Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Histamine is a biogenic amine that has tremendous influence over a variety of physiological and pathologic processes through different histamine receptors. Thus far, four pharmacologically distinct histamine receptors have been identified and cloned, all of which are members of the G-protein-coupled receptor family of proteins. Cloning of the first three histamine receptors, the H₁, H₂, and H₃ receptors, was reported previously (Gantz et al., 1991; Yamashita et al., 1991; Lovenberg et al., 1999). Recently, the fourth histamine receptor, the H₄ receptor, was cloned independently by several groups (Oda et al., 2000; Liu et al., 2001; Morse et al., 2001; Nguyen et al., 2001; Zhu et al., 2001). The H₄ receptor is preferentially expressed in tissues of immunological relevance, and its expression seems to be regulated by interleukin-10 or -13 (Morse et al., 2001). Understanding the molecular mechanism for the interaction between histamine and the H₄ receptor will probably be useful for the development of selective H₄ antagonists and for elucidating and modulating its function in the future.

Site-directed mutagenesis studies were performed previously to investigate the molecular basis for binding of histamine and histaminergic antagonists to H₁ and H₂ receptors. In transmembrane region (TM) 3, a conserved aspartic acid [amino acid position 107 (3.32) in human H₁ and position 98 (3.32) in human H₂ receptor] is essential for the binding of both histamine and basic antagonists for both receptors (Gantz et al., 1992; Ohta et al., 1994). In the H₁ receptor, antagonists have been shown to vary in the strength of interaction with this aspartate (Nonaka et al., 1998). In the H₃ receptor, TM3 residues adjacent to Asp¹¹⁴ (3.32) have been shown to form the basis for species-specific binding of antagonists (Ligneaux et al., 2000; Lovenberg et al., 2000). Residues in TM5 of the guinea pig H₁ and human H₂ receptors have also been shown to be required for histamine binding (Gantz et al., 1992; Leurs et al., 1994). Asn²⁰⁷ (5.46) in TM5 of the H₁ receptor is involved in hydrogen binding with the N nitrogen atom of histamine, whereas Asp¹⁸⁶ (5.42) in TM5 of the H₂ receptor is connected to the same N nitrogen atom of histamine by an ion pair (Gantz et al., 1992; Leurs et al., 1994). In addition, Thr¹⁹⁰ (5.46) in TM5 of the H₂ receptor was shown to be important in establishing the kinetics of histamine binding and action (Gantz et al., 1992). This residue is thought to participate in hydrogen binding to the N nitrogen atom of histamine (Gantz et al., 1992). These findings are consistent with prior expectations that all three histamine nitrogens participate in binding and receptor activation (Weinstein et al., 1976).

In this study, we carried out computer modeling of the H₄ receptor based on its primary sequence and examined the putative histamine-binding pocket. The resulting model suggests three potential interactions between the H₄ receptor and histamine. Asp⁹⁴ (3.32) in TM3 of the H₄ receptor, which matches the conserved Asp residues in TM3 of both the H₁ and H₂ receptor, could interact with cationic amine moiety of histamine. The imidazole ring of histamine would be predicted to interact with the side chains of Glu¹⁸² (5.46) and either Thr¹⁷⁸ (5.42) or Ser¹⁷⁹ (5.43) in TM5 of the H₄ receptor. In addition, our computer model of the H₄ receptor suggested that Asn¹⁴⁷ (4.57) in TM4 and Ser³²⁰ (6.52) in TM6 could be important for histamine binding. These two residues seem to reside at the opening of the putative binding pocket with their side chains pointing inward, suggesting the potential for a role in guiding histamine into the binding site. Interestingly, corresponding amino acid positions in TM4 of the H₁, H₂, and H₃ receptors are occupied by bulkier residues (Tyr, Trp, and Phe, respectively). A Phe is found in the H₁ and H₂ receptor in amino acid positions corresponding to Ser³²⁰ (6.52) in TM6 of the H₄ receptor (Fig. 1). The bulkier side chains at these positions in the H₁, H₂, and H₃ receptors have the potential to impede access of histamine to the binding pocket. The molecular model would predict that the difference in the identities of the residues at these two sites between H₄ and the other three described histamine receptors have potential implications for subtype-specific differences in histamine interaction.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Alignment of amino acid sequences of the third, fourth, fifth, and sixth transmembrane domain in human histamine H1, H2, H3, and H4 receptors. The bold and underlined residues in the H4 receptor are predicated to be involved in the action of histamine. These residues are also numbered using the Ballesteros and Weinstein index system modified by van Rhee and Jacobson (Ballesteros and Weinstein, 1995; van Rhee and Jacobson, 1996).

