Novel chalcone-based fluorescent human histamine h(3) receptor ligands as pharmacological tools.

Novel fluorescent chalcone-based ligands at human histamine H(3) receptors (hH(3)R) have been designed, synthesized, and characterized. Compounds described are non-imidazole analogs of ciproxifan with a tetralone motif. Tetralones as chemical precursors and related fluorescent chalcones exhibit affinities at hH(3)R in the same concentration range like the reference antagonist ciproxifan (hH(3)R pK(i) value of 7.2). Fluorescence characterization of our novel ligands shows emission maxima about 570 nm for yellow fluorescent chalcones and ≥600 nm for the red fluorescent derivatives. Interferences to cellular autofluorescence could be excluded. All synthesized chalcone compounds could be used to visualize hH(3)R proteins in stably transfected HEK-293 cells using confocal laser scanning fluorescence microscopy. These novel fluorescent ligands possess high potential to be used as pharmacological tools for hH(3)R visualization in different tissues.

Introduction

Histaminergic receptors belong to class A of membrane bound G-protein-coupled receptors (GPCRs). They consist of four subtypes, the histamine H1, H2, H3, and H4 receptors (Walter and Stark, 2012). The H3 receptor has a neurotransmitter function. High receptor densities could be found in different areas of the central nervous system (Martinez-Mir et al., 1990; Sander et al., 2008). Possible indications of H3 receptor antagonists/inverse agonists could be the treatment of cognitive and sleep disorders as well as schizophrenia, epilepsy, adipositas, and neuropathic pain (Girard et al., 2004). To get information on etiopathology and accumulation or depletion of human histamine H3 receptors (hH3R) and to accelerate the clinical development of pharmaceuticals in the screening of drugs it is interesting to design labeled H3 receptor ligands. To date several different techniques have been used to measure receptor occurrence in tissues. Radioactive competition is often used for instance in brain slices with [3H](R)-alpha-methylhistamine (Martinez-Mir et al., 1990), [3H]Nα-methylhistamine (Le et al., 2009), [3H]-A-349821 (Miller et al., 2009), [125I]iodoproxyfan (Ligneau et al., 1994; Stark et al., 1996), and [125I]iodophenpropit (Jansen et al., 2000). Adversely, analysis of ex vivo autoradioactivity often takes longer time (Le et al., 2009) and synthesis and storage of radioligands causes high costs and special equipment/rooms. Radioactive exposure as generally known is very harmful. Fluorescent ligands are preferred over radioligands in terms of safety precautions and often applicability. Fluorescent ligands and their use for the localization and detection of GPCRs is still a topical area of investigation (Kuder and Kiec-Kononowicz, 2008). Attempts to design fluorescent human hH3R were established with motif structure elements of Sangers reagent, dansyl, NBD, cyanoisoindol, and tetramethylrhodamine groups (Amon et al., 2006, 2007; Cowart et al., 2006; Kuder et al., 2010). Most of these compounds showed high affinity at histamine H3 receptors (hH3R Ki: 0.1–10 nM), but their fluorescence absorption and emission wavelengths were mainly between 300 and 500 nm. In this wavelength range interactions with cellular autofluorescence occur and in addition to this problem, most of these ligands possess low fluorescence intensities. A PhD thesis from the working group of Prof. Buschauer (University Regensburg, Erdmann, 2010) presented further fluorescent human histamine H3 ligands. Applied fluorophores possess good fluorescent features but are expensive or difficult to synthesize. The aim of this study was the optimization of the fluorescent properties taking advantage of already published fluorescent ligands as lead structures. A literature survey indicated bio-active fluorescent benzylidine tetralones (Kamakshi et al., 2010) possessing different antibacterial activity and useful physicochemical data (Al-Ansari, 1998). Tomecková et al. (2004) reported on related cyclic chalcone analogs demonstrating biological effects on mitochondrial outer membrane via fluorescence microscopy. These results motivated us to use chalcones as fluorescent element for labeling of histamine H3 receptor ligands to generate novel fluorescent pharmaceutical tools (Figure 1).

Figure 1

Design strategy for novel fluorescent human histamine H.

