N¹-(3-Cyclohexylbutanoyl)-N²-[3-(1H-imidazol-4-yl)propyl]guanidine (UR-AK57), a Potent Partial Agonist for the Human Histamine H₁- and H₂-Receptors

Histamine (HIS) (1) (see Fig. 1) is a neurotransmitter and autacoid and acts through H₁-, H₂-, H₃-, and H₄-receptors (H₁R–H₄R) (Hill et al., 1997; Hough, 2001; Bakker et al., 2002). The H₁R couples to G_q-proteins to mediate phospholipase C activation and plays a role in the regulation of alertness and as mediator of type 1 allergic reactions (Hill et al., 1997; Bakker et al., 2002). The H₂R couples to G_s-proteins to mediate adenylyl cyclase activation and regulates H⁺ secretion in gastric parietal cells, cardiac contractility, and various myeloid cell functions (Klinker et al., 1996; Hill et al., 1997; Bakker et al., 2002).

It has been difficult to establish relevant native test systems for the analysis of the human H₁R (hH₁R) and human H₂R (hH₂R), because there are unexplained pharmacological differences in the properties of hH₁R and hH₂R in native cells relative to standard guinea pig test organs (Burde et al., 1990; Seifert et al., 1994; Klinker et al., 1996). To facilitate the comparison of histamine receptors under identical experimental conditions, we established expression systems for the H₁R and H₂R in Sf9 insect cells (Kelley et al., 2001; Houston et al., 2002). Sf9 cells express the H₁R and H₂R at high levels and can be cultured in large quantities. GPCR/G-protein coupling in Sf9 membranes is monitored with high sensitivity using the steady-state GTPase assay. This assay assesses GPCR/G-protein coupling at a proximal point of the signaling cascade, avoiding potential bias introduced by assessing more downstream events such as effector activation or changes in gene expression. For the H₁R, coupling of the GPCR to insect cell G_q-proteins is determined, and the GTPase signal is amplified by RGS proteins (Houston et al., 2002; Seifert et al., 2003). For the H₂R, fusion proteins of GPCR and mammalian G_sα proteins ensure defined 1:1 stoichiometry of the coupling partners and their efficient interaction (Seifert et al., 1999; Kelley et al., 2001). By measuring GTP hydrolysis, potencies and efficacies of H₂R agonists are assessed in an expression level-independent manner (Seifert et al., 1999; Kelley et al., 2001; Wenzel-Seifert et al., 2001).

Both H₁R and H₂R agonists are important pharmacological tools for studying the role of the H₁R and H₂R, respectively, in (patho)physiological processes (Bakker et al., 2002; Dove et al., 2004; Pertz et al., 2004). H₁R agonists are divided into three classes: 1) small agonists derived from HIS, such as 2-methylhistamine and 2-(2-thiazolyl)ethanamine, 2) HIS derivatives with a bulkier aromatic substituent at position 2 of the imidazole ring, such as 2-(3-bromophenyl)histamine, and 3) the histaprodifens (Bakker et al., 2002; Pertz et al., 2004). Unfortunately, bulky H₁R agonists exhibit considerably lower potency and efficacy at the hH₁R than at the guinea pig H₁R (gpH₁R), limiting their usefulness as tools for studying the hH₁R (Seifert et al., 2003). The molecular basis for the differences in pharmacological properties between hH₁R and gpH₁R has recently been elucidated (Bruysters et al., 2005). A further complication is that, at concentrations in the range of 10 μM to 1 mM, 2-phenylhistamines may activate G-proteins directly, i.e., in a receptor-independent manner (Seifert et al., 1994; Hagelüken et al., 1995; Klinker et al., 1996).

H₂R agonists are divided into two classes: 1) small agonists derived from HIS (1), such as dimaprit and amthamine, 2) long-chained and more bulky molecules, such as the guanidines arpromidine and impromidine (Bakker et al., 2002; Dove et al., 2004), and 3) the recently introduced N^G-acylated imidazolylpropylguanidines (AIPGs), which are less basic than guanidines (Xie et al., 2006). Similar to the situation with H₁R species isoforms, bulky H₂R agonists are considerably less potent and efficacious at hH₂R than at gpH₂R, reducing their value as probes to examine hH₂R (Kelley et al., 2001; Wenzel-Seifert et al., 2001; Xie et al., 2006). The pharmacological differences between hH₂R and gpH₂R are attributable to two amino acid differences in transmembrane (TM) domains 1 and 7 (Kelley et al., 2001; Dove et al., 2004).