To explore the hypotheses regarding the importance of the residues mentioned above for H₄ receptor binding and activation by histamine, we mutated these residues individually and in combination. The ability of each mutant and the wild-type receptor to bind and respond to histamine was measured. The results suggest a model for subtype-specific differences in the mechanism of interaction between histamine and its receptor.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Modeling of the Histamine H₄ Receptor. A molecular model of the human H₄ receptor was constructed from the structure of rhodopsin (Palczewski et al., 2000), using a method described previously for modeling the human melanin-concentrating hormone receptor (MacDonald et al., 2000). Briefly, the model was built with the Look program (Molecular Application Group, Palo Alto, CA), which uses the SEGMED program (Levitt, 1992). A model of histamine was docked to the H₄ receptor homology model at a site corresponding to its expected binding pocket in H₁ and H₂ receptors, based on prior mutagenesis data. The resulting complex model was refined by 1000 steps of molecular mechanics minimization with the Insight II/Discover program (Accelrys, San Diego, CA).

Site-Directed Mutagenesis. The human H₄ receptor cDNA (Morse et al., 2001) was subcloned into the Nhe-1 and Not-1 sites of the mammalian expression vector pME18-CD8-Flag, which allows an expressed protein to be epitope-tagged with N-terminal FLAG peptide and includes a signal peptide sequence derived from CD8 that promotes efficient expression. All the point mutations were introduced using a QuikChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA). The full-length wild-type and mutant cDNA sequences were verified using the cycle-sequencing method with the ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems, Foster City, CA).

Cell Culture, Transfection, and Expression. Serum-free medium-adapted (SFM) HEK-293 F cells (Invitrogen, Carlsbad, CA) were grown at 37°C in a humidified atmosphere with 5% CO₂ in Dulbecco's modified Eagle's medium containing 10% (v/v) fetal bovine serum. Cells were transiently transfected with the wild-type or mutant H₄ receptor cDNA in pME18-CD8-Flag using LipofectAMINE 2000 reagent (Invitrogen). Receptor expression on the cell surface was determined by flow cytometric analysis using a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA). Briefly, cells were harvested in 5 mM ice-cold EDTA in PBS 24 h after transfection, washed twice with PBS, and stained with biotinylated anti-FLAG M₂ monoclonal antibody (Sigma-Aldrich, St. Louis, MO) on ice for 30 min. After being washed twice with PBS, the cells were then stained with phycoerythrin-conjugated streptavidin (PharMingen, San Diego, CA) for 30 min on ice, followed by two washes with cold PBS before being analyzed.

Membrane Preparation. Twenty-four hours after transfection, the cells were harvested in 50 mM ice-cold Tris-HCl, pH 7.5, and homogenized with a homogenizer (setting 2, 30 s; Polytron; Brinkmann Instruments, Westbury, NY). The homogenate was centrifuged for 5 min at 1,000g to remove nuclei and unbroken cells. The supernatant was centrifuged at 50,000g for 10 min, and the resulting pellet was resuspended in 50 mM ice-cold Tris-HCl, pH 7.5. The protein concentration of the membrane preparation was measured by using BCA Assay Reagent (Pierce Chemical, Rockford, IL).

Histamine H₄ Receptor-Binding Studies. For saturation binding, membrane proteins (40-60 µg) were incubated in a total volume of 200 µl of 50 mM Tris-HCl, pH 7.5, with a range of [³H]histamine dihydrochloride (Amersham Biosciences, Piscataway, NJ) concentrations for 1 h at 30°C. Nonspecific binding was determined by inclusion of 1 mM histamine. The bound radioactivity was separated by filtration through polyethyleneimine-treated GF/B filters (Packard BioScience, Meriden, CT) with a Filtermate 196 harvester (Packard BioScience). The filters were washed eight times with 50 mM ice-cold Tris-HCl, pH 7.5, and radioactivity retained on the filters was measured by liquid scintillation counting with a TopCount (Packard BioScience) at 34% efficiency. All experiments were performed in triplicate. The binding data were evaluated with Prism (GraphPad Software, San Diego, CA) and analyzed for one- and two-site fits. A single-binding-site model best described all curves.