Materials and Methods

Chemistry

All reagents and solvents were purchased from VWR (Darmstadt, Germany), Sigma-Aldrich (Steinheim, Germany), Alfa Aesar (Ward Hill, MA, USA), Perkin Elmer Life and Analytical Sciences (Rodgau, Germany), and Acros Organics (Geel, Belgium), and were used without further purification (unless otherwise stated). 1H and 13C NMR spectra were recorded on a AV 250 Spektrometer (5.9 T; 1H: 250 MHz; 13C: 63 MHz), AV 300 Spektrometer (7.1 T; 1H: 300 MHz; 13C: 75 MHz), or AV 400 Spektrometer (0.4 T; 1H: 400 MHz, 13C: 100 MHz): Bruker (Rheinstetten, Germany). Electro-spray-ionization MS (ESI MS) was performed on a: VG Platform II: Fisons Instruments (Manchester, GB) and nano-ESI (nESI): Mariner Workstation TOF: Applied Biosystems (Carlsbad, CA, USA). High resolution MS (HRMS) was recorded on a LTQ Orbitrap XL: Thermo Fisher Scientific (Waltham, MA, USA). Elemental analyses (C, H, N) were measured on a Vario MicroCube: Elementar Heraeus (Hanau, Deutschland) and were within ±0.4% of the theoretical values for all final compounds. Preparative column chromatography was performed on silica gel 60 F254, coat thickness: 0.2 mm (VWR, Darmstadt, Germany). The microwave oven used was a Biotage Initiator 2.0 (400 W): Biotage (Uppsala, Sweden). For detailed synthesis procedures and analytical data see Section “Appendix.”

The initializing precursor 3-(piperidin-1-yl)propan-1-ol hydrochloride was synthesized by alkylation of piperidine with 3-chloropropan-1-ol as described in the literature (Apelt et al., 2005) and chlorinated (Sander et al., 2010). 6-Hydroxy-1-tetralone was commercially available whereas its regioisomer 7-hydroxy-1-tetralone was prepared from the corresponding methoxy derivative (Scheme 1). Ether cleavage has been carried out with para-toluenesulfonic acid and 1-butyl-3-methyl-1H-imidazolium bromide as ionic liquid under microwave condition with yields greater than 90% as described before (Boovanahalli et al., 2003). Synthesis of compound 1 and 2 started with the alkylation of 6- and 7-hydroxy-1-tetralone with 1-(3-chloropropyl)piperidine hydrochlorid (Scheme 1) by Williamson ether reaction, respectively (Williamson, 1851). Compound 1 and 2 were converted into fluorescent chalcones (compound 3, 4, 5, and 6) via aldol condensation with appropriate aldehyde.

Scheme 1

Synthesis of compounds 1–6. (a) para-Toluenesulfonic acid, 1-butyl-3-methyl-1H-imidazolium bromide, microwave irradiation, 160°C, 1 h, η = 97%; (b), 1-(3-chloropropyl)piperidine hydrochlorid, K2CO3, KI, acetone, 60°C, 48 h, η = 71–94%; (c) 4-(dimethylamino)benzaldehyde, ethanol/water, NaOH, RT, over night, η = 63–85%; (d) 4-(dimethylamino)cinnamic aldehyde, ethanol/water, NaOH, RT, over night, η = 69–89%.

Alkylation of 4-(hydroxymethyl)phenol with 1-(3-chloropropyl)piperidine hydrochlorid was the initializing step in the synthesis of compounds with elongated spacer B (Scheme 2). Resulting (4-(3-(piperidin-1-yl)propoxy)phenyl)methanol was chlorinated with thionylchloride. After alkylation with 6- and 7-hydroxy-1-tetralone fluorescent chalcones were synthesized with corresponding aldehydes (compounds 9–12), respectively. All tetralone and chalcone derivatives were purified by column chromatography. Oily products were crystallized as salts of oxalic acid in ethanol.

Scheme 2

Synthesis of compound 7–12. (a) K2CO3, KI, acetone, 60°C, 48 h, η = 65–71%; (b) 4-(dimethylamino)benzaldehyde, ethanol/water, NaOH, RT, over night, η = 85–91%; (c) 4-(dimethylamino)cinnamic aldehyde, ethanol/water, NaOH, RT, over night, η = 93–95%.

