Synthesis and characterization of 5,6,7,8-tetrahydroquinoline C5a receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2008.03.049

A novel series of substituted 2-aryl-5-amino-5,6,7,8-tetrahydroquinoline C5a receptor antagonists is reported. Synthetic routes were developed that allow the substituents on the tetrahydroquinoline core to be efficiently varied, facilitating determination of structure-activity relationships. Members of the series display high binding affinity for the C5a receptor and are potent functional antagonists.

Full text of DOI:10.1016/j.bmcl.2008.03.049

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2544–2548
Synthesis and characterization of 5,6,7,8-tetrahydroquinoline
C5a receptor antagonists
J. Kent Barbay,a,* Yong Gong,a Mieke Buntinx,b Jian Li,a Concha Claes,b
Pamela J. Hornby,a, Guy Van Lommen,b Jean Van Wauweb and Wei Hea,à
aDrug Discovery, Johnson & Johnson PRD, East Coast RED, Spring House, PA 19477-0776, USA
bDrug Discovery, Johnson & Johnson PRD, RED EU, Turnhoutseweg 30, B-2340, Beerse, Belgium
Received 14 February 2008; revised 14 March 2008; accepted 17 March 2008
Available online 20 March 2008
Abstract—A novel series of substituted 2-aryl-5-amino-5,6,7,8-tetrahydroquinoline C5a receptor antagonists is reported. Synthetic
routes were developed that allow the substituents on the tetrahydroquinoline core to be efficiently varied, facilitating determination
of structure–activity relationships. Members of the series display high binding affinity for the C5a receptor and are potent functional
antagonists.
ꢀ 2008 Elsevier Ltd. All rights reserved.
The complement system plays an important role in the
body’s immune defense.1 Activation of complement trig-
gers a cascade involving proteolysis of serum proteins to
active peptides. The anaphylatoxin C5a, formed by pro-
teolysis of complement component 5 (C5), is a potent
pro-inflammatory mediator involved in recruitment
and activation of leukocytes. The functional effects of
C5a are mediated by interaction with the G-protein cou-
pled C5a receptor (C5aR, CD88) expressed by target
cells.2
blocks C5 proteolysis, inhibiting the formation of C5a
and C5b. Considerable effort has also been directed to-
ward the discovery of small molecule drugs capable of
blocking the complement response,5 and in particular
C5aR signaling.6
3-Substituted-6-arylpyridine C5aR antagonists (general
structure 1, Fig. 1) were discovered by scientists at
Neurogen Corporation.7 Aided in part by molecular
modeling, we hypothesized that these antagonists could
be biased toward their proposed bioactive conformation
by the addition of a cyclic tether (converting the pyri-
dine core to a 5,6,7,8-tetrahydroquinoline).8 We report
here the synthesis, C5aR binding and functional antag-
onism, and pharmacokinetic properties of the members
of the resulting series.
Inappropriate and/or excessive complement activation
has been implicated in the pathology of a number of
autoimmune and inflammatory diseases, and therapeu-
tic agents capable of attenuating the complement re-
sponse have been widely sought.3 Eculizumab
(SolirisTM, Alexion Pharmaceuticals, Inc.), the first mar-
keted complement directed therapy, recently received
FDA approval for treatment of paroxysmal nocturnal
hemoglobinuria (PNH).4 This C5-directed antibody
In the first phase of this study, the amino group at the 5-
position was varied. The preferred synthetic route for
varying this substituent incorporates the amino group
Keywords: C5a receptor antagonist; C5aR; C5a; Complement;
Tetrahydroquinoline.
*
Corresponding author. Tel.: +1 215 628 5033; fax: +1 215 628
4985; e-mail addresses: kbarbay@prdus.jnj.com; kentbarbay@
comcast.net
Present address: Centocor, 145 King of Prussia Road, Radnor, PA
19087, USA.
Figure 1. Generalized structures of disclosed 3-substituted-6-aryl
pyridine C5a receptor antagonists7 (1) and proposed 5,6,7,8-tetrahy-
droquinolines (2).
à
Present address: Adolor Corporation, 700 Pennsylvania Drive,
Exton, PA 19341, USA.
0960-894X/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.03.049