Ca²⁺ Mobilization Assay. HEK-293 SFM cells were transiently cotransfected overnight in DMEM and 10% FCS with the wild-type or mutant H₄ receptor cDNAs in pME18-CD8-Flag (0.5 µg/cm²) and the chimeric G_q/i protein cDNA in pCDNA 3 (0.05 µg/cm²) (Morse et al., 2001) using LipofectAMINE 2000 reagent (1.5 µl/cm²). Twenty-four hours after transfection, the cells were harvested and reseeded at 5 × 10⁵ cells/well in DMEM and 10% FCS in the poly(D-lysine)-treated, 96-well, clear-bottomed black plates (BD Biosciences). Forty-eight hours after transfection, the cells were loaded for 1 h with 4 µM Fluo-3AM (Molecular Probes, Eugene, OR) in loading buffer (10% FCS and 20 mM HEPES in DMEM). After being washed extensively with washing buffer (Hanks' balanced salt solution and 20 mM HEPES, pH 7.4) to remove excess dye, the cells were evaluated for agonist-induced intracellular mobilization using a Fluorescent Imaging Plate Reader (Molecular Devices, Menlo Park, CA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Prediction of Histamine H₄ Receptor Interaction Based on Receptor Modeling. The initial model of histamine docked into the hypothetical binding site in the H₄ receptor is shown in Fig. 2. Histamine is predicted to bind in a pocket formed by residues in TM3 through TM6, anchored by an ion pair between the side chain of Asp⁹⁴ (3.32) in TM3 and the cationic amino group of histamine. In TM5, Thr¹⁷⁸ (5.42) and/or Ser¹⁷⁹ (5.43) could form a hydrogen bond with the imidazole N nitrogen, whereas Glu¹⁸² (5.46) could form an ion pair with the protonated imidazole N nitrogen. Finally, Asn¹⁴⁷ (4.57) in TM4 and Ser³²⁰ (6.52) in TM6 point toward the central histamine-binding cavity and may facilitate the binding interaction.

View larger version (43K): [in this window] [in a new window]   Fig. 2.   A molecular model of human H4 receptor-histamine complex. A molecular model of the human H4 receptor was constructed from the structure of bovine rhodopsin and refined by 1000 steps of molecular mechanics minimization with the Insight II/Discover program. Asp94 (3.32) in TM3 is expected to form an ion pair to the cationic amino group of histamine. Thr178 (5.42) and Ser179 in TM5 could form a hydrogen bond to the imidazole N nitrogen. Glu182 (5.46) in TM5 could form an ion pair to the protonated imidazole N nitrogen. Asn147 (4.57) in TM4 and Ser320 (6.52) in TM6 point toward the central histamine-binding cavity.

Investigation of the Histamine-Binding Site by Site-Specific Mutagenesis. To experimentally explore the interaction of histamine with the amino acids of the H₄ receptor that our model predicted would be important, these residues were mutated individually or in combination (see Table 1 for the list of mutants). During the process of subcloning the wild-type H₄ receptor cDNA into a mammalian expression vector, we introduced a FLAG epitope at the N terminus of the receptor to facilitate examination of cell-surface expression. This construct was subsequently used to generate the mutant receptors used in this study. Transfection of the FLAG-H₄ receptor in HEK-293 SFM cells resulted in the appearance of cell-surface FLAG staining and high-affinity binding sites for [³H]histamine, with a K_D of 15.3 nM (Fig. 3; Table 1). The K_D of the FLAG-H₄ receptor obtained in this study agrees with that reported previously for the H₄ receptor without the FLAG epitope (Oda et al., 2000; Morse et al., 2001; Zhu et al., 2001). This indicates that a FLAG epitope at the N terminus of the H₄ receptor does not affect histamine binding.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Saturation binding of [3H]histamine to membranes from HEK-293 SFM cells expressing the N-terminal FLAG-tagged H4 receptor after transient transfection. Inset, Scatchard transformation of saturation binding data. Each data point is the mean of triplicate determinations and is representative of four independent experiments. The data were analyzed by nonlinear regression analysis using single-site binding models as described under Materials and Methods.

