Novel 2,3,4,5-tetrahydro-1H-3-benzazepines with high affinity and selectivity for the dopamine D3 receptor

Article abstract of DOI:10.1016/S0960-894X(00)00505-9

Starting from the dopamine D₃ receptor antagonist SB-277011 1, a series of 2,3,4,5-tetrahydro-1H-3-benzazepines has been identified with high affinity for the dopamine D₃ receptor and selectivity over the D₂ receptor. The 3-acetamido-2-fluorocinnamide derivative 20 gave high D₃ receptor affinity (pK(i) 8.4) with 130-fold selectivity over the D₂ receptor. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00505-9

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2553±2555
Novel 2,3,4,5-Tetrahydro-1H-3-benzazepines with High Anity
and Selectivity for the Dopamine D3 Receptor
Nigel E. Austin, Kim Y. Avenell, Izzy Boy®eld, Clive L. Branch, Michael S. Hadley,
Phillip Je€rey, Christopher N. Johnson,* Gregor J. Macdonald, David J. Nash,
Graham J. Riley, Alexander B. Smith, Geo€rey Stemp, Kevin M. Thewlis,
Antonio K. K. Vong and Martyn Wood
SmithKline Beecham Pharmaceuticals, New Frontiers Science Park, Third Avenue, Harlow, Essex, CM19 5AW, UK
Received 19 July 2000; revised 21 August 2000; accepted 6 September 2000
AbstractÐStarting from the dopamine D3 receptor antagonist SB-277011 1, a series of 2,3,4,5-tetrahydro-1H-3-benzazepines has
been identi®ed with high anity for the dopamine D3 receptor and selectivity over the D2 receptor. The 3-acetamido-2-¯uoro-
cinnamide derivative 20 gave high D3 receptor anity (pKi 8.4) with 130-fold selectivity over the D2 receptor. # 2000 Elsevier
Science Ltd. All rights reserved.
Dopaminergic neurotransmission is mediated by ®ve
receptor subtypes (D1±D5) which can be grouped into
two receptor families. D1-like receptors include the D1
and D5 subtypes, whereas D2-like receptors include the
D2, D3 and D4 subtypes. Clinically e€ective anti-
psychotic agents share the property of dopamine D2 and
D3 receptor antagonism. At the doses used in the clinic
these drugs occupy D3 as well as D2 receptors and their
antipsychotic e€ects could therefore be mediated via D2
and/or D3 receptors. Blockade of dopamine D2 recep-
tors in the striatum leads to serious extrapyramidal side-
e€ects, which result in poor patient compliance and
consequently poor control of the disease. However,
dopamine D3 receptors are preferentially located in
limbic brain regions, such as the nucleus accumbens,
where dopamine receptor blockade has been associated
with antipsychotic activity. A selective dopamine D3
receptor antagonist therefore o€ers the potential for an
e€ective antipsychotic therapy, free of the serious side-
e€ects of currently available drugs.1 4 The presence of
the dopamine D3 receptor in projection regions of the
mesocorticolimbic system also suggests a potential
therapeutic role in reinforcement processes and drug
abuse.5 Recently, we described the novel tetra-
hydroisoquinoline SB-277011 1, a selective dopamine D3
receptor antagonist with high oral bioavailability and CNS
penetration in the rat.6 As part of the investigation of
structure±activity relationships around SB-277011, it was
important to investigate whether modi®cation of the
tetrahydroisoquinoline ring would lead to retention of
dopamine D3 receptor anity and selectivity.
Inspection of molecular models suggested that a 7-sub-
stituted 2,3,4,5-tetrahydro-1H-3-benzazepine would give a
good overlap with the 6-substituted tetrahydroisoquino-
line moiety of 1 (Fig. 1). However, the modelling also
indicated the di€erent spatial requirements of the ben-
zazepine ring which might adversely a€ect D3 anity.
This rationale led to the preparation of the 7-cyanotetra-
hydrobenzazepines 7±20 as shown in Scheme 1.
The known 7-methoxy-2,3,4,5-tetrahydro-1H-3-benz-
azepine 27 was converted to the corresponding 7-cyano
derivative 4 in ®ve steps, the key stage involving a pal-
ladium catalysed cyanide displacement of the Boc-pro-
tected tri¯uoromethylsulfonate 3. Reductive alkylation
of 4 with aldehyde 56 in the presence of NaBH(OAc)3
introduced the cyclohexylethyl linker, and subsequent
treatment with tri¯uoroacetic acid gave the primary
amine 6. Coupling of 6 to the appropriate acid in the
*Corresponding author. Tel.: +44-1279-627735; fax: +44-1279-627
896; e-mail: christopher_n_johnson@sbphrd.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00505-9

