A novel non-phenolic dibenzazecine derivative with nanomolar affinities for dopamine receptors

Article abstract of DOI:10.1002/cbdv.201000317

Dibenzazecines are a novel class of dopamine receptor antagonists, characterized by their high affinities as well as their tendency for D ₁ selectivity. Hitherto, the most active dibenzazecines were phenolic in nature; a 3-OH substituent was found to result in the highest affinities. However, the phenolic nature of these compounds mostly renders them unsuitable for in vivo application, due to the poor pharmacokinetic profile, imparted by the phenolic group. A novel dibenzazecine derivative was prepared, with methylenedioxy moiety, connecting C(2) amd C(3), instead of the 3-OH group. The newly synthesized derivative 3 showed high affinities similar to the lead LE404, displaying nanomolar affinities for all dopamine receptor subtypes. Its dibrominated derivative 4, though exhibiting almost a fivefold decrease in affinities, still displayed nanomolar ones for all dopamine receptors, except for D₄. In a functional Ca²⁺ assay, both compounds 3 and 4 were found to possess antagonistic properties towards the dopamine receptors.

Full text of DOI:10.1002/cbdv.201000317

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
431
A Novel Non-phenolic Dibenzazecine Derivative with Nanomolar Affinities
for Dopamine Receptors
by Dina Robaaa), Shams ElDin AbulAzmb), Jochen Lehmanna)c), and Christoph Enzensperger*a)
a) Institut fꢀr Pharmazie, Lehrstuhl fꢀr Pharmazeutische/Medizinische Chemie, Friedrich Schiller
Universitꢁt Jena, Philosophenweg 14, D-07743 Jena (e-mail: ch.enzensperger@uni-jena.de)
b) Department of Medicinal Chemistry, Faculty of Pharmacy, University of Alexandria, Alexandria,
Egypt
c) College of Pharmacy, King Saud University, Riyadh, Saudi Arabia
Dibenzazecines are a novel class of dopamine receptor antagonists, characterized by their high
affinities as well as their tendency for D1 selectivity. Hitherto, the most active dibenzazecines were
phenolic in nature; a 3-OH substituent was found to result in the highest affinities. However, the phenolic
nature of these compounds mostly renders them unsuitable for in vivo application, due to the poor
pharmacokinetic profile, imparted by the phenolic group. A novel dibenzazecine derivative was
prepared, with methylenedioxy moiety, connecting C(2) amd C(3), instead of the 3-OH group. The
newly synthesized derivative 3 showed high affinities similar to the lead LE404, displaying nanomolar
affinities for all dopamine receptor subtypes. Its dibrominated derivative 4, though exhibiting almost a
fivefold decrease in affinities, still displayed nanomolar ones for all dopamine receptors, except for D4. In
a functional Ca2þ assay, both compounds 3 and 4 were found to possess antagonistic properties towards
the dopamine receptors.
Introduction. – Dopamine is a regulator of numerous physiological functions in the
central nervous system (CNS) and the periphery. Dysfunctions of the dopaminergic
system have been implicated in the pathogenesis of several neurological disorders,
including schizophrenia, Parkinsonꢂs disease, depression, attention-deficit hyperactiv-
ity disorder (ADHD), as well as drug and alcohol addiction [1][2].
Dibenz[d,g]azecines are a novel class of dopamine receptor antagonists, which are
distinguished by their new chemical structure, as well as their high affinities for
dopamine receptors, predominantly for the D1 family [3][4]. SAR Studies on this new
class of compounds revealed that derivatives that encompass a dibenz[d,g]azecine
moiety as the backbone structure and bear a OH substituent at C(3) (e.g., compound 1;
LE404; Fig.), display the highest affinities, specifically for human D1 and D5 receptors
[3–5]. Changing the position of the OH group or replacing it by a MeO group both led
to decreased affinities [4]. Surprisingly, a 2,3-dihydroxy substitution pattern (com-
pound 2; LE403; Fig.), which exhibits the highest structural resemblance to dopamine,
the natural ligand, showed ca. 100–1000-fold lower affinities than the monohydroxy-
lated LE404 1 [3].
However, the high affinities of these dopamine antagonists are coupled with
drawbacks. Phenols and especially dihydroxylated compounds tend to be chemically
unstable. Due to their phenolic nature, these compounds are predisposed to poor
ꢃ 2011 Verlag Helvetica Chimica Acta AG, Zꢀrich

