _1A receptor (38) residues at a homologous position as Phe⁴³⁶ are also involved in ligand binding.

Because of the perturbed protein expression of the Trp¹⁶¹  Ala receptor, no conclusive decision could be made regarding its role in the binding of H₁ antagonists. Mutation of Trp¹⁶¹ with either Phe or Met is well allowed for high affinity [³H]mepyramine binding or the interaction with histamine. A Trp¹⁹² residue at homologous position of Trp¹⁶¹ in the m3 receptor is important for both agonist and antagonist binding (39), suggesting that Trp¹⁶¹ could be involved in an hydrophobic interaction with the cis-ring of the H₁ antagonists as well. However, Trp¹⁶¹ is 96% conserved throughout the GPCR family, suggesting a fundamental role in GPCR architecture (40). We can therefore not exclude the possibility that a hydrophobic amino acid at position 161 is simply essential to adopt a functional GPCR conformation.

Based on the results of the site-directed mutagenesis studies we concluded that the three-dimensional H₁ receptor model had some predictive value for receptor-ligand interactions. To challenge our three-dimensional model the zwitterionic ligands acrivastine and d-cetirizine were fitted into the H₁ receptor model. Both ligands contain a carboxylate moiety in either the trans-ring (acrivastine) or connected with a spacer to the amine-function (d-cetirizine). Modeling studies indicated a possible interaction of the negatively charged group with the amine of Lys²⁰⁰ in TM V. The Lys²⁰⁰  Ala H₁ receptor mutant shows normal binding of the classical H₁ antagonist [³H] mepyramine (10). Our present findings strongly suggest that Lys²⁰⁰ is directly involved in the binding of the carboxylate group of acrivastine and d-cetirizine. The Lys²⁰⁰  Ala mutation results in a 30- (acrivastine) to 10-fold (d-cetirizine) loss of affinity of the zwitterionic antagonists, whereas the affinities of acrivastine and cetirizine analogs lacking the negatively charged carboxylate group, triprolidine and meclozine, respectively, are not affected by the Lys²⁰⁰  Ala mutation. The Lys²⁰⁰ residue was also mutated to Met, which resembles the Lys residue sterically but does not contain a protonated amine function. As predicted, a huge reduction of the affinity of acrivastine was observed for the Lys²⁰⁰  Met receptor mutant. Moreover, a basic Arg residue is able to functionally replace Lys²⁰⁰  Ala as shown by the high affinity of acrivastine for the Lys²⁰⁰  Arg H₁ receptor. These data provide strong evidence for the hypothesized ionic interaction between the carboxylate of acrivastine and the protonated amine function of Lys²⁰⁰. As observed previously for the interaction of histamine with the Lys²⁰⁰  Ala mutant receptor (10), histamine was less responsive at the Lys²⁰⁰  Met and Lys²⁰⁰  Arg H₁ receptor as well. Whereas this is not surprising for the Met mutant (10), the observations with the Lys²⁰⁰  Arg H₁ receptor indicate that the longer arginine side chain and the larger guanidinium group cannot optimally accommodate the imidazole ring of histamine. This observation will be important for the refinement of a three-dimensional model for the agonist-binding site.

In view of the emerging cardiotoxicity of several second generation H₁ antagonists (41) and the current interest to combine potent nonsedative H₁ antagonism with other anti-allergic activities (3), the identification of the role of Lys²⁰⁰ will be of importance for the design of potent "third generation" H₁ antagonists. The introduction of an carboxylate group in the structure of the H₁ antagonists has by empirical approach been found to be a very effective way to limit central nervous system penetration and to derive nonsedative H₁ antagonists like acrivastine (42) and cetirizine (43). Consequently, this structural element can also been found in other new H₁ antagonists, e.g. carebastine, fexofenadine, KF-15766, KW-4679, levocabastine, and pibaxizine (for review see Ref. 3). Besides leading to favorable pharmacokinetics, the carboxylate group also discriminates the receptor binding of second generation H₁ antagonists acrivastine and cetirizine from the first generation H₁ antagonists via its interaction with Lys²⁰⁰. Because Lys²⁰⁰ is an unique residue for the H₁ receptor, interaction of ligands with this residue via a carboxylate group will lead, in addition to limited brain penetration, probably also to good H₁ receptor selectivity. For future design of nonsedative H₁ antagonists the introduction of a carboxylate group capable of interacting with Lys²⁰⁰ could therefore be favorable.

In conclusion, our study shows that combining known interacting amino acids of the receptor (Asp¹¹⁶) with a pharmacophore for the H₁ antagonist-binding site and verification by site-directed mutagenesis results in the identification of Trp¹⁶⁷, Phe⁴³³, and Phe⁴³⁶ in TM IV and VI as probable interaction points for the aromatic rings of H₁ antagonists. Moreover, the use of the three-dimensional receptor model allowed the identification of Lys²⁰⁰ in TM V as specific anchor point for some nonsedative, zwitterionic H₁ antagonists.

	ACKNOWLEDGEMENTS

We thank Drs. Edwin Roovers and Yvonne van de Vrede for technical assistance.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed. Tel.: 31-20-4447579; Fax: 31-20-4447610; E-mail: leurs@chem.vu.nl.

² K. Wieland, H. Timmerman, and R. Leurs, unpublished observations.

³ A. M. Ter Laak and R. Kühne, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: GPCR, G-protein coupled receptor; BR, bacteriorhodopsin; TM, transmembrane domain.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				A. Strasser, H.-J. Wittmann, and R. Seifert Ligand-Specific Contribution of the N Terminus and E2-Loop to Pharmacological Properties of the Histamine H1-Receptor J. Pharmacol. Exp. Ther., September 1, 2008; 326(3): 783 - 791. [Abstract] [Full Text] [PDF]

				A. Strasser, B. Striegl, H.-J. Wittmann, and R. Seifert Pharmacological Profile of Histaprodifens at Four Recombinant Histamine H1 Receptor Species Isoforms J. Pharmacol. Exp. Ther., January 1, 2008; 324(1): 60 - 71. [Abstract] [Full Text] [PDF]

				R. A. Bakker, A. Jongejan, K. Sansuk, U. Hacksell, H. Timmerman, M. R. Brann, D. M. Weiner, L. Pardo, and R. Leurs Constitutively Active Mutants of the Histamine H1 Receptor Suggest a Conserved Hydrophobic Asparagine-Cage That Constrains the Activation of Class A G Protein-Coupled Receptors Mol. Pharmacol., January 1, 2008; 73(1): 94 - 103. [Abstract] [Full Text] [PDF]

				M. R. Braden, J. C. Parrish, J. C. Naylor, and D. E. Nichols Molecular Interaction of Serotonin 5-HT2A Receptor Residues Phe339(6.51) and Phe340(6.52) with Superpotent N-Benzyl Phenethylamine Agonists Mol. Pharmacol., December 1, 2006; 70(6): 1956 - 1964. [Abstract] [Full Text] [PDF]

M. Bruysters, A. Jongejan, A. Akdemir, R. A. Bakker, and R. Leurs A Gq/11-coupled Mutant Histamine H1 Receptor F435A Activated Solely by Synthetic Ligands (RASSL) J. Biol. Chem., October 14, 2005; 280(41): 34741 - 34746.