The AIPGs characterized so far possess two ring systems, i.e., two phenyl rings, a phenyl and a pyridyl ring, a phenyl and an imidazolyl ring, or a phenyl ring and a thiazole ring (Xie et al., 2006). In our present study, we examined AIPGs substituted with a single phenyl ring (2–8) or a single cyclohexyl ring (9–14), possessing various linker lengths and alkanoyl chain branching between the acylguanidine moiety and the ring system (Fig. 1). Within this series of AIPGs, N¹-(3-cyclohexylbutanoyl)-N²-[3-(1H-imidazol-4-yl)propyl-]guanidine (UR-AK57; 14) is the most potent hH₂R agonist identified so far, and surprisingly, this compound is also a potent partial hH₁R agonist.

Materials and Methods

Materials. Construction of baculoviruses encoding hH₂R-G_sαS, gpH₂R-G_sαS, hH₁R, and gpH₁R was described previously (Kelley et al., 2001; Seifert et al., 2003). Baculoviruses encoding RGS proteins 4 and 19 were a gift from Dr. E. Ross (Department of Pharmacology, University of Southwestern Medical Center, Dallas, TX). AIPGs 2-14 were prepared according to the procedure described by Ghorai (2005). Structures of synthesized compounds were confirmed by ¹H NMR spectroscopy and high-resolution mass spectrometry. Purity of compounds was >98% as determined by high-performance liquid chromatography or capillary electrophoresis (Schuster et al., 1997). Stock solutions of compounds 2-14 (10 mM) were prepared in dimethyl sulfoxide and stored at –20°C. Under these conditions, compounds were stable for at least 2 years (longer periods of time were not studied). Further dilutions of compounds 2-14 were prepared in distilled water. Sources of other materials are described elsewhere (Kelley et al., 2001; Houston et al., 2002; Seifert et al., 2003). Baculovirus infection and culture of Sf9 cells and membrane preparation were performed as described previously (Kelley et al., 2001). H₂R-G_sα expression levels were 5 to 6 pmol/mg as assessed by immunoblotting using the M1 monoclonal antibody and β₂-adrenoceptor expressed at defined levels as standard (Kelley et al., 2001). H₁R expression levels were 4 to 6 pmol/mg as assessed by [³H]mepyramine saturation binding (Seifert et al., 2003).

Steady-State GTPase Activity Assay. GTP hydrolysis in Sf9 membranes expressing H₂R-G_sα fusion proteins or H₁R isoforms plus RGS proteins was determined as described previously (Kelley et al., 2001; Seifert et al., 2003). In brief, assay tubes (100 μl) contained Sf9 membranes (10 μg of protein/tube); various ligands; 1.0 mM MgCl₂, 0.1 mM EDTA, 0.1 mM ATP, 100 nM GTP, 1 mM adenylyl imidodiphosphate, 5 mM creatine phosphate, 40 μg of creatine kinase, and 0.2% (w/v) bovine serum albumin in 50 mM Tris/HCl, pH 7.4; and [γ-³²P]GTP (0.2–0.5 μCi/tube). Reactions were conducted for 20 min at 25°C and terminated by the addition of a 900-μl slurry consisting of 5% (w/v) activated charcoal and 50 mM NaH₂PO₄, pH 2.0. ³²P_i in supernatant fluids of reaction mixtures was determined by liquid scintillation counting.

[³H]Mepyramine Binding Assay. [³H]Mepyramine competition binding experiments with Sf9 membranes expressing hH₁R or gpH₁R plus RGS proteins were performed as described previously (Seifert et al., 2003). In brief, assay tubes (500 μl) contained membranes (20–25 μg of protein/tube), 2 nM [³H]mepyramine, and unlabeled ligands in binding buffer (12.5 mM MgCl₂, 1 mM EDTA, and 75 mM Tris/HCl, pH 7.4). Bound radioligand was separated from free radioligand by filtration through GF/C filters, and filter-bound radioactivity was determined by liquid scintillation counting.