432
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
Figure. The 3-hydroxybenzazecine derivative LE404 (1), the less active 2,3-dihydroxy analog LE403 (2),
and the target methylenedioxy analogs 3 and 4
pharmacokinetics (low bioavailability), which would restrict their utility for in vivo
experiments and render them unsuitable for clinical development. Accordingly, our
preliminary in vivo evaluations have shown that compound 1 (LE404) has a low oral
bioavailability when administered to rats [6]. Studies on other classes of phenolic
dopamine antagonists, in particular SCH23390 and SCH39166, have led to similar
results [7][8]. Both compounds showed very low bioavailability besides a short
duration of action. The latter fact has been attributed to the fast biotransformation of
these phenolic compounds, occurring mainly by glucuronidation at the phenolic
function [7][8]. Analogs lacking this phenolic group have indeed been found to exhibit
better pharmacokinetics and longer duration of action, which is more advantageous for
in vivo application [9].
Aim of the present study was to prepare a dibenz[d,g]azecine in which C(2) and
C(3) are connected through a methylenedioxy, ꢀOCH2Oꢀ, moiety. This was appealing
for the following reasons.
ꢁ Our previous studies have shown that the nature, number, and position of the
substituents at the benzene ring significantly affect the binding affinities. It was hence
interesting to investigate the influence of this spatially rather confined group on the
affinities. A ꢀOCH2Oꢀ moiety shares similar electronic properties with the 2,3-
dihydroxy and 2,3-dimethoxy substituents, but is spatially distinct from all previously
examined substitution patterns.
ꢁ Compared with the previously synthesized phenolic dibenzazecines, the target
compound 3 is presumably chemically more stable and conceivably of better
pharmacokinetics.
Results. – Chemistry. As outlined in the Scheme, the target dibenzazecine 3 was
obtained via its pentacyclic precursor quinolizine 7. Our previously established
procedure, according to which the amine was reacted with 2-(2-bromoethyl)benzalde-
hyde in the presence of CF3COOH (TFA) [10], only yielded the intermediary
isoquinolinium salt 6 and not the expected quinolizine 7. Attempts to cyclize the
obtained isoquinolinium salt by PictetꢀSpengler cyclization with 6n HCl were
unsuccessful. The quinolizine 7 was thus prepared through the reaction of the
precursor amine 5 with 2-(chloroethyl)benzoyl chloride, and the obtained intermediate
was directly cyclized under BischlerꢀNapieralski conditions and subsequently reduced.

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
433
Scheme
a) TFA, dioxane, reflux, 12 h. b) 6n HCl, reflux, 10 h. c) Et3N, CH2Cl2, r.t., 90 min. d) POCl3/MeCN 1:2,
reflux, 36 h; NaBH4, MeOH, reflux, 2 h. e) MeI, acetone. f) Na, liq. NH3, ꢀ408, 10 min. g) ClCOOEt,
dry THF, ꢀ658, 4 h, then NaCNBH3, ꢀ658 to r.t.; h) LiAlH4, dry THF, reflux, 2 h. i) Br2, AlCl3, CH2Cl2,
r.t., 10 h.