For fluorescence characterization all corresponding histamine H3 receptor ligands were dissolved in buffer (12.5 mM MgCl2, 100 mM NaCl, and 75 mM Tris/HCl, pH 7.4) at a concentration of 10 mM. Fluorescence absorption and emission spectra were recorded on a Fluorolog HORIBA JOBIN YVON fluorometer (HORIBA scientific, Kyoto, Japan) at room temperature. Spectra were analyzed with FluorEssence™ (HORIBA scientific, Kyoto, Japan) for Windows®.

Binding studies

Determination of human histamine H3 receptor affinity

HEK-293 cells stably expressing the recombinant human histamine H3 receptor were used for membrane extraction. In brief, membrane protein concentration was determined by the method of Bradford (1976). Competition binding experiments were carried out as described before (Kottke et al., 2011). Membranes were incubated with [3H]Nα-methylhistamine (2 nM) and test ligand in a concentration range between 0.01 nM and 100 μM. For determination of non-specific binding 10 μM of pitolisant was used. All values are means of at least three independent measurements, each in triplicates and on seven different concentrations. Binding data were analyzed by the software GraphPad Prism™ (2000, version 3.02, San Diego, CA, USA).

Determination of human histamine H1 receptor affinity

CHO-K1 cells expressing the human histamine H1 receptor were used for membrane extraction. In brief, human histamine H1 receptor radioligand competition binding assay was performed as described before (Rossbach et al., 2011). Membranes were incubated with 1 nM of [3H]pyrilamine and test ligand in a concentration range from 100 nM to 100 μM. For determination of non-specific binding 10 μM of chlorpheniramine was used. All values are means of at least two independent measurements, each in triplicates and four different concentrations. Binding data were analyzed by the software GraphPad Prism™ (2000, version 3.02, San Diego, CA, USA).

Determination of human histamine H4 receptor affinity

Sf9 cells transiently expressing the human histamine H4 receptor and co-expressed with G-protein Gαi/o and Gβ1γ2 subunits were used for membrane extraction. Competition binding experiments were carried out as described before (Kottke et al., 2011). Membranes were incubated with 10 nM [3H]histamine and test ligand in a concentration range from 0.01 nM to 100 μM. Non-specific binding was determined by using 10 μM JNJ-7777120. All values are means of at least two independent measurements, each in triplicates and four different concentrations. Binding data were analyzed by the software GraphPad PrismTM (2000, version 3.02, San Diego, CA, USA).

Cell culture

Slides were coated with poly-d-lysine (50 μg/mL, Sigma-Aldrich, Steinheim, Germany) and cooled for 2 h. HEK-293 cells stably expressing the recombinant human histamine receptor (mean protein content: 10 μg/mL; Bmax: 300–900 fmol/mg) were seeded (100,000 cells/mL) and grown in 2 mL of Dulbecco's modified Eagle's medium (without phenol red) with 2 mM glutamine, 10 mM HEPES, 10% fetal bovine serum, and 10 μL/mL penicillin/streptomycin in an atmosphere of 5% CO2 at 37°C for 48 h in six-well plates with poly-d-lysine coated slides. The medium was removed, and the slides were incubated with 3% BSA in buffer (12.5 mM MgCl2, 100 mM NaCl, and 75 mM Tris/HCl, pH 7.4) for 30 min at 37°C. Afterward the slides were washed with PBS buffer and incubated with 2 mL 200 nM of appropriate fluorescent H3 ligand in buffer (12.5 mM MgCl2, 100 mM NaCl, and 75 mM Tris/HCl, pH 7.4) for 1 h at 37°C. The ligand solution was removed, washed with PBS buffer, and the cells were fixed in methanol for 20 min at a temperature of −20°C. After equilibration with PBS buffer the nucleus was stained with 100 nM blue fluorescent DAPI (4′,6-diamidino-2-phenylindole; Kapuscinski, 1995). The slides were mounted on cover slips with 30 μL Mowiol. For reasons of comparison HEK-293 cells were cultured as blind reference and treated with respective ligand in a similar manner.