Miscellaneous. Protein concentrations were determined using the Bio-Rad DC protein assay kit (Bio-Rad, Hercules, CA). All analyses of experimental data were performed with the Prism 4.02 software (GraphPad Software, Inc., San Diego, CA). K_i and K_B values were calculated using the Cheng and Prusoff equation (Cheng and Prusoff, 1973). Statistical comparisons in Table 1 were performed with the Student's t test.

Results

Agonist Potencies and Efficacies of AIPGs 2–14 at hH₂R-G_sαS and gpH₂R-G_sαS in the GTPase Assay. In membranes expressing hH₂R-G_sαS (Fig. 2A) and gpH₂R-G_sαS (Fig. 2B), HIS activated GTP hydrolysis with an EC₅₀ value of ∼1 μM. UR-AK57 was a 50-fold more potent agonist (EC₅₀, 23 nM) at hH₂R than HIS. At gpH₂R, UR-AK57 activated GTP hydrolysis 130-fold more potently (EC₅₀, 9 nM) than HIS. At hH₂R, UR-AK57 was almost a full agonist (E_max, 0.87), and at gpH₂R, UR-AK57 was a full agonist (E_max, 1.11). UR-AK57 constitutes the most potent hH₂R agonist identified so far and surpasses the previous leader, UR-PG55B, which is substituted with two p-fluorophenyl groups, in terms of potency by 2-fold (Xie et al., 2006). Moreover, UR-AK57 clearly surpasses UR-PG55B (E_max, 0.61) in terms of efficacy (Xie et al., 2006). Shortening of the connecting chain (13 versus 14) reduced potency and efficacy. Exchange of the 3-cyclohexylbutanoyl group in compound 14 against a 4-cyclohexylbutanoyl residue (12) had little effect on efficacy and potency; the same was true for shortening of the connecting alkanoyl chain between the guanidine moiety and the cyclohexyl ring (12→11→10→9).

Exchange of cyclohexyl against phenyl (14→8) reduced hH₂R potency without affecting efficacy. Shortening of the connecting chain (8→7) was also well tolerated. In the series of compounds with a phenyl ring and a connecting chain ranging from tetramethylene to none (6→2), minor changes in potency, with the exception of compound 2, and variable effects on efficacy were noted.

Overall, as is true for guanidines (Kelley et al., 2001) and AIPGs with two aromatic ring systems (Xie et al., 2006), AIPGs with a single ring system exhibited higher potencies and efficacies at gpH₂R-G_sαS than at hH₂R-G_sαS (Table 1 and Fig. 3). However, among the series of aryl/diarylalkylguanidines (“guanidines”), AIPGs with two aromatic substituents and AIPGs with one substituent, the systematic difference between hH₂R and gpH₂R was the smallest [compare Tables 1 and 3 of this study with Fig. 6 and Table 2 in Kelley et al. (2001) and Fig. 2 and Table 1 in Xie et al. (2006)]. Most notably, among compounds 2–14, six derivatives (46% of the compound pool) (3, 4, 6, 9, 11, and 12) exhibited potencies that varied by just 2-fold between hH₂R and gpH₂R, whereas for guanidines, only one of nine compounds (11% of the compound pool) (BU-E-43) fell in this group (Kelley et al., 2001). For AIPGs substituted with two aromatic ring systems, just three of 12 compounds (25% of the compound pool) (UR-PG137, UR-PG55B, and UR-PG153) were in this range (Xie et al., 2006).

Structure-Activity Relationships for the Partial hH₁R Agonism and gpH₁R Antagonism of AIPGs 2–14 in the GTPase assay. In membranes expressing hH₁R (Fig. 2C) and gpH₁R (Fig. 2D), HIS activated GTP hydrolysis with an EC₅₀ value of ∼200 nM. At hH₁R, UR-AK57 was a similarly potent agonist (EC₅₀, 280 nM) as HIS. The stimulatory effect of UR-AK57 on GTP hydrolysis catalyzed by gpH₁R was too small to assess potency. At hH₁R, UR-AK57 was a moderately strong partial agonist (E_max, 0.56), and at gpH₁R, UR-AK57 was only a very weak partial agonist (E_max, 0.13). Among all of the AIPGs studied, compound 14 was the most potent and efficacious partial hH₁R agonist. Chain shortening (9, 10, and 13), chain elongation (12), and methyl group removal (11) reduced agonist potency and efficacy. Exchange of the cyclohexyl ring (9-14) against a phenyl ring (2-8) reduced hH₁R agonism as well. Collectively, a 3-substituted butanoyl moiety connecting the guanidino group and the cyclohexyl ring is favorable for hH₁R agonism.