434
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
Conversion of the quinolizine 7 to the target dibenzazecine 3 was similarly knotty.
Conventionally at that step, the quinolizine derivative 7 was quaternized with MeI and
the obtained quaternary salt was cleaved under Birch conditions to the corresponding
azecine. In this case, reduction of the quaternary salt 8 with Na in liquid NH3 gave a 1:1
mixture of two different compounds having the same molecular weight. These
substances were identified by means of GC/MS as the phenolic azecines 1 (LE404) and
9, which was confirmed by spiking the GC/MS with reference samples of 1 and 9, both
available in our laboratory from previous work. The phenomenon of the cleavage of the
ꢀOCH2Oꢀ moiety under various reductive conditions has been already described
[11][12]. For instance, reduction under Birch conditions (Na and liquid NH3) was
found to result in the cleavage of methylenedioxy derivatives, yielding the respective p-
phenolic compounds [11]. It has been suggested that the nature of the substituent
positioned opposite to theꢀOCH2Oꢀmoiety directs the site of cleavage, which leads to
the formation of either the m- or p-phenolic compound. In our case, however, cleavage
of the ꢀOCH2Oꢀ group occurred at two different positions and not at only one
position, as observed with reported methylenedioxy derivatives [11]. This, of course,
may be attributed to the fact that both substituents across the ꢀOCH2Oꢀ group exert
equal effects.
To accomplish ring opening and avoid degradation of the methylenedioxy function,
the quinolizine 7 was first converted to the carbamate 10 according to [13], which was
subsequently reduced with LiAlH4 to give the target compound 3. This azecine was
further dibrominated to examine the impact of the additional Br-atoms on the affinities.
Unfortunately, the monobrominated derivatives could not be obtained, even when
equimolar amounts of Br2 were used.
Pharmacology. The target compounds 3 and 4, as well as the precursor quinolizine 7
were screened for their affinities for the human cloned dopamine receptor subtypes D1,
D2L, D3, D4.4, and D5. These receptors were stably expressed in HEK293 or CHO cells;
[3H]SCH23390 and [3H]spiperone were used as radioligands at the D1- and D2-like
receptors, respectively. Ki Values are given in nanomolar units (Table). In addition, the
compounds were tested in an intracellular Ca2þ assay, to determine their functionality
at the D1 and D2 receptors. For this purpose, HEK293 cells, stably expressing the
respective D-receptor, were loaded with a fluorescent dye, and, after preincubation
with rising concentrations of the test compound, an agonist (SKF 38393 for D1 and
quinpirole for D2) was injected, and the Ca2þ-induced fluorescence was measured with
a microplate reader. The ability of the test compound to suppress the agonist-induced
Ca2þ influx is an indication of antagonistic or inverse agonistic properties at the
receptor. Both the radioligand-binding assay and the Ca2þ assay have been described in
detail in [10].
Discussion. – The newly synthesized dibenzazecine derivative 3 proved to be a
highly active dopamine antagonist (functional Ca2þ assay), showing comparatively
high affinities at the different dopamine receptors as the lead compound 1 (LE404), all
lying in the nanomolar range. The D1-selectivity profile, encountered with most
members of this class of dopamine antagonists, was, however, lost. Compound 3
displayed equal affinities for the D1 and D2 subtypes, and showed the highest selectivity
for the D5 receptor (D1/D5 ꢂ7). With respect to the affinities, compound 3 is superior to

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
435

436
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
437
previously synthesized 1- and 2-monohydroxy derivatives, 14 and 12, respectively, and
the 1-, 2- and 3-monomethoxy derivatives, 15, 13, and 11, respectively, and similarly