View larger version (38K): [in this window] [in a new window]   Fig. 4.   Surface expression of the FLAG-tagged H4 receptor and mutants in HEK-293 SFM cell after transient transfection. Cells were transiently transfected with cDNAs of the FLAG-tagged wild-type and mutant H4 receptors, as indicated. The transfected cells were stained first with either anti-FLAG biotinylated M2 monoclonal antibody (solid lines) or biotin mouse IgG1 (dotted lines) and then with phycoerythrin-conjugated streptavidin. The cells were analyzed using a FACSCalibur flow cytometer.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Binding of [3H]histamine to HEK-293 SFM cell membranes after transient transfection with mutated histamine H4 receptors. Data points shown are the mean of triplicate determinations and are representative of at least three independent experiments. The data were analyzed by nonlinear regression analysis (Graphpad Prism). A, Asn147 (4.57)  Tyr and Asn147 (4.57)  Ala mutants. B, Thr178 (5.42) Ala, Ser179 (5.43)  Ala, and Thr178 (5.42) Ser179 (5.43)  Ala Ala mutants. C, Glu182 (5.46)  Asn mutants. D, Ser320 (6.52)  Ala and Ser320  Phe mutants. The wild-type receptor-binding curve is shown in each figure for comparison.

Histamine-Induced Ca²⁺ Flux after Stimulation of Wild-Type and Mutant H₄ Receptors. The functional capacity of mutant H₄ receptors was examined by measuring histamine-induced Ca²⁺ mobilization in HEK-293 SFM cells transiently cotransfected with constructs expressing the receptors and a chimeric G_q/i protein, as described previously (Morse et al., 2001). In dose-response studies, cells expressing both the wild-type H₄ receptor with the N-terminal FLAG epitope and G_q/i exhibited Ca²⁺ mobilization in response to histamine treatment. The histamine dose for half-maximal response (EC₅₀) (21 ± 0.6 nM) was similar to those published previously (Oda et al., 2000; Morse et al., 2001) (Table 1; Fig. 6). As expected from the results of the [³H]histamine-binding assay, the mutant H₄ receptors that did not demonstrate binding of [³H]histamine [Asp⁹⁴ (3.32)  Ala, Asp⁹⁴ (3.32)  Glu, Asp⁹⁴ (3.32)  Asn, Glu¹⁸² (5.46)  Ala, and Glu¹⁸² (5.46)  Gln receptors] also did not stimulate histamine-induced Ca²⁺ mobilization. On the other hand, histamine-induced Ca²⁺ mobilization was observed in all the cells that expressed the mutant H₄ receptors capable of [³H]histamine binding, although the EC₅₀ varied with the different mutant receptors. The Thr¹⁷⁸ (5.42)  Ala, Ser¹⁷⁹ (5.43)  Ala, and Thr¹⁷⁸ (5.42)/Ser¹⁷⁹ (5.43)  Ala/Ala mutant H₄ receptors all demonstrated EC₅₀ values slightly higher (four times) than that of the wild-type receptor. These three mutations did not alter the maximal histamine-induced Ca²⁺ mobilization (Table 1; Fig. 6). In contrast, mutation of Glu¹⁸² (5.46)  Asp resulted in a 50% reduction in maximal Ca²⁺ mobilization by histamine and a 20-fold increase in the EC₅₀ for histamine, compared with the wild-type receptor (Table 1; Fig. 6). For the Asn¹⁴⁷ (4.57)  Tyr mutant receptor, no change in EC₅₀ was observed, although the maximal responses to histamine were reduced by about 60%. Changing the same residue to Ala, however, increased the EC₅₀ for histamine by 7-fold and decreased the maximal stimulation by 50% (Table 1; Fig. 6). Interestingly, both mutations of Ser³²⁰ (6.52) also seemed to affect the signaling of the H₄ receptor. Although mutation of this residue to Ala resulted in some reduction of histamine potency, mutation to Phe resulted in a 250-fold increase in the EC₅₀ for histamine. In addition, both mutations consistently exhibited an increase in maximal Ca²⁺ flux of 60 and 100% for both Ser³²⁰ (6.52)  Ala and Ser³²⁰ (6.52)  Phe, respectively. The increased efficacy of histamine at these mutant receptors did not seem to correlate with increased [³H]histamine binding; in experiments where binding and functional assays were carried out on the same batch of transfected cells, with the wild-type receptor assayed in parallel, the B_max values for [³H]histamine were reduced in the Ser³²⁰ (6.52)  Phe mutant receptor by 50%, whereas the maximal Ca²⁺ mobilization observed was double that observed with the wild-type receptor.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Histamine-induced Ca2+ mobilization in HEK-293 SFM cells coexpressing wild-type or mutant H4 receptors and Gq/i chimeric G-protein. Data are presented as relative change of fluorescence unit. Each data point is the mean of triplicate determinations and is representative of at least three independent experiments. A, the average maximal fluorescence values for the wild type, Thr178 (5.42)  Ala, Thr178 (5.42) Ser179 (5.43)  Ala Ala, and Ser179 (5.43)  Ala mutant H4 receptors were 3500, 3192, 3192, and 3146 units, respectively. B, the average maximal fluorescence value for the Glu182 (5.46)  Asp mutant H4 receptor was 1886 units. C, the average maximal fluorescence value for the Asn147 (4.57)  Ala and Asn147 (4.57)  Tyr mutant receptors were 1273 and 1727 units, respectively. D, the average maximal fluorescence value for the Ser320 (6.52)  Phe and Ser320 (6.52)  Ala mutant receptors were 6941 and 5716 units, respectively.