Fluorescence microscopy

Cell preparations were fluorescently visualized with a confocal laser scanning microscope (Leica TCS SP5, Wetzlar, Germany). Fluorescence intensities were adjusted minimizing autofluorescence and all slices were measured with the same intensity to generate comparable images. All images were recorded with a 60× oil immersions objective. Fluorescence colors were adapted to appropriate fluorescence emission wavelengths. Fluorescent ligands were excited with a 488 nm multiline argon laser and emission was detected between 520 and 790 nm. DAPI-stained nuclei of the HEK-293 cells were excited with a 405 nm diode laser and emission was measured between 420 and 500 nm. Images were recorded in sequential mode to avoid interferences.

Results and Discussion

Design of novel human histamine H3 ligands

Historically the first potent histamine H3 receptor ligands were derivatized from the endogenous ligand histamine. Ciproxifan as parent compound for these lead structures possesses a basic imidazole ring as core moiety (Figure 1). Spacer A is a propyl group, followed by a polar ether functionality linking the phenyl group which is then connected to structurally different affinity enhancing elements. The imidazole-containing ciproxifan exerts an affinity at the human histamine H3 receptor in the medium nanomolar concentration range (Table 1; Ligneau et al., 2000). Since the imidazole group may potentially be associated with interaction to the cytochrome P450 enzyme system, the change from imidazole to more robust piperidine systems in the western part of the molecule has been performed resulting in an improved pharmacological profile (Lazewska et al., 2006, 2008; Sander et al., 2008). In 2001 the piperidine analog of ciproxifan, UCL-2190 (Figure 1), was published (Meier et al., 2001). Likewise we introduced the piperidine core element in our newly designed compounds, kept spacer A constant with three methylene groups as well as the benzyl ether spacer. For compound 7–12 spacer B was elongated by the benzyl ether. The keto group in the lipophilic residue of ciproxifan was incorporated into a tetrahydronaphthalene system (1, 2, 7, and 8) which were converted to fluorescent chalcones (3, 4, 5, 6, 9, 10, 11, and 12).

Table 1

Human histamine H.

Compound	Structure	hH1R Ki [nM]a	hH3R Ki [nM]b	hH4R Ki [nM]c	λmax Abs. [nm]d	λmax Em. [nm]d	Stokes shift [nm]
Ciproxifan		>10,000e	46 ± 4f	612 ± 32g
1		3,473 ± 475	68 ± 12	100,535 ± 21,305
2		3,717 ± 873	89 ± 8	169,895 ± 118,942
3		791 ± 192	70 ± 21	12,370 ± 3,055	405	584	179
4		2,142 ± 886	149 ± 29	13,725 ± 1,110	397	583	186
5		881 ± 203	94 ± 26	5,634 ± 2,100	468	670	202
6		1,974 ± 89	101 ± 13	8,704 ± 204	467	674	207
7		1,968 ± 416	19 ± 5	37,260 ± 13,209
8		3,686 ± 950	17 ± 5	43,875 ± 9,581
9		1,768 ± 472	41 ± 4	12,598 ± 6,608	405	567	162
10		4,228 ± 1,121	97 ± 34	15,610 ± 5,841	405	583	178
11		1,374 ± 10	68 ± 3	6,400 ± 2,189	467	600	133
12		1,837 ± 837	87 ± 20	10,984 ± 3,403	467	607	140

.

.

During design of novel fluorescent pharmacological tools it is important to generate compounds with high affinity at the target receptor, high fluorescence intensities, and emission wavelengths near the infrared wavelength range. It is difficult to retain binding affinity and low molecular weight with structurally larger fluorophoric elements in small molecule GPCR ligands. Fluorophores emitting with high wavelengths frequently consist of bulky structures. Al-Ansari (1998) and Kamakshi et al. (2010) described the synthesis of unbulky, fluorescent benzylidene tetralones which have been used as fluorophore lead element (Figure 1).

Determination of human histamine H3 receptor affinity

Affinities of the newly designed fluorescent chalcones (3, 4, 5, 6, 9, 10, 11, and 12) at the human histamine H3 receptor are in a comparable nanomolar concentration range like that of ciproxifan (Table 1). Tetralones with elongated aromatic spacer B exhibit slightly lower Ki values than those of the shorter tetralones (1 → 7; 2 → 8) most probably due to increased lipophilic interactions. Elongation of the tetralones with appropriate aldehydes gives no further improvement in binding properties.

