sulfonyl-guanidines 3.102 and 3.103. Unlike the acylguanidine group (pKa ≈ 8)7, the sulfonylguanidine group (pKa ≈ 0.4)7 is virtually uncharged at physiological pH. The sulfonylguanidine analog of UR-AK51 (3.103) showed a more than 50-fold decrease in potency at the hH4R (KB = 1,400 nM) and, contrary to UR-AK51, had no agonistic activity. The same was observed for the sulfonylguanidine 3.102 (acylguanidine analog at the hH4R: EC50 = 6.1 nM, Emax = 0.40).
3.102 and 3.103 acted as moderate inverse agonists at the hH3R and showed partial agonism at the hH2R (Emax ≈ 0.50). However, potencies were reduced by two orders of magnitude relative to the acylated guanidines at the H2R subtype (EC50 = 12,000 nM). These findings show that removing basicity from the acylguanidines is not tolerated at the different HR subtypes. In terms of hH2R activity, this can be well explained by the putative interaction hH2R model. An ionic interaction between the positively charged acylguanidine group and the highly conserved aspartate residue of helix 3 is assumed.38 Similar interactions can be expected for the other HR subtypes. Obviously, for the acylguanidines an ionic interaction is
crucial for potent hH2/3/4R binding since the structurally closely related, non-protonated sulfonylguanidine analogs 3.102 and 3.103 display drastically reduced potencies. In contrast, likewise non-basic cyanoguanidines such as UR-PI376 are able to potently activate the hH4R. Apparently, for these cyanoguanidine-type compounds an ionic interaction with the receptor is not required. Therefore, distinct binding modes for the acylguanidine-type and cyanoguanidine-type agonists at the hH4R can be presumed. Nevertheless, it has to be considered that next to the physicochemical differences of the acylguanidines and sulfonyl-guanidines also steric factors, as for example the different orientation of the oxygen atoms, may play a role in altering the pharmacological activities at the HR subtypes.
Cyanoguanidine-type H4R agonists 57
Cyanoguanidine-type H4R agonists 59
3.3.2 Affinities of UR-PI376 (3.66) for the hH1R, hH2R, hH3R and hH4R subtypes in radioligand binding experiments
Figure 3.4. Displacement of [3H]mepyramine (5 nM), [3H]tiotidine (10 nM), [3H]Nα-methylhistamine (3 nM) and [3H]histamine (10 nM) with UR-PI376 from Sf9 insect cell membranes expressing the hH1R + RGS4, hH2R-GsαS fusion protein, hH3R + Giα2 + Gβ1γ2 + RGS4 or hH4R-RGS19 fusion protein + Giα2 + Gβ1γ2. Radioligand binding was determined as described under Pharmacological methods. Data were analyzed for best fit to one site (monophasic) competition curves. Data points shown are the means of n independent experiments each performed in duplicate.
UR-PI376 (3.66) was identified as the most potent and selective hH4R agonist in the cyano-guanidine series by using steady-state GTPase assays (EC50 = 34 nM, Emax = 0.93, ≈ 30-fold selectivity toward the hH3R). Nevertheless, agonist potencies determined in functional assays depend on different factors as for example the G-protein availability.6 This can cause deviations between potencies received from functional experiments and binding studies. For this reason dissociation constants (KI values) of UR-PI376 at the different hHR subtypes were determined. UR-PI376 was found to displace [3H]mepyramine from the hH1R, [3H]tiotidine from the hH2R, [3H]Nα-methylhistamine from the hH3R and [3H]histamine from the hH4R giving monophasic competition binding curves (Figure 3.4). In accordance with the results from the GTPase assay, UR-PI376 bound with the highest affinity to the hH4R (KI = 58 nM) and, as expected, showed a remarkably lower affinity for the hH3R (KI = 530 nM).
However, the affinity of UR-PI376 for the hH4R was slightly reduced compared to the EC50 value determined in the GTPase activity assay, while the contrary was observed for the affinity of UR-PI376 to the hH3R. Therefore, selectivity of UR-PI376 for the hH4R over the hH3R is about three times lower than in the functional experiments. At the hH1R and hH2R, the compound displayed low affinity resulting in a 450- and 70-fold selectivity for the hH4R,
KI (UR-PI376) [nM] n
Cyanoguanidine-type H4R agonists 61
respectively. Overall, the determined KI values and their rank orders are in agreement with the potencies evaluated in the functional GTPase assays and confirm UR-PI376 to be a highly affine and selective ligand for hH4R.