Most AIPGs, particularly compound 14, exhibited much lower efficacies at gpH₁R than at hH₁R (Table 2). In fact, AIPGs were gpH₁R antagonists with affinities in the range of 0.5 to 2 μM. UR-AK57 (14) exhibited ∼3-fold higher affinity for hH₁R than for gpH₁R in the GTPase assay. Other AIPGs exhibited up to 15-fold higher affinity for gpH₁R than for hH₁R.

Inhibition of UR-AK57 (Compound 14)-Stimulated GTP Hydrolysis at hH₁R by H₁R Antagonists. H₁R agonists of the 2-phenylistamine class are cationic-amphiphilic compounds and efficient direct G-protein activators in some systems (Seifert et al., 1994; Hagelüken et al., 1995). To exclude direct G-protein activation as mechanism for the GTPase stimulation by UR-AK57, which is cationic-amphiphilic as well, we studied the effects of the first-generation H₁R antagonists mepyramine, promethazine, diphenhydramine, triprolidine, and cyproheptadine, as well as the second-generation H₁R antagonists terfenadine and fexofenadine, on GTP hydrolysis stimulated by UR-AK57 (1 μM) (Fig. 4). H₁R antagonists inhibited GTP hydrolysis in the order of potency: promethazine > cyproheptadine > triprolidine > mepyramine > diphenhydramine > terfenadine ≫ fexofenadine. This order of potency fits exactly to the one observed for HIS-stimulated GTP hydrolysis at hH₁R (Seifert et al., 2003).

Affinities of HIS and AIPGs for hH₁R and gpH₁R in [³H]Mepyramine Competition Binding Experiments. In the GTPase assay, we determined the agonist potencies of AIPGs at hH₁R, but for the gpH₁R, antagonist affinities had to be determined. Because agonist potencies depend on several factors, including G-protein availability, those values cannot directly be compared with antagonist potencies (Seifert et al., 1999). Therefore, we compared potencies of representative AIPGs in the [³H]mepyramine competition binding assay. All compounds inhibited [³H]mepyramine binding according to monophasic isotherms (Table 3) that were insensitive to guanine nucleotides (data not shown). The latter findings indicate that ternary complex formation is not detected in this system, probably because of the low expression level of the insect G_q-protein (Houston et al., 2002). Among all of the compounds studied, UR-AK57 (14) exhibited the highest affinity for hH₁R. Chain shortening between the guanidino group and the cyclohexyl group (10 and 13) and substitution of the cyclohexyl ring by a phenyl ring (5, 6, and 8) were unfavorable, whereas hH₁R tolerated chain elongation (12). At gpH₁R, UR-AK22 (6) exhibited the highest affinity within the compound pool, whereas UR-AK57 (14) ranged among the low-affinity compounds at this receptor. gpH₁R tolerated a phenyl ring (5, 6, and 8), a methylene linker (10), and a trimethylene linker (12) better than hH₁R. In contrast, hH₁R tolerated the methyl-branched chain (13 and 14) better than gpH₁R.

Discussion

Historically, the availability of a generally applicable and reliable analysis system for the hH₂R was a substantial problem (Klinker et al., 1996; Dove et al., 2004). During the past years, our group has established fusion proteins of the hH₂R and G_sα as the standard model for the analysis of both agonists and antagonists (Kelley et al., 2001; Wenzel-Seifert et al., 2001; Houston et al., 2002; Xie et al., 2006). UR-AK57 (14) is the most potent hH₂R agonist known so far, and among all of the bulky hH₂R agonists examined, it also exhibits one of the highest efficacies (Table 1) (Kelley et al., 2001; Xie et al., 2006). Thus, a 3-substituted butanoyl moiety connecting the cyclohexyl substituent and the guanidino group is optimal in affording high potency and efficacy at hH₂R. Probably, the cyclohexyl ring and butanoyl moiety of UR-AK57 form hydrophobic interactions with amino acid residues in TM3, 6, and 7, and Ala-271 in TM7 of hH₂R may be of particular importance in this respect (Kelley et al., 2001). It is noteworthy that hH₂R tolerated alterations of linker length between the acylguanidino group and the phenyl or cylohexyl ring quite well (Table 1), indicative of conformational flexibility of hH₂R with this particular compound class. With diarylalkylguanidines as ligands, hH₂R exhibited lower overall conformational flexibility than gpH₂R (Kelley et al., 2001), but those guanidines are also bulkier than the AIPGs studied herein. Thus, among the guanidines and AIPGs studied so far, UR-AK57 possesses the optimal properties in terms of hH₂R potency and efficacy.