active to the 3-hydroxy compound 1. Compared with the 2,3-dihydroxy and the 2,3-
dimethoxy derivative, 2 and 16, respectively, 3 showed a tremendous improvement in
affinities; almost 100-fold increase in the affinity for D1 and 1000-fold for D2 and D5 was
observed. The dibromo derivative 4 displayed the highest affinity for the D5 receptor;
for the D1, D2, and D3, it showed similar but somewhat lower affinities, while the D4
affinities were strongly decreased.
Previous SAR studies have already indicated that a OH group at C(3) of the
dibenzazecine ring is crucial for high affinities. It presumably interacts through H-bond
formation with a serine residue (Ser 5.46 or Ser 5.50) in the receptor-binding pocket, as
already observed with other dopamine antagonists at the D2 receptor [15]. Derivatives
with a MeO substituent especially at C(2) or C(3) showed significantly decreased
affinities. This could be attributed to the fact that a MeO group, which rotates freely, is
spatially too bulky for the receptor binding site. A 2,3-dimethoxy substitution pattern
requires much more space due the free internal rotation of both MeO groups, hence,
resulting in a pronounced decrease in affinities. Conversely, aꢀOCH2Oꢀmoiety, which
requires less space due to its restricted rotation, showed similarly high affinities as the
highly active 3-OH derivative 1. From our observations, it can be concluded that the
phenolic compounds rather act as H-acceptor than as H-donor.
Taken together, a ꢀOCH2Oꢀ moiety connecting C(2) and C(3) of the dibenzaze-
cine ring is not only favored in terms of chemical stability and pharmacokinetics but
also in regards to the binding affinities to dopamine receptors.
Experimental Part
General. TLC: silica gel F254 plates (Merck). M.p.: in open cap. tubes, with a Gallenkamp melting
point apparatus; uncorrected. 1H- and 13C-NMR spectra: Bruker Advance 250 spectrometer (250 MHz)
and Advance 400 spectrometer (400 MHz); d in ppm, J in Hz. MS: by GC/MS, using a Hewlett Packard
GCD-Plus (G1800C) apparatus (HP-5MS column; J&W Scientific); m/z (rel. %). Purities of the
compounds were determined by elemental analysis, performed on a Hereaus Vario EL apparatus. All
values for C, H, and N were found to be within ꢃ0.4. All compounds showed >95% purity.
2-[2-(1,3-Benzodioxol-5-yl)ethyl]-3,4-dihydroisoquinolinium Trifluoroacetate (6). A soln. of 2-(1,3-
benzodioxol-5-yl)ethylamine (5; 1.65 g, 10 mmol), 2-(2-bromoethyl)benzaldehyde [16] (3.2 g, 15 mmol),
and CF3COOH (TFA; 1.7 g, 15 mmol) in dioxane (40 ml) was heated under reflux for 12 h. The
isoquinolinium salt, and not the expected quinolizinium salt, separated as a yellow solid. This was filtered
off, washed with dioxane, then Et2O, and dried to give 6 (1.2 g, 36%). M.p. 214–2168. 1H-NMR
(250 MHz, CD3OD): 3.19 (t, J¼7, CH2(4)); 3.28 (t, J¼8, CH2CH2N); 4.13 (t, J¼8, CH2CH2N); 4.27 (t,
J¼7, CH2(3)); 5.93 (s, OCH2O); 6.74ꢀ6.81 (mc, 2 arom. H); 6.89 (s, 1 arom. H); 7.48–7.56 (m, 2 arom.
H); 7.73–7.83 (m, 2 arom. H); 8.91 (s, Nþ¼CH).
5,8,9,14b-Tetrahydro-6H-[1,3]dioxolo[4,5-g]isoquinolino[1,2-a]isoquinoline (7). To an ice-cooled
soln. of 5 (2.97 g, 18 mmol) in CH2Cl2 (30 ml) was added Et3N (5.45 g, 54 mmol), followed by a soln. of 2-
(chloroethyl)benzoyl chloride (4.02 g, 19.8 mmol) in CH2Cl2 (10 ml). The mixture was stirred at r.t. for
90 min, subsequently washed with 10% HCl (2ꢄ30 ml), followed by brine (30 ml). The org. layer was
dried (Na2SO4) and evaporated under reduced pressure to yield a brown oil. This crude amide was
dissolved in a mixture of POCl3/MeCN (15 ml/30 ml), and the mixture was heated under reflux for 36 h.
The solvents were then removed under reduced pressure, and the remaining POCl3 was removed by
washing with petroleum ether (40–608), followed by Et2O. The residue was dissolved in MeOH (150 ml)
without prior purification, then NaBH4 (4 g) was carefully added to the ice-cooled soln. The mixture was