Pharmacological Analysis of Asn¹⁴⁷  Tyr and Ser³²⁰  Phe Mutant Receptors. Because the residues Asn¹⁴⁷ (4.57) and Ser³²⁰ (6.52) did not seem to contribute strongly to histamine binding but did seem to affect histamine signaling, we investigated the ability of other H₄ agonists to activate the Asn¹⁴⁷ (4.57)  Tyr and Ser³²⁰ (6.52)  Phe mutant receptors. In general, the mutant receptors responded similarly to the histamine derivatives (R)-()--methylhistamine, (S)-(+)--methylhistamine, imetit, and imepip, compared with histamine itself (Table 2; Fig. 7). Thus, the maximum response was reduced for all compounds at the Asn¹⁴⁷ (4.57)  Tyr mutant compared with the wild type, although R-()--methylhistamine was least affected. The EC₅₀ values for the agonists were reduced as well, with the exception of S-(+)--methylhistamine, which exhibited somewhat higher potency at the mutant receptor compared with the wild type (Table 2). For the Ser³²⁰ (6.52)  Phe mutation, all compounds exhibited higher maximal responses, whereas the potency of all the compounds was reduced compared with the wild type (Table 2). This mutation, however, had the greatest effect on the action of histamine compared with the other compounds.

View this table: [in this window] [in a new window]   TABLE 2 Comparison of the agonist induced intracellular Ca2+ mobilization in HEK-293 SFM cells expressing the wild-type and mutant histamine H4 receptors HEK-293 SFM cells expressing wild type, Ser320  Phe, or Asn147  Tyr mutant histamine H4 receptors were treated with increasing concentrations of the indicated histamine H4 receptor agonists. The resulting intracellular Ca2+ mobilization was measured by Fluorescent Imaging Plate Reader assay. Data were calculated as the mean ± S.E. of three independent experiments. Efficacy is expressed as percentage of the response of the wild-type receptor to each agonist.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Agonist-induced Ca2+ mobilization in HEK-293 SFM cells coexpressing wild-type or mutant (Ser320  Phe or Asn147  Tyr) H4 receptors and Gq/i chimeric G-protein. Cells transiently transfected with the indicated receptor constructs were treated with increasing concentrations of R-()--methylhistamine (A), S-(+)--methylhistamine (B), imetit (C), or immepip (D). Data are presented as relative change of fluorescence. Each data point is the mean of quadruplicate determinations and is typical of three independent experiments. The data were analyzed by nonlinear regression analysis using a sigmoidal dose-response model as described under Materials and Methods.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Recently, a fourth member of the histamine family of G-protein-coupled receptors has been identified and characterized by several groups. The H₄ histamine receptor exhibits the highest degree of similarity to the H₃ histamine receptor, and comparison of the sequences of the four histamine receptors reveals that a number of amino acids that have been implicated in histamine binding to the other receptor subtypes are conserved in the H₄ receptor. To begin to delineate the histamine-binding site on the H₄ receptor, molecular modeling and site-directed mutagenesis were carried out to determine the involvement of specific amino acid residues in histamine binding and receptor activation.