The recently studied AIPGs with two aromatic ring substituents surpass UR-AK57 in terms of potency at gpH₂R-G_sαS [EC₅₀ of UR-AK57 (9 nM) versus EC₅₀ of UR-PG80 (6 nM)]. The opposite is true for hH₂R [EC₅₀ of UR-AK57 (23 nM) versus EC₅₀ of UR-PG80 (78 nM)]. These data support the notion that hH₂R accommodates the 3-cyclohexylbutanoyl moiety of UR-AK57 particularly well. It is conceivable that Ala-271 is crucial in mediating the high-affinity interactions of phenyl- and cyclohexyl-substituted AIPGs with hH₂R, whereas gpH₂R bears an aspartate residue at this position, rendering hydrophobic interactions impossible (Kelley et al., 2001). However, in gpH₂R, alternative hydrophobic interactions of AIPGs with other as yet unidentified amino acids in TM3, 6, and 7 must take place since those compounds exhibit high affinity for gpH₂R as well.

Bulky guanidines are moderately potent H₁R antagonists, with arpromidine exhibiting a K_i value of 33 nM at gpH₁Rin the [³H]mepyramine competition binding assay (Seifert et al., 2003). At hH₁R, arpromidine is 10-fold less potent than at gpH₁R (K_i, 350 nM) (Seifert et al., 2003). Structural differences in TM2 play a crucial role for the differences in affinity of guanidines at the two H₁R receptor isoforms (Bruysters et al., 2005). The exchange of the guanidino group against an acylguanidino group decreases affinity for gpH₁R ∼300-fold (K_i of UR-PG136 in the [³H]mepyramine competition binding assay, 9.6 μM). Similar changes were observed for arpromidine versus UR-PG136 at hH₁R (Xie et al., 2006). These data show that AIPGs ensure excellent selectivity (>100-fold) for H₂R isoforms relative to H₁R isoforms. However, although in terms of affinity for H₁R isoforms AIPGs substituted with two aromatic ring systems were not particularly interesting, we noted that those compounds were weak partial hH₁R agonists with preference for gpH₁R in terms of agonist efficacy (Xie et al., 2006).

Based on those observations, we examined the effects of AIPGs substituted with a single aromatic/aliphatic substituent on H₁R isoforms. Unexpectedly, UR-AK57 turned out to be a moderately strong partial hH₁R agonist exhibiting a potency that approaches that of HIS. In fact, in terms of efficacy (E_max, 0.56) and potency (EC₅₀, 280 nM) at hH₁R, UR-AK57 is comparable to the most potent derivatives of the 2-phenylhistamine class, which are classic H₁R agonists (Bakker et al., 2002; Pertz et al., 2004). Specifically, 2-(3-bromophenyl)histamine exhibits an efficacy of 0.73 and a potency of 210 nM at hH₁R (Seifert et al., 2003). In marked contrast, UR-AK57 is only a very weak partial agonist at gpH₁R, with lower apparent affinity than for hH₁R in GTPase experiments (Table 2). Thus, UR-AK57 is the first synthetic H₁R agonist with higher potency and efficacy for hH₁R than gpH₁R.

Because AIPGs are cationic-amphiphilic and compounds with such properties can activate G-proteins directly (Seifert et al., 1994; Hagelüken et al., 1995), it was important to exclude the possibility of direct G-protein activation by AIPGs. Direct G-protein-stimulatory effects of histamine receptor ligands are usually observed at concentrations of >10 to 100 μM (Seifert et al., 1994; Hagelüken et al., 1995), but the stimulatory effects of UR-AK57 on GTP hydrolysis in membranes expressing hH₁R were already apparent at a concentration as low as 100 nM (Fig. 2C). The different concentration ranges argue against direct G-protein stimulation playing a part in the GTPase activation in hH₁R-expressing Sf9 membranes. The largely different effects of UR-AK57 on GTPase activity in Sf9 membranes expressing hH₁R and gpH₁R (Fig. 2, compare C with D) also corroborate the notion that the stimulatory effects of the compound on hH₁R are not due to direct G-protein activation, because Sf9 membranes harboring hH₁R and gpH₁R express the same type of endogenous G_q-protein (Houston et al., 2002). Finally, the studies with H₁R antagonists (Fig. 4) provided definitive proof that the pronounced stimulatory effect of UR-AK57 on GTPase activity is due to H₁R activation and not due to receptor-independent G-protein activation.