The variation from 6- to 7-hydroxy-1-tetralone derivatives causes no remarkable difference in binding and does not result in higher affinities at the human histamine H3 receptor.

Determination of human histamine receptor selectivity

For selectivity validation the new compounds have been screened on affinity at related human histamine H1 and H4 receptors. All ligands possess highest affinity at the human histamine H3 receptor demonstrating their receptor preference (Table 1). Affinity at the human H1 receptor is about one log unit and at the human histamine H4 receptor about two log units higher than that at the human histamine H3 receptor.

Compound 8 demonstrated one of the highest selectivity over all 12 compounds. Compound 9 is the most selective compound in the fluorescent chalcone series whereas compound 5 the less selective chalcone. These data further prove the beneficial effects of elongation of spacer B via benzyl ether concerning selectivity and affinity.

Fluorescence characterization

Kamakshi et al. (2010) described the fluorophore lead structure (Figure 1) as compound with intense emission signals. Absorption maximum was found at 406 nm. The fluorophore lead is described as charge transfer compound where the N,N-dimethyl group behaves as the electron donor moiety and the carbonyl group as the electron acceptor moiety. These two groups form a donor–acceptor complex upon excitation. Fluorescence measurement confirmed the absorption maxima (compounds 3, 9, and 10: 405 nm; cf. Appendix).

Introduction of an additional vinyl group resulted in an increased maximum in excitation and emission wavelength. (cf. 3 → 5: Δλmax Abs.: 63 nm, Δλmax Em.: 102 nm). Compounds with propenone element (3, 4, 9, 10) emit in the yellow area of the fluorescence spectrum and compounds with pentadienone elements (5, 6, 11, 12) in the red one. It is important to design fluorophores which emit fluorescence higher than 500 nm. In the blue spectroscopic area (300–450 nm) high autofluorescence occurs and causes difficulties in detection of the actual target. In fluorescence visualization it is the aim to achieve emission in the near infrared wavelength area to improve the signal-to-noise ratio. Therefore it is a need to use fluorophores with high quantum yields and low quenching properties. To avoid self-quenching it is important to synthesize fluorophores with large Stokes shifts. All compounds fulfill this requirement (Table 1).

Elongation of spacer B via benzyl ether group causes decrease of emission wavelengths of red emitting chalcones. Introduction of a phenylether group can influence the donor–acceptor complex of the fluorophore. Elongated compounds show higher fluorescence emission intensities (Figure 2; cf. Appendix).

Figure 2

Fluorescence emission spectra of selected chalcone derivatives.

For the measurement of fluorescence absorption and emission buffer (12.5 mM MgCl2, 100 mM NaCl, and 75 mM Tris/HCl, pH 7.4) has been used to simulate assay conditions and simultaneously find the ideal settings for confocal laser microscope studies.

Visualization of human histamine H3 receptors

To minimize unspecific binding cells were incubated for 30 min with 3% BSA solution. Fluorescent histamine H3 receptor ligands were added in a concentration range of two up to three times of Ki-value at the human histamine H3 receptor (200 nM) on hH3-HEK-293 and HEK-293 cells. After washing steps and staining of the nucleus with blue fluorescent DAPI as reported in literature (Kapuscinski, 1995), slides were fixed with Mowiol and analyzed with a confocal laser scanning microscope (Figure 3).

Figure 3

Confocal laser scanning microscope images of fluorescent human histamine H. All images were taken on a Leica TCS SP5 microscope. Fluorescent chalcones were incubated for 1 h and after washing steps nuclei were stained with blue fluorescent DAPI. (A) Shows a HEK-293 cell labeled with compound 3 and (B) shows the DAPI-stained nucleus of the same cell. (C) Is an overlay of image (A,B). (D) Shows compound 3 on an hH3-HEK-293 cell, (E) the DAPI-stained nucleus of the same cell and (F) an overlay of (D) and (E). Red fluorescent compound5 on an hH3-HEK-293 cell is shown in (G), cell nucleus in (H), and the overlay of(G) and(H) in (I). (J) Shows an hH3-HEK-293 cell stained with compound 9, (K) its nucleus, and (L) the overlay of (J,K).