Previous mutagenesis studies on the H₁ and H₂ receptors (Gantz et al., 1992; Ohta et al., 1994), the  and  adrenergic receptors (Strader et al., 1987; Strader et al., 1988; Wang et al., 1991), and the M1 muscarinic acetylcholine receptor (Fraser et al., 1989) all argue for a critical role of a conserved aspartic acid residue in TM3 in mediating ligand binding, presumably by providing a negative counter-ion for the protonated amine group of the ligand. In the H₄ receptor, molecular modeling indicated that Asp⁹⁴ (3.32) in TM3 might serve such a role in the binding of histamine. In the present study, mutation of Asp⁹⁴ (3.32) to either Ala or Asn to eliminate the negative charge on the side chain of the amino acid, or altering the position of the carboxylate group in the side chain of Asp⁹⁴ (3.32) by mutating it to Glu, eliminated [³H]histamine binding and histamine-induced intracellular Ca²⁺ mobilization in cells expressing the mutants. Nevertheless, flow cytometric analysis indicated that each of these mutant receptor proteins was expressed on the cell surface at levels comparable with that of wild-type H₄ receptor. These results indicate that, as for other biogenic amine receptors, the conserved Asp in TM3 plays a critical role in agonist binding and receptor activation.

The fifth transmembrane domain has also been shown to play a critical role in ligand binding to histamine and other biogenic amine receptors. Asn¹⁹⁸ (5.46) in the H₁ histamine receptor or Asp¹⁸⁶ (5.42) in the H₂ histamine receptor have been proposed to interact with the protonated nitrogen atom of histamine (Ganz et al., 1992; Ohta et al., 1994). These two positions are considered to be homologous and form the basis for explaining the relatively flipped position of histamine in the H₁ and H₂ receptor. In TM5 of the human H₄ receptor, position 5.46 is occupied by a Glu residue, Glu¹⁸². Molecular modeling of the H₄ receptor indicated that this residue had the potential to interact with the N nitrogen atom of the histamine imidazole ring by either a hydrogen bond or an ion pair. Substitution of Glu¹⁸² (5.46) with Ala or Gln resulted in a mutant H₄ receptor that could no longer bind to [³H]histamine and mediate histamine signaling, even when the mutant receptors were expressed on the cell surface at roughly the same level as the wild-type receptor. Mutation of Glu¹⁸² (5.46) to Asp reduced the affinity for [³H]histamine binding and reduced the potency and efficacy of histamine-stimulated Ca²⁺ mobilization. These results indicate that Glu¹⁸² (5.46) in TM5 is another essential element of the H₄ receptor for histamine binding. The mechanism for the interaction between Glu¹⁸² (5.46) and the N nitrogen atom of histamine is likely to be an ion pair, similar to that proposed between Asp¹⁹⁰ in the TM5 of the H₂ receptor and histamine, because Asn cannot substitute for Asp at this position. The reduced affinity for histamine observed with the Asp¹⁸² mutant H₄ receptor would be consistent with this assignment, with the reduction in affinity arising from the increased distance between the negatively charged oxygen atom and the N nitrogen atom of histamine.