2554
N. E. Austin et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2553±2555
benzazepine series 14 (Table 2), with an encouraging
75-fold selectivity over D2.
The 3- and 2-¯uoro analogues, 15 and 16 respectively,
had a similar pro®le to that of 14 but the 3-methoxy
analogue 17 had signi®cantly higher D3 anity (pKi 8.5)
and 180-fold selectivity over D2. Unfortunately 17 gave
no exposure following oral administration in the rat,
presumably due to metabolic instability of the 3-meth-
oxyphenyl moiety either by O-demethylation or hydro-
xylation of the activated phenyl ring. The 3-acetyl
cinnamide 18 also gave high D3 anity, but signi®cantly
reduced selectivity over the D2 receptor. However, the
3-acetamidocinnamide 19 retained D3 anity and had
130-fold selectivity over D2. Following oral administra-
tion to rats, the systemic exposure of 19 was sig-
ni®cantly lower than that of SB-277011. In an attempt
to increase metabolic stability, a deactivating ¯uoro
substituent was introduced. The resulting disubstituted
cinnamide 20 had similar D3 anity and selectivity over
D2 compared with 19, and most encouragingly, 20
Figure 1. Overlap of a 7-cyano-2,3,4,5-tetrahydro-1H-benzazepine
(yellow) with SB-277011 (magenta).
presence of EDC and HOBT in dichloromethane gave
®nal compounds 7±20, which were puri®ed by chroma-
tography and isolated as their hydrochloride salts.
The e€ect of variation of the aryl carboxamide moiety
on dopamine D3 receptor anity and selectivity over
the dopamine D2 receptor was investigated8 (Tables 1
and 2), with the additional objectives of minimising the
potential for cytochrome P450 interactions and
optimising the pharmacokinetic pro®le.
The 4-quinolinylcarboxamide 7 gave reduced dopamine
D3 receptor anity (pKi 7.6) compared to SB-277011
(Table 1). The 1-naphthylcarboxamide 8 had a similar
pro®le, but the 5-quinolinyl derivative 9 was 100-fold
selective for D3 over the D2 receptor. A 2-indolyl-
carboxamide moiety had previously been shown to give
high dopamine D3 anity in the tetrahydroisoquino-
line series,6 so the corresponding benzazepine 10 was
prepared. Although 10 maintained D3 anity, selectiv-
ity over D2 was only 20-fold. In contrast, the 2-naphthyl
derivative 11 had >100-fold selectivity. Isosteric
replacement of the 2-naphthyl moiety was investigated
in order to reduce lipophilicity and improve aqueous
solubility. The 7-carbostyryl isostere 12 encouragingly
had D3 anity equivalent to that of 1, with 50-fold
selectivity over D2. Benzoxazinone 13 (D3 pKi 8.2)
achieved the target of ꢀ100-fold selectivity over D2,
but systemic exposure of 13 following oral
administration to rats was signi®cantly lower than that
of SB-277011.6,10
Table 1. Anities (pKi) of aryl carboxamide derivatives at dopamine
D3 and D2 receptors
b
b
Compounda
R
D3
D2
SB-277011 (1)
7
8
9
10
11
Ð
8.0
7.6
7.7
7.9
7.8
7.7
6.0
5.9
5.9
5.7
6.5
5.5
4-Quinolinyl
1-Naphthyl
5-Quinolinyl
2-Indolyl
2-Naphthyl
12
13
8.0
8.2
6.3
6.1
It had previously been observed in the 6-cyanotetra-
hydroisoquinoline series that a 4-¯uorocinnamide moiety
gave higher dopamine D3 receptor anity than the 4-
A corresponding
increase was seen with the 4-¯uorocinnamide in the
aAll new compounds gave satisfactory analytical and/or mass spectral
data.9
quinolinylcarboxamide group.6,11
bAll values represent the mean of at least three experiments, each
within 0.3 of the mean.
Scheme 1. Reagents: (i) 48% aq HBr, re¯ux, 3 h, 99%; (ii) (Boc)2O, NEt3, aq THF, 20 ꢁC, 18 h, 93%; (iii) (CF3SO2)2O, NEt3, CH2Cl2, 20 ꢁC, 18 h,
92%; (iv) Zn(CN)2, Pd(PPh3)4, DMF, 100 ꢁC, 4 h, 97%; (v) CF3CO2H, CH2Cl2, 20 ꢁC, 4 h, 100%; (vi) NaBH(OAc)3, ClCH2CH2Cl, 0 ꢁC, 18 h,
80%; (vii) EDC, HOBT, CH2Cl2, 20 ꢁC, 18 h, 60±85%.