Chalcones synthesized from 4-(dimethylamino)benzaldehyde result in yellow fluorescent ligands like compound3 and 9. Chalcones synthesized from 4-(dimethylamino)cinnamic aldehyde in red emitting ligands like compound5. Due to their maximum emission wavelengths ligand color on the confocal laser scanning microscope was adjusted. Fluorescence images were recorded in sequential mode to avoid interfering of emission wavelengths of ligand and DAPI fluorescence. In the first step DAPI-stained nuclei were recorded (Figures 3B,E,H,K) between 420 and 500 nm, then fluorescent ligand emission was measured between 520 and 790 nm (Figures 3A,D,G,J). Appropriate images were over-laid to co-localize fluorescent ligand and fluorescent nucleus (Figures 3C,F,I,L). Figure 3D shows the enrichment of the yellow emitting compound3 in the outer membrane of hH3-HEK-293 cells. Figure 3G shows the enrichment in same regions in hH3-HEK-293 cells of the red emitting compound5 which emits more intensive (Figure 2). The most human histamine H3 receptor selective and most intense compound (9) is shown in Figures 3J–L. Fluorescent human histamine H3 ligands stain the outer membrane of hH3-HEK-293 cells where human hH3R are mainly expressed. In HEK-293 cells which do not have human hH3R, no enrichment of ligand in the outer cell membrane can be identified (Figures 3A,C; cf. Appendix) indicating selectivity to the aimed receptor (Table 1). The fluorescent ligands on HEK-293 cells are diffusely distributed and emit in a low manner if compared to images 3D, 3G, and 3J where outer membranes emit strong fluorescence (cf. Appendix). Differences between hH3- and HEK-293 cells are facile to detect. Fluorescence emission of chalcone derivatives on HEK-293 cells is in the same range as autofluorescence (cf. Appendix) and therefore negligible.

Conclusion

In our work we achieved the design of selective and fluorescent human histamine H3 receptor ligands. Affinities at the human histamine H3 receptor of all newly synthesized compounds are in the nanomolar concentration range. All 12 compounds showed high H3 receptor preference in comparison to their binding properties at H1 and H4 receptors.

The fluorescent chalcone ligands facilitate visualization of human hH3R on hH3–HEK cells. Their fluorescence characteristics (fluorescence emission maxima higher than 500 nm) avoid appearance of interference with autofluorescence. Intense light signals label the localization of targeted receptor. Reference HEK-293 cells which do not over-express human hH3R show negligible binding of fluorescent ligands demonstrating low unspecific binding.

Compounds 3, 5, and 9 showed excellent, but slightly diverse properties concerning absorption, emission, Stokes shift, intensity and H3 receptor affinity and selectivity. Best compound concerning wavelength represents compound 5 (λmax Abs./Em.: 468/670 nm) but it is not as selective as compound 9 which emits with highest intensity of all synthesized compounds in the yellow area of spectrum. Compound 9 has the highest affinity at the human histamine H3 receptor compared to the other chalcone derivatives in this series.

Our new fluorescent ligands open new possibilities for non-radioactive alternatives for novel binding assays and novel visualization tools. They are precious tools for detailed pharmacological analyses on detection of receptors in cells and tissues. Studies on different tissues with these novel ligands are envisaged to provide further evidence of their convenience as pharmacological tools.

More detailed information is available in the Appendix.

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Table A1

Fluorescence images of fluorescent compounds.

Compound	Cell line	Ligand image	DAPI image	Overlay of ligand and DAPI image
3	hH3-HEK-293
3	HEK-293
4	hH3-HEK-293
4	HEK-293
5	hH3-HEK-293
5	HEK-293
6	hH3-HEK-293
6	HEK-293
9	hH3-HEK-293
9	HEK-293
10	hH3-HEK-293
10	HEK-293
11	hH3-HEK-293
11	HEK-293
12	hH3-HEK-293
12	HEK-293
–	H3-HEK-293
–	HEK-293