Other residues in TM5 have also been shown to be involved in the binding of histamine and other biogenic amines to their receptors. In the ₂ adrenergic receptor, two Ser residues in TM5, corresponding to Ser¹⁷⁹ (5.43) and Glu¹⁸² (5.46) in the H₄ receptor, have been shown to be involved in hydrogen bonding to the meta- and para-hydroxyl groups of the catechol ring of epinephrine (Strader et al., 1989a). In the ₁-adrenergic receptor, Ser¹⁸⁸ (5.42) is critical for binding to the meta-hydroxyl of the endogenous agonists (Hwa and Perez, 1996). Likewise, in the D1 dopamine receptor, mutation of either Ser¹⁹⁸ (5.42) or Ser¹⁹⁹ (5.43) to Ala disrupts agonist binding (Pollock et al., 1992). In the dopamine D2 receptor, Ser¹⁹³ (5.42) contributes notably to the binding of dopamine and Ser¹⁹⁴ (5.43) is absolutely required for activation of agonists as a result of bonding with the -hydroxyl group of catecholamines (Cox et al., 1992). For serotonin receptors, Ser 5.43 of the human 5-HT₄ was proposed to interact with serotonin through a hydrogen bond (Mialet et al., 2000); and mutation of the analogous Ser 5.43 in the rat 5-HT_2A receptor to alanine caused a 6-fold decrease in 5-HT binding affinity (Shapiro et al., 2000). In the 5-HT_1A receptor, substitution of Ser¹⁹⁸ (5.42) or Thr¹⁹⁹ (5.43) with alanine resulted in a significant reduction of serotonin binding (Ho et al., 1992). The analogous residue in the rat M₃ muscarinic receptor, Thr²³⁴ (5.42), also affects acetylcholine binding affinity and the ability of the receptor to stimulate agonist-dependent phosphatidylinositol hydrolysis (West et al., 1992). Taken together, these studies clearly demonstrate the conservation of the critical role of TM5 Ser and/or Thr residues in biogenic amine binding. In the case of histamine receptors, however, this role is not as well conserved. Although Thr¹⁹⁰ (5.46) in TM5 of the H₂ histamine receptor has been proposed to interact with the N nitrogen of the histamine imidazole ring by a hydrogen bond and seems important for establishing the kinetics of histamine binding and activation (Gantz et al., 1992), the homologous TM5 Thr (5.42) of the human and guinea pig H₁ receptors (Thr¹⁷⁴ and Thr²⁰³, respectively) are not required for histamine binding (Leurs et al., 1994; Ohta et al., 1994). In the H₄ receptor, the corresponding Thr¹⁷⁸ (5.42) and the adjacent Ser¹⁷⁹ (5.43) are predicted by computer modeling to be appropriately positioned to form a hydrogen bond with the N nitrogen of histamine. The present study demonstrates, however, that substitution of Ala at these two sites, alone or in combination, does not dramatically alter the affinity of the H₄ receptor for histamine or the ability of the mutant receptors to mediate histamine-induced signaling. Therefore, it seems that in the H₄ receptor, Thr¹⁷⁹ (5.43) and Ser¹⁷⁸ (5.42) do not play an essential role in histamine binding or signaling.

The present results demonstrate that histamine interacts with the H₄ receptor either in different orientations or by different mechanisms compared with the H₁ and H₂ receptors, due to the differences in the chemical nature and location of residues in TM5 that interact with the N nitrogen atom of the histamine imidazole ring. Thus, although Asp¹⁸⁶ (5.42) of the H₂ receptor and Glu¹⁸² (5.46) of the H₄ receptor bind the same nitrogen atom by the same mechanism, histamine must adopt a different orientation in these two receptors. In contrast, histamine binds to the H₁ and H₄ receptors in the same orientation, but Asn¹⁹⁸ (5.46) in the H₁ receptor and Glu¹⁸² (5.46) in the H₄ interact with the N nitrogen atom by either a hydrogen bond or an ion pair, respectively. Furthermore, although interactions with the potential H-bond donors/acceptors in TM5 are not essential for histamine binding to the H₁ and H₄ receptors, the homologous Thr residue (5.46) in the H₂ receptor was demonstrated to be important in the interaction with the histamine N nitrogen atom (Gantz et al., 1992).

The computer modeling studies of the H₄ receptor also revealed two additional amino acid residues, Asn¹⁴⁷ (4.57) and Ser³²⁰ (6.52), that were predicted to be in positions that could allow interaction with histamine in the predicted binding pocket. Comparison with the analogous residues in other biogenic amine receptors reveals that Asn¹⁴⁷ (4.57) in TM4 of the H₄ receptor is unique; this position is occupied by Trp, Phe, or Tyr in the H₁, H₂, and H₃ receptors, respectively. In TM6, the position occupied by Ser³²⁰ (6.52) in the H₄ receptor is generally found to be Phe in other biogenic amine receptors, except for the H₃ receptor, which has Thr at this position. In the ₂ adrenergic receptor, the Phe at the corresponding position (6.52) was suggested to be involved in forming an aromatic-aromatic interaction with the catecholamine phenyl ring of norepinephrine and important for the receptor agonist binding (Strader et al., 1989b). The adjacent Phe (6.51) in the _1B adrenergic receptor was found to be necessary not only for agonist binding but also for agonist potency and efficacy (Chen et al., 1999).