In the [³H]mepyramine competition binding assay, UR-AK57 exhibited similar potency at hH₁R and gpH₁R (K_i, ∼1 μM) (Table 3). The higher apparent affinity of the compound for hH₁R in the GTPase assay (EC₅₀, 280 nM) (Table 2) is probably due to the fact that, in those studies, only the high-affinity (G-protein-coupled) UR-AK57-liganded hH₁Ris assessed, whereas in the [³H]mepyramine competition binding assay, only the low-affinity (G-protein-uncoupled) UR-AK57-liganded hH₁R is assessed. This affinity difference between the two assays probably reflects the relative paucity of available G-proteins as coupling partners, which are detected with greater sensitivity in the GTPase assay than in the binding assay. The similar affinity of UR-AK57 at gpH₁R in the GTPase and [³H]mepyramine competition binding assays compared with the different apparent affinities of this compound in the corresponding assays with hH₁R (Tables 2 and 3) further support the notion of a specific agonist action of UR-AK57 on hH₁R.

The structure-activity relationships of AIPGs for interaction with hH₁R and gpH₁R are different in terms of agonist efficacy and affinity in the GTPase and [³H]mepyramine competition binding assay (Tables 2 and 3). Most importantly, a 3-substituted butanoyl moiety as is present in 14 is favorable for hH₁R agonism. These data indicate that it may become possible to synthesize bulky H₁R agonists with even greater preference for hH₁R relative to gpH₁R than UR-AK57.

Our present study demonstrates that the notion of bulky agonists exhibiting higher potencies and efficacies at gpH₁R and gpH₂R than at hH₁R and hH₂R, respectively (Kelley et al., 2001; Seifert et al., 2003; Xie et al., 2006), is actually not true. Specifically, several AIPGs substituted with a phenyl or cyclohexyl ring exhibit similar potencies at hH₂R and gpH₂R and include the most potent hH₂R agonist identified so far (Fig. 3 and Table 1). In terms of efficacy at hH₂R, UR-AK57 comes close to a full agonist as well. Most strikingly, UR-AK57 is also a potent and moderately strong partial hH₁R agonist, with much higher efficacy than at gpH₁R (Fig. 2, C and D). Thus, UR-AK57 constitutes an interesting starting point for the development of potent and efficacious hH₂R and hH₁R agonists.

Not only may H₂R agonists be a good starting point for the development of H₁R agonists but, conversely, H₁R agonists may also serve as template for the development of H₂R agonists. This notion is supported by the finding that N^α-(imidazolylethyl)histaprodifen, originally synthesized as H₁Ragonist (Pertz et al., 2004), is a potent partial hH₁R agonist (EC₅₀, 0.24 μM; E_max, 0.84) and a potent partial hH₂R agonist (EC₅₀, 0.57 μM; E_max, 0.39) (Seifert et al., 2003). Our present data emphasize the importance of examining all potential ligands for the H₁R and H₂R both in the agonist and antagonist mode for each receptor subtype and species isoform and not to extrapolate the putative ligand properties from previous studies obtained, even with closely related compounds. We assume that the numerous compounds designed for agonistic activity at H₁R and H₂R (Bakker et al., 2002; Pertz et al., 2004; Dove et al., 2004) still hold many surprising pharmacological properties that have been missed so far because of incomplete analyses. In future studies, we will systematically analyze agonist and antagonist effects of guanidines, AIPGs, and histaprodifens at H₁R and H₂R species isoforms. In this analysis, we will include the human and guinea pig histamine receptor and the rat receptor as recent data point to unique pharmacological properties of the rat H₂R (Xie et al., 2006). In terms of future compound synthesis, pharmacophoric elements of the histaprodifens and 2-phenylhistamines will be combined with structural elements of compounds 2–14. Finally, the compounds analyzed in this paper will have to be examined in native systems.