N. E. Austin et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2553±2555
2555
Table 2. Anities (pKi) of cinnamide derivatives at dopamine D3
and D2 receptors
References and Notes
1. Sokolo€, P.; Giros, B.; Martres, M.-P.; Bouthenet, M.-L.;
Schwartz, J.-C. Nature 1990, 347, 146.
2. Schwartz, J.-C.; Levesque, D.; Martres, M.-P.; Sokolo€, P.
Clin. Neuropharmacol. 1993, 16, 295.
3. Shafer, R. A.; Levant, B. Psychopharmacology 1998, 135, 1.
4. Joyce, J. N.; Meador-Woodru€, J. H. Neuropsycho-
pharmacology 1997, 16, 375.
5. Pilla, M.; Perachon, S.; Sautel, F.; Garrido, F.; Mann, A.;
Wermuth, C. G.; Schwartz, J.-C.; Everitt, B. J.; Sokolo€, P.
Nature 1999, 400, 371.
6. Stemp, G.; Ashmeade, T.; Branch, C. L.; Hadley, M. S.;
Hunter, A. J.; Johnson, C. N.; Nash, D. J.; Thewlis, K. M.;
Vong, A. K. K.; Austin, N. E.; Je€rey, P.; Avenell, K. Y.;
Boy®eld, I.; Hagan, J. J.; Middlemiss, D. N.; Reavill, C.; Riley,
G. J.; Routledge, C.; Wood, M. J. Med. Chem. 2000, 43, 1878.
7. Kanao, M.; Hashizume, T.; Ichikawa, Y.; Irie, K.; Satoh,
Y.; Isoda, S. Chem. Pharm. Bull. 1982, 30, 180.
Compounda
X
D3
D2
b
b
14
15
16
17
18
19
20
4-F
3-F
2-F
3-OMe
8.2
8.2
8.0
8.5
8.4
8.5
8.4
6.4
6.4
6.2
6.2
6.7
6.3
6.3
3-COMe
3-NHCOMe
2-F, 3-NHCOMe
a,bSee footnotes to Table 1.
8. Compounds were evaluated in binding assays using dis-
placement of 125I-iodosulpride from human D3 and D2 recep-
tors, expressed in CHO cells.
showed equivalent systemic exposure to SB-277011
following oral administration in the rat.
9. 1H NMR spectra were recorded at 250 MHz in CDCl3 as
solvent. Compound 20, mp 274.5±275.5 ꢁC dec. (HCl salt);
mass spectrum (API+): Found 503 (MH+). C30H35FN4O2
requires 502. 1H (free base) d: 1.04±1.20 (m, 5H), 1.40±1.50 (m,
2H), 1.70±1.82 (m, 2H), 2.00±2.10 (m, 2H), 2.24 (s, 3H), 2.40±
2.52 (m, 2H), 2.60±2.70 (m, 4H), 2.90±3.00 (m, 4H), 3.80±3.90
(m, 1H), 5.45 (d, J=5 Hz, 1H), 6.42 (d, J=10 Hz, 1H), 7.08±
7.25 (m, 3H), 7.30±7.45 (m, 3H), 7.71 (d, J=10 Hz, 1H), 8.30
(m, 1H).
Conclusions
Using the dopamine D3 receptor antagonist SB-277011
as a starting point, and from examination of molecular
models, a series of 2,3,4,5-tetrahydro-1H-3-benzaze-
pines has been identi®ed with high anity for the
dopamine D3 receptor and selectivity over the D2
receptor. Where comparisons are available, the benza-
zepines have slightly lower dopamine D3 receptor a-
nity than the corresponding tetrahydroisoquinolines,6
although selectivity over the D2 receptor is similar.
From the series of benzazepines described above, the 3-
acetamido-2-¯uorocinnamide derivative 20 has been
identi®ed having 100-fold selectivity over the D2 recep-
tor. Subsequent evaluation showed that 20 was >100-
fold selective over a range of other receptors with a
clean P450 pro®le, had equivalent systemic exposure to
SB-277011 in the rat, and was CNS penetrant,10 with a
brain:blood ratio of 0.4:1. Compound 20 therefore
represents a structurally distinct tool compound for the
continued investigation of the role of dopamine D3
receptors in the CNS.
10. Systemic exposure following oral administration and CNS
penetration at steady-state were investigated in the rat. For
determination of systemic exposure, compounds were sus-
pended in 1% (w/v) methylcellulose aq at a concentration of
1.5 mg free base/mL and administered at a target dose of 3 mg
free base/kg. Blood samples were removed at 0.5, 1, 2, 3, 4, 6
and 10 h post-dose and analysed by LC/MS/MS. For CNS
penetration studies, compounds were dissolved in 2% (v/v)
DMSO in 5% (w/v) dextrose aq and administered at a con-
stant infusion rate over 12 h at a target dose rate of 0.3 mg free
base/kg/h. Blood samples were removed during the latter part
of the infusion to con®rm steady-state blood concentrations.
Blood and brain samples were analysed by LC/MS/MS.
11. Austin, N. E.; Avenell, K. Y.; Boy®eld, I.; Branch, C. L.;
Coldwell, M. C.; Hadley, M. S.; Je€rey, P.; Johns, A.; John-
son, C. N.; Nash, D. J.; Riley, G. J.; Smith, S. A.; Stacey, R.
C.; Stemp, G.; Thewlis, K. M.; Vong, A. K. K. Bioorg. Med.
Chem. Lett. 1999, 9, 179.