3
944
M. Nimczick et al. / Bioorg. Med. Chem. 22 (2014) 3938–3946
spacer at the imidazole, the bivalent compound 31a with this
spacer at the amide is now an antagonist.
or linking groups, are needed and may serve as the basis for contin-
uative investigation of the hCB R. According to these results, very
careful design of bivalent ligands for the hCB R is necessary, espe-
2
Altogether, the potency of all bivalent ligands decreases to the
2
micromolar range at the hCB
2
R compared to parent ligand 13a.
cially when agonist properties are planned to be maintained. The
results obtained in this study might be of general relevance and
interest for the design of bivalent GPCR agonists.
But also the standard ligands had shown lower affinities in the
functional steady-state GTPase assay32 and for the parent ligand
1
3a a 49-fold lower potency than the literature value26 was deter-
mined. Therefore, it is reasonable to expect higher affinities for the
uni- and bivalent ligands in radioligand binding studies. Neverthe-
4
4
4
. Experimental section
.1. Chemistry
less, at the hCB
range. Thus, they largely lose selectivity to the hCB
1
R they are potent in the micromolar or higher
R, but still
2
show affinity. Due to the coupling with a spacer, the efficacies
change and the compounds are no agonists any more. However,
the influences of bivalency, spacer length and connecting position
are observable. For bivalent ligands with the spacer at the imidaz-
ole, bivalency only changes the efficacy at both receptors and has
no effect on the potency compared to the corresponding univalent
compounds. For the short spacer at the amide, the efficacy at both
receptors is changed and the potency is increased due bivalency. In
contrast, for the long spacer at the amide, the bivalency has no
influence either on potency or on efficacy. The spacer length leads
to a change in the efficacy at both receptors. In addition, also the
connecting position has an influence on the efficacy. The bivalent
ligands with a short spacer at the imidazole act as antagonists
.1.1. General methods
Unless indicated otherwise, all chemicals and solvents were
purchased from commercial suppliers and used without further
purification. THF was dried by reflux over sodium overnight and
freshly distilled before use. For reactions with dry, freshly distilled
THF oven-dried glassware was used. Reactions over argon or nitro-
gen atmosphere were carried out by filling the reaction appara-
tuses by a gas flow of the corresponding, commercially available
gas and afterwards closing the filled reaction system with a gas-
filled balloon. Thin layer chromatography (TLC) was performed
on silica gel 60 on aluminum foils with fluorescent indicator
2
3
54 nm. For detection iodine vapor and UV light (254 nm and
66 nm) were used. Preparative TLC was performed with silica
(
2 1
at the hCB R) or moderate inverse agonists (at the hCB R), and
gel 60 GF254 for preparative thin layer chromatography (Merck
KGaA). For column chromatography silica gel 60 (particle size:
with a long spacer as inverse agonist or full inverse agonist,
respectively. For bivalent ligands connected at the amide, the short
0
.063–0.200 mm or 0.035–0.070 mm) was used. The purity of the
2
spacer results in inverse agonism (at the hCB R) or full inverse
target compounds was confirmed by HPLC (degassing unit:
DGU20A3R, gradient solvent delivery unit: LC-20AB with two LC-
agonism (at the hCB R), and the long spacer results in antagonism
1
or moderate inverse agonism, respectively. In the end, a connection
of the parent structure with a spacer at the imidazole or at
the amide always causes a decrease of potency and a change of
2
0AD pumps, column: Synergi 4u Fusion-RP 80A 150 ꢂ 4.6 mm
from Phenomenex Incorporation, UV/vis detector: SPD-20A, analy-
sis software: LabSolutions version 5, everything except the column
2 1
efficacy at the hCB R and hCB R. This suggests that the receptor–
from Shimadzu Corporation; method parameters: A: H
CF COOH, B: CH OH with 0.1% CF COOH, B 10% to 80% 10 min, B
0% 5 min, B 80% to 10% 3 min, flow rate: 1.0 mL min , UV detec-
tion: 254 nm). The compounds were characterized by a combina-
2
O with 0.1%
ligand-interactions at both receptors are highly sensitive for
changes of the parent structure at both assumed ‘preferred’ attach-
ment positions.
3
3
3
ꢀ1
8
1
13
tion of
H
NMR,
C NMR, mass spectrometry (MS), high
1
13
3
. Conclusion
resolution MS (HRMS) and elemental analysis. H NMR and
C
NMR spectra were recorded on a Bruker AVANCE 300 MHz spec-
trometer (Bruker BioSpin GmbH) with 300 MHz and 75.5 MHz,
respectively or a Bruker AVANCE 400 MHz spectrometer (Bruker
BioSpin GmbH) with 400 MHz and 100 MHz, respectively. NMRs
In summary, two series of homobivalent ligands with different
attachment points containing the parent ligand (hCB
agonist 13a) and their corresponding univalent ligands were
designed and synthesized to target the hCB R. Biological evaluation
2
R selective
2
3
were determined in CDCl and chemical shifts are expressed in
1
13
of these compounds in a functional steady-state GTPase assay
demonstrates that changes of the parent structure at the imidazole
or amide have a crucial effect on the potency and efficacy at the
ppm relative to the solvent peak ( H NMR: d 7.26 ppm; C NMR:
d 77.16 ppm). For MS and HRMS different ionization techniques
like ESI, EI or CI were used. MS were performed for ESI with TSQ
7000 (ThermoQuest Finnigan) or Agilent 1100 series LC/MSD ion
trap mass spectrometer (Agilent Technologies), for EI and CI with
MAT SSQ 710 A (Finnigan). HRMS were measured with Q-TOF
6540 UHD (Agilent Technologies). Elemental analysis for carbon,
hydrogen and nitrogen were undertaken using a Vario micro cube
(Elementar Analysensysteme GmbH).
hCB
2
R and hCB
1
R. They still showed a micromolar activity at both
R and no agonistic behav-
receptors, but no selectivity to the hCB
2
ior. Thus, the receptor–ligand-interaction is very sensitive for
changes of the parent structure at these positions. However, it
was shown that bivalency had a pronounced influence on the effect
of the ligands at the receptors. For bivalent ligands coupled at the
imidazole and the bivalent ligand with a short spacer at the amide,
bivalency changed the efficacy compared to the univalent com-
pounds. Additionally, for the short bivalent compound 31b the
bivalency was able to increase the potency compared to 26b. In
contrast, for the long spacer at the amide, bivalency did not change
potency or efficacy. Furthermore, it was shown that the spacer
length and the attachment position also altered the efficacy of
these bivalent ligands at the receptors. In addition, the methyl
ester 13c indicates that the ester function at position 5 of the benz-
4.1.2. General procedure A for amide formation to prepare 8,
22a–b, 27a–b
To a solution of 4-fluoro-3-nitrobenzoic acid (7) and a catalytic
amount of DMF in CH
dichloride in CH Cl at 0 °C. The mixture was stirred at room tem-
perature for 2 h. Then, the solvent was removed in vacuo, some
CH Cl was added and the solvent was removed again. The brown,
oily residue was immediately dissolved in CH Cl . This solution
was slowly added to a solution of the corresponding amine and
triethylamine in CH Cl at 0 °C. The mixture was stirred overnight
at room temperature. After removal of solvent in vacuo the crude
product was dispersed in concentrated Na CO solution, and
2 2
Cl was slowly added a solution of oxalyl
2
2
2
2
2
2
imidazole is tolerated with a higher selectivity to the hCB
thus bivalent ligands with an ester coupling might be possible.
Further research of bivalent ligands targeting the hCB R, maybe
applying other parent structures, modifying the spacer structure
2
R, and
2
2
2
2
3