Because the affinity of the H₄ receptor for [³H]histamine was only slightly affected by mutations, as shown in our [³H]histamine-binding assay, we conclude that Asn¹⁴⁷ (4.57) and Ser³²⁰ (6.52) of the H₄ receptor are not critical for histamine binding. The same conclusion has been made regarding the corresponding serine (4.57) in TM4 of the human 5-HT₄ receptor (Mialet et al., 2000). This conclusion contrasts with our model, which suggests that increased side-chain volume at these two sites could impede histamine binding to the H₄ receptor. However, both Asn¹⁴⁷ (4.57) and Ser³²⁰ (6.52) of the H₄ receptor seem to be involved in activation of the H₄ receptor by histamine. Replacing Asn¹⁴⁷ (4.57) of the H₄ receptor with Tyr to mimic the H₃ receptor seems to be detrimental to histamine signaling through the H₄ receptor, as evidenced by a reduction of 50% in the ability of the receptor to respond to histamine with no change in levels of receptor expression. In contrast, changing Ser³²⁰ (6.52) of the H₄ receptor to Phe to imitate the H₁, H₂, and other biogenic amine receptors had the unexpected effect of greatly reducing the potency of histamine while at the same time doubling the maximal response of the receptor to histamine treatment. Although the increased efficacy could be due to an increase in receptor expression, paradoxically, this receptor mutant exhibited a reduction in maximal [³H]histamine binding (and a slight change in affinity) that did not seem to be correlated with reduced surface expression as assessed by anti-FLAG antibody staining.

The altered signaling of the Asn¹⁴⁷ (4.57)  Tyr and Ser³²⁰ (6.52)  Phe receptors was not specific for histamine. Examination of the response to several histamine analogs generally revealed the same pattern of activity at both mutant receptors as seen with histamine, albeit to differing extents. Thus, although the maximal responses for all compounds were reduced at the Asn¹⁴⁷ (4.57)  Tyr mutant, this mutation had the smallest effect on R-()-methylhistamine and the greatest effect on imepip. On the other hand, mutation of Ser³²⁰ (6.52)  Phe resulted in decreased potency and increased efficacy for all compounds, although the effects were greatest for histamine.

The observed reduction in maximal [³H]histamine binding and histamine potency at the Ser³²⁰ (6.52) mutant might be explained by altered interactions with G-proteins leading to an increase in the proportion of low-affinity histamine-binding states, which may not be detected with the ligand and conditions used in the present study. To fully address this issue, however, the development of an H₄ receptor antagonist radioligand will be required. In any case, the H₄ receptor clearly differs from the H₁ and H₂ receptors in the required chemical or physical nature of residue position 6.52.

Previous studies on rhodopsin and ₂ adrenergic receptor have suggested that a critical step in agonist-induced activation of G-protein-coupled receptors is the movement of TM6 and its cytoplasmic extension, which is important for G-protein coupling. For example, photoactivation of rhodopsin involves rotation and tilting of TM6 relative to TM3 (Farrens et al., 1996; Lin and Sakmar, 1996; Dunham and Farrens, 1999). It has been similarly suggested that TM6 of the ₂ adrenergic receptor would move in response to agonist stimulation (Javitch et al., 1997; Jensen et al., 2001). Recently, the movement of TM6 after agonist activation of the ₂ adrenergic receptor was investigated by fluorescence spectroscopy, and the results suggested clockwise rotation of TM6 and/or tilting of the cytoplasmic end of TM6 toward TM5 (Ghanouni et al., 2001).

Specific residues in TM6 have been suggested to be involved in the movement of this TM region upon agonist activation. Grånäs et al. (1998) have suggested that Phe³³¹ (6.52) in the 5HT_1B receptor, analogous to Ser³²⁰ (6.52) in the H₄ receptor, is involved in the conformational changes associated with signal transduction. In the _1B adrenergic receptor, a highly conserved Phe³⁰³ (6.44) in TM6 has been postulated to be a key residue in coupling TM6 movement to G-protein activation (Chen et al., 2000, 2002). It remains unclear whether the demonstrated effects of the mutations at the H₄ Ser³²⁰ (6.52) are due to nonspecific global or local changes in the receptor structure or disruption of particular intramolecular interactions that serve to regulate the ability of histamine to promote H₄ receptor signal transduction. Given the findings mentioned above, which suggest that agonist-induced TM6 movement for G-protein coupling critically depends on the identity of individual residues in TM6, it is possible that Ser³²⁰ (6.52) in the histamine H₄ receptor is involved TM6 movement and subsequent G-protein coupling on histamine activation. This hypothesis is currently under further investigation.

Taken together, the unique way histamine interacts with residues in the binding pocket of the H4 receptor offers potential for the development of subtype-selective antagonists. These are likely to provide further insights into the physiological role of this newly discovered histamine receptor subtype.