438
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
heated under reflux for 2 h, the solvent was subsequently removed under reduced pressure, and the
residue was treated with H2O (100 ml), and the mixture was extracted with Et2O (2ꢄ100 ml). The
collected org. layers were dried (Na2SO4) and evaporated under reduced pressure. The obtained oil was
purified by column chromatography (CC; AcOEt/MeOH 9 :1) to yield 7 (0.52 g, 10%). Yellowish white
oil, which solidified on standing. M.p. 122–1238. 1H-NMR (250 MHz, CDCl3): 2.74–2.94, 3.04–3.26 (2m,
CH2(5,6,8,9)); 5.00 (s, HꢀC(14b)); 5.89, 5.91 (2s, CH2(12)); 6.61 (s, HꢀC(10)); 6.69 (s, HꢀC(14)); 7.11–
7.26 (m, HꢀC(1–4)). GC/MS: 278 (100), 250 (25), 191 (8), 178 (14), 165 (16), 148 (11), 130 (11), 115
(24), 103 (19), 89 (23), 77 (39), 63 (30). Anal. calc. for C18H17NO2 ·0.1 AcOEt: C 76.79, H 6.23, N 4.86;
found: C 76.73, H 6.19, N 4.94.
Ethyl 5,8,9,15-Tetrahydrobenzo[d][1,3]benzodioxolo[5,6-g]azecine-7(6H)-carboxylate (10). A stir-
red soln. of 7 (0.78 g, 2.7 mmol) in dry THF (50 ml) was cooled in MeOH/dry ice at ꢀ658. While keeping
the mixture under N2, ClCOOEt (1.63 g, 15 mmol) was added, and stirring was continued for 4 h. Then, a
soln. of NaCNBH3 (0.59 g; 9.4 mmol) in dry THF (10 ml) was slowly added at ꢀ658, and the mixture was
stirred overnight, while allowing it to reach r.t. The mixture was subsequently treated with 2n NaOH
(120 ml), the THF layer was separated, washed with brine, and finally the org. layer was evaporated
under reduced pressure to give a yellow waxy solid. The obtained product was purified by CC (AcOEt/
hexane, 1:2) to yield 10 (0.65 g, 69%). White solid. M.p. 1108. 1H-NMR (400 MHz, CDCl3): 0.77, 0.97 (2t,
J¼11, MeCH2); 2.35, 2.48 (2t, J¼9, CH2(5)); 2.86–2.94 (m, CH2(9)); 3.40–3.57 (m, CH2(6,8)); 3.74, 3.86
(2q, J¼11, MeCH2); 3.98 (s, CH2(15)); 5.89 (s, CH2(12)); 6.57, 6.61, 6.64, 6.74 (4s, HꢀC(10,14)); 7.03–7.26
(m, HꢀC(1–4)). 13C-NMR (100 MHz, CDCl3): 14.35, 14.33 (MeCH2); 32.63, 33.09 (C(5)); 33.82, 34.17
(C(9)); 37.94, 38.44 (C(15)); 51.74, 51.86 (C(6)); 52.85, 53.69 (C(8)); 60.80, 60.90 (MeCH2); 100.78
(C(12)); 126.23, 126.47, 126.68, 126.75 (C(2,3)); 129.97, 130.92 (C(1)); 131.06, 132.08 (C(4)); 132.39,
132.77 (C(9a)); 133.46, 133.67 (C(14a)); 139.53, 139.60 (C(4a)); 140.00, 140.34 (C(15a)); 146.07, 146.26,
146.56 (C(10a,13a)); 156.38 (C¼O). GC/MS: 353 (28), 280 (9), 248 (31), 237 (16), 220 (72), 207 (12), 191
(17), 178 (70), 165 (100), 152 (47), 139 (19), 128 (30), 115 (64), 104 (70), 91 (41), 77 (55), 65 (41). Anal.
calc. for C21H23NO4: C 71.37, H 6.56, N 3.96; found: C 71.17, H 6.58, N 3.87.
5,6,7,8,9,15-Hexahydro-7-methylbenzo[d][1,3]benzodioxolo[5,6-g]azecine (3). A soln. of 10 (0.8 g,
2.3 mmol) in dry THF (20 ml) was slowly added to an ice-cooled, stirred suspension of LiAlH4 (0.25 mg,
9 mmol) in dry THF (75 ml), while keeping the reaction under N2. After the addition was completed, the
mixture was heated under reflux for 2 h. It was then cooled in an ice-bath, and the reaction was quenched
by careful addition of sat. potassium sodium tartrate soln., until no further H2 evolved. The resulting
suspension was then filtered off, and the filtrate was evaporated under reduced pressure to give 3 (0.65 g,
97%). White waxy solid. The obtained product was purified by CC (AcOEt/MeOH 9 :1). M.p. 70–718.
1H-NMR (250 MHz, CDCl3): 2.26 (s, NMe); 2.64–2.71 (mc, CH2(5,6,8,9)); 4.35 (s, CH2(15)); 5.88 (s,
CH2(12)); 6.56 (s, HꢀC(10)); 6.75 (s, HꢀC(14)); 7.04–7.33 (m, HꢀC(1–4)). 13C-NMR (62.5 MHz,
CDCl3): 34.46, 34.48 (C(5,9)); 38.06 (C(15)); 46.92 (NMe); 60.70, 60.74 (C(6,8)); 100.66 (C(12)); 109.99
(C(10,14)); 126.10, 126.29 (C(2,3)); 130.52, 130.89 (C(1,4)); 133.63, 133.88 (C(9a,14a)); 141.02, 141.09
(C(4a,15a)); 145.78, 145.84 (C(10a,13a)). GC/MS: 295 (12), 237 (30), 223 (38), 207 (10), 190 (32), 178
(60), 165 (100), 152 (39), 139 (16), 128 (21), 115 (55), 103 (33), 89 (43), 71 (60), 63 (41). Anal. calc. for
C19H21NO2 ·0.1 AcOEt: C 76.60, H 7.22, N 4.60; found: C 76.78, H 7.28, N 4.57.
10,14-Dibromo-5,6,7,8,9,15-hexahydro-7-methylbenzo[d][1,3]benzodioxolo[5,6-g]azecine (4). To a
suspension of 3 (0.15 g, 0.5 mmol) and anh. AlCl3 (0.03 g, 0.23 mmol) in CH2Cl2 (4 ml) was slowly added
a soln. of Br2 (0.4 g, 2.5 mmol) in CH2Cl2 (2 ml). After 10 h of stirring at r.t., the mixture was poured into
a sat. soln. of Na2S2O5 (50 ml) and extracted with CH2Cl2 (2ꢄ15 ml). The extract was washed with sat.
NaHCO3 soln. (2ꢄ50 ml) and H2O (2ꢄ50 ml). The org. layer was dried (Na2SO4) and evaporated under
reduced pressure, leaving a brownish solid, which was crystallized from MeOH to give 4 (0.15 g, 67.6%).
Yellow crystals. M.p. 135–1368. 1H-NMR (250 MHz, CDCl3): 2.35 (s, NMe); 2.65–2.78 (m,
CH2(5,6,8,9)); 4.86 (s, CH2(15)); 6.09 (s, CH2(12)); 7.03–7.26 (m, HꢀC(1–4)). 13C-NMR (62.5 MHz,
CDCl3): 32.71 (C(5)); 34.88 (C(9)); 37.95 (C(15)); 46.54 (NMe); 58.89 (C(6)); 60.79 (C(8)); 101.26
(C(12)); 104.23, 104.43 (C(10,14)); 126.24 (C(2)); 126.43 (C(3)); 130.15 (C(4)); 130.79 (C(1)); 135.17
(C(9a)); 135.59 (C(14a)); 139.31 (C(4a)); 141.35 (C(15a)); 144.20, 144.27 (C(10a,13a)). Anal. calc. for
C19H19Br2NO2: C 50.36, H 4.23, Br 35.26, N 3.09; found: C 50.20, H 4.29, Br 35.08, N 2.96.