3.3.3 Inhibition of the UR-PI376 (3.66) stimulated GTP hydrolysis at the hH4R by standard H4R antagonists
Figure 3.5. Inhibition of the UR-PI376 stimulated GTP hydrolysis at the hH4R by the H4R antagonists thioperamide, iodophenpropit and JNJ 7777120. Steady-state GTPase activity in Sf9 insect cell membranes expressing the hH4R-RGS19 fusion protein + Giα2 + Gβ1γ2 was determined as described under Pharmacological methods. Reaction mixtures contained 1 µM of UR-PI376. Data were analyzed by nonlinear regression and were best fit to sigmoid concentration/response curves. Data points shown are the means of two independent experiments each performed in duplicate.a KI values of the reference H4R antagonists were taken from Lim et al.1
The functional pharmacological activities of all compounds were determined in steady-state GTPase activity assays using membranes of Sf9 insect cells expressing the respective HR subtype. When using membranes instead of intact cells, G-proteins are directly accessible to the evaluated compounds. Therefore, the possibility of direct, receptor independent G-protein activation has to be taken into account. For many compounds like peptides, local anesthetics, β-adrenoceptor antagonists but also for cationic-amphiphilic HR ligands, direct G-protein activation has been reported.41-43 UR-PI376 with its polar and basic imidazole ring and more lipophilic side chain somehow features cationic-amphiphilic properties. To exclude a direct G-protein activation, GTPase activity was stimulated with UR-PI376 (1 µM) and the effect of increasing concentrations of the H4R antagonists thioperamide, iodophenpropit and JNJ7777120 was evaluated. As shown in Figure 3.5, all antagonists suppressed the UR-PI376 induced GTP hydrolysis in a dose-dependent manner. Thioperamide was more
Compound KB [nM] KI [nM]a Thioperamide 62 ± 4.6 125 Iodophenpropit 30 ± 2.4 13 JNJ 7777120 12 ± 2.5 16
-10 -8 -6 -4 -2
2 3 4 5
Thioperamide Iodophenpropit JNJ 7777120
H4R antagonist (log M) GTPase activity [pmol min-1 mg-1 ]
effective than the other standard H4R antagonists due to its pronounced inverse agonistic activity at the hH4R (Table 3.1). The KB values determined for thioperamide, iodophenpropit and JNJ 7777120 against UR-PI376 in the functional assay were in good accordance with KI
values from binding studies reported in literature (Figure 3.5). These results indicate the H4R antagonists to compete with UR-PI376 for the same binding site at the hH4R and suggest the GTPase activation through UR-PI376 to be receptor mediated. Direct G-protein activation by HR receptor ligands was reported to occur at concentrations > 10 µM.41-43 By contrast, UR-PI376 stimulated GTPase activation in membranes expressing the hH4R was detectable at concentrations ≥ 10 nM. The inverse agonistic activity of UR-PI376 at the hH3R also supports a receptor dependent G-protein activation as both the hH3R and hH4R were coexpressed with Giα2. Finally, UR-PI376 was found to displace [3H]histamine from the hH4R (Figure 3.4). These data confirm UR-PI376 to act as a hH4R agonist in the GTPase assay, whereas direct G-protein stimulation can definitely be ruled out.
3.3.4 Potencies and efficacies of selected compounds at the guinea pig ileum (gpH1R) and guinea pig right atrium (gpH2R)
Selected compounds were investigated on the isolated guinea pig (gp) ileum for H1R activity and on the isolated spontaneously beating guinea pig right atrium for H2R activity using histamine as the reference agonist (Table 3.2).
The sulfonylguanidines 3.102 and 3.103 both showed very poor activities at the gpH2R.
Contrary to 3.102, which was a very weak antagonist, the sulfonyl analog of UR-AK51 (3.103) exerted weak partial agonism (Emax = 0.38). As observed for the hH2R in the GTPase assay, replacing the basic acylguanidine moiety with a non-basic sulfonylguanidine group was not tolerated at gpH2R. The potent cyanoguanidine-type hH4R agonist UR-PI376 (3.66) was a very poor gpH2R antagonist. This result is consistent with the poor hH2R activity determined in the GTPase assay. At the guinea pig ileum, all evaluated compounds acted as weak gpH1R antagonists.
Cyanoguanidine-type H4R agonists 63
Table 3.2. Pharmacological activities of selected compounds at the guinea pig ileum (gpH1R) and the guinea pig right atrium (gpH2R).
a number of experiments,b pEC50 were calculated from the mean shift ΔpEC50 of the agonist curve relative to the histamine reference curve by equation: pEC50 = 6.00 + 0.13 + ∆pEC50. Summand 0.13 represents the mean desensitization observed for control organs when two successive curves for histamine were performed (0.13 ± 0.02, n = 16). The SEM given for pEC50 is the SEM calculated for
∆pEC50, c efficacy, maximal response (%), relative to the maximal increase in heart rate induced by the reference compound histamine; d Emax (histamine) at 50 µM 3.66: 0.80 ± 0.01; e Emax (histamine) at 30 µM 3.102: 0.93 ± 0.02; n.d.: not determined.
3.3.5 Summary and Conclusion
Starting from the NG-acylated imidazolylpropylguanidine UR-AK51, which is lacking HR subtype selectivity and shows high activities at hH2/3/4Rs, several classes of compounds with structurally related polar central groups were prepared. The aim was to obtain potent hH4R agonists with improved selectivity compared to UR-AK51 and to acquire information about structure-activity and structure-selectivity relationships of the compounds at the distinct HR subtypes.
Replacing the acylguanidine moiety in UR-AK51 with a cyanoguanidine group produced 3.42 that exhibited moderate partial agonism at the hH4R. Altering the chain length between the imidazole ring and the cyanoguanidine group revealed a tetramethylene spacer to be optimal for high potency and efficacy at the hH4R. By contrast, two- and five-membered carbon chains were not tolerated with respect to hH4R agonistic activity. The residue in the “eastern part” of the imidazoylbutylcyanoguanidines was found to be sensitive toward variations.
Minor modifications, for example altering the chain length or fluorine substitution of the phenyl ring, substantially reduced hH4R activity. Otherwise, introduction of small alkyl residues was well accepted and revealed that larger substituents in the “eastern part” of the compounds are required for selectivity over the hH3R. At the hH1R and hH2R, most cyano-guanidines showed negligible efficacies and just low potencies. Compared to hH1R and hH2R higher activities were observed for the hH3R, but almost all imidazolylbutylcyanoguanidines were most potent at the hH4R. Introduction of a methyl group at position 5 of the imidazole ring resulted in nearly total loss of hH4R activity. The possibility of direct, receptor
gpH1R gpH2R