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
439
We thank Bꢀrbel Schmalwasser, Petra Wiecha, and Heidi Traber for skillful technical assistance in
performing the pharmacological assays. Further, we would like to thank the DFG for the financial
support for the project (EN 875/1-1).
REFERENCES
[1] D. Vallone, R. Picetti, E. Borrelli, Neurosci. Biobehav. Rev. 2000, 24, 125.
[2] G. Emilien, J. M. Maloteaux, M. Geurts, K. Hoogenberg, S. Cragg, Pharmacol. Ther. 1999, 84, 133.
[3] B. Hoefgen, M. Decker, P. Mohr, A. M. Schramm, S. A. Rostom, H. El Subbagh, P. M. Schweikert,
D. R. Rudolf, M. U. Kassack, J. Lehmann, J. Med. Chem. 2006, 49, 760.
[4] P. Mohr, M. Decker, C. Enzensperger, J. Lehmann, J. Med. Chem. 2006, 49, 2110.
[5] M. Schulze, F. K. Mꢀller, J. M. Mason, H. Gorls, J. Lehmann, C. Enzensperger, Bioorg. Med. Chem.
2009, 17, 6898.
[6] M. Schulze, O. Siol, D. Robaa, F. K. Mꢀller, C. Fleck, J. Lehmann, J. Pharmacol. Exp. Ther.,
submitted.
[7] C. E. Tedford, V. B. Ruperto, A. Barnett, Drug Metab. Dispos. 1991, 19, 1152.
[8] C. E. Tedford, V. B. Ruperto, V. L. Coffin, M. Cohen, M. Libonati, A. Barnett, Drug Dev. Res. 1992,
26, 389.
[9] W. L. Wu, D. A. Burnett, R. Spring, W. J. Greenlee, M. Smith, L. Favreau, A. Fawzi, H. Zhang, J. E.
Lachowicz, J. Med. Chem. 2005, 48, 680.
[10] D. Robaa, C. Enzensperger, S. D. Abul Azm, E. S. El Khawass, O. El Sayed, J. Lehmann, J. Med.
Chem. 2010, 53, 2646.
[11] A. J. Birch, J. Chem. Soc. 1947, 102.
[12] E. Schwenk, D. Papa, J. Org. Chem. 1945, 10, 232.
[13] M. J. Calverley, J. Chem. Soc., Chem. Commun. 1981, 1209.
[14] C. Enzensperger, F. K. Mꢀller, B. Schmalwasser, P. Wiecha, H. Traber, J. Lehmann, J. Med. Chem.
2007, 50, 4528.
[15] M. Y. Kalani, N. Vaidehi, S. E. Hall, R. J. Trabanino, P. L. Freddolino, M. A. Kalani, W. B. Floriano,
V. W. Kam, W. A. Goddard, Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 3815.
[16] P. C. B. Page, B. R. Buckley, L. F. Appleby, P. A. Alsters, Synthesis 2005, 3405.
Received November 11, 2010