J. K. Barbay et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2544–2548
2545
at the last step (Scheme 1). Cyclocondensation of 3-ami-
no-2-cyclohexen-1-one (3) with malonic acid bis(2,4,6-
trichlorophenyl) ester (4) and chlorination of the crude
bis-phenol intermediate with phosphorus(V) oxychlo-
ride afforded 5. Suzuki cross-coupling with 2,6-diet-
hylphenylboronic acid occurred predominantly at the
2-position. Refluxing intermediate 6 with sodium meth-
oxide in methanol resulted in substitution of the 4-
chloro group. Installation of the amino group at the 5-
position was accomplished by reduction of the ketone,
activation of alcohol 8 as the corresponding chloride,
and displacement with an amine. The displacement reac-
tions tended to be sluggish; excess amine and potassium
carbonate (4 equiv each) were used to drive the reaction
and decrease the amount of elimination byproduct
formed. The reactions were typically conducted at ambi-
ent temperature using an extended reaction period (1–
3 d); higher temperatures promoted elimination. In sev-
eral instances in which it was desired to independently
vary the two amino substituents, or due to commercial
availability of the amine starting material, a primary
amine was used in the displacement reaction and the sec-
ond alkyl group (Me or Et) was subsequently installed
by reductive amination (compounds 28, 33–34, 36–37,
and 42, Table 2).
of these substituents was installed at the last step.9 These
routes proved more challenging than that of Scheme 1;
displacements of the chlorides prepared from alcohols
11 and 14 with N-ethyl-1-naphthylamine required heat-
ing and were less efficient than those using the chloride
derived from 8. Nevertheless, preparation of 2-chloro-
tetrahydroquinoline 12 allowed installation of the 2-po-
sition aryl substituent in the final step by a microwave-
accelerated Suzuki reaction, allowing rapid generation
of structure–activity relationships (Scheme 2a). Two
analogs varying at the 4-position were produced by
the route shown in Scheme 2b (60–61), while others
(62–63) were prepared using the route shown in Scheme
1, with alternative nucleophiles substituted for sodium
methoxide.
Compounds were screened for their ability to block
C5a-induced intracellular calcium mobilization in the
human monocytic cell line U937.10 An initial set of ana-
logs that revealed potential for C5a receptor antagonist
activity within the series incorporated 1,2,3,4-tetrahy-
droisoquinolines at the 5-position (Table 1). The unsub-
stituted tetrahydroisoquinoline analog 17 had an IC50 of
1.1 lM;
a brief screen of substitutions achieved
Table 1. C5aR antagonist activity of 5-position tetrahydro-
isoquinolines
In order to efficiently prepare analogs that varied at the
2- and 4-position, syntheses were explored in which each
Compounda
R
C5a-induced Ca2+
flux IC50 (lM)
17
18
19
20
21
22
23
24
25
26
H
5-Cl
1.1
1.7
5-CF3
6-OMe,7-OMe
7-Cl
>10
>10
1.4
7-F
0.88
>10
>10
0.84
0.27
7-OMe
7-CF3
8-CO2Me
8-CF3
Scheme 1. Reagents and conditions: (a) PhBr, reflux, Ar = 2,4,6-
trichlorophenyl; (b) POCl3, reflux; (c) 2,6-DiEtPhB(OH)2, Pd(PPh3)4,
Na2CO3, H2O, toluene, reflux; (d) NaOMe, MeOH, reflux; (e) NaBH4,
MeOH; (f) SOCl2, CH2Cl2; (g) R1R2NH, K2CO3, CH3CN.
a Except where noted, compounds were tested as racemates.
Scheme 2. Reagents and conditions: (a) NaOMe, MeOH, 23 ꢁC (44% 10, 40% separable 2-methoxy-4-chloro regioisomer); (b) NaBH4, MeOH; (c)
SOCl2, CH2Cl2; (d) N-ethyl-1-naphthylamine, K2CO3, CH3CN, reflux; (e) ArB(OH)2, PdCl2(PPh3)2, Na2CO3, H2O, CH3CN, lW 120–160 ꢁC,
10 min; (f) N-ethyl-1-naphthylamine, K2CO3, CH3CN, 110 ꢁC (sealed tube); (g) ROH, RONa, lW, 155–175 ꢁC.

2546
J. K. Barbay et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2544–2548
Table 2. C5aR antagonist activity of 5-position naphthylamines and analogs
Compound
R1
R2
C5a-induced Ca2+ flux IC50 (lM)
27
28
1-Naphthyl
1-Naphthyl
Et
0.10
0.17
0.19
6.2
Me
2-OH–Et
Acetyl
H
29
1-Naphthyl
1-Naphthyl
1-Naphthyl
30a
31
>10
>10
>10
1.5
32
33
H
2-Naphthyl
Et
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
34
35
2-Me-1-naphthyl
6-OMe-1-naphthyl
5-OMe-1-naphthyl
4-OMe-1-naphthyl
(S)-1,2,3,4-Tetrahydro-1-naphthylb
(S)-1,2,3,4-Tetrahydro-1-naphthylc
(R)-1,2,3,4-Tetrahydro-1-naphthylb
(R)-1,2,3,4-Tetrahydro-1-naphthylc
5,6,7,8-Tetrahydro-1-naphthyl
Phenyl
0.18
>10
0.53
0.82
5.5
36
37
38
39
40
41
1.9
8.1
42
43
0.22
7.9
Me
Et
44
45
Benzyl
>10
>10
0.027
>10
–CH2-1-naphthyl
1-Naphthyl
1-Naphthyl
Me
Et
(À)-27d
(+)-27d
Et
a Prepared by the reaction of N-acetyl-1-naphthylamine with the chloride derived from 8 (NaH, THF, 23 ꢁC).
b Less polar diastereomer (SiO2 TLC, 25% EtOAc:hexanes); tetrahydroquinoline 5-position absolute stereochemistry undetermined.
c More polar diastereomer.
d Enantiomeric excesses (ee): (À)-27 >99%, (+)-27 98.6% by analytical chiral SFC (chiralcel OD-H column, 0.46 · 50 cm; mobile phase 70:30
CO2:modifier (0.2% i-Pr2NH in MeOH), flow rate 3 mL/min, 50 ꢁC).
Table 3. C5aR antagonist activity of analogs varying at the 2-position
Table 4. C5aR antagonist activity of analogs varying at the 4-position
Compound
R
C5a-induced
Ca2+ flux IC50 (lM)
Compound
X
C5a-induced Ca2+ flux IC50 (lM)
60
61
62
63
64a
15
OEt
OCH2Ph
0.067
0.27
O-Cyclopentyl 0.037
46
47
48
49
50
51
52
53
54
55
56
57
58
59
12
2-Me,6-Me phenyl
2-Me,4-OMe,6-Me phenyl
2-OMe,6-OMe phenyl
2-OMe,6-Cl phenyl
2-Me phenyl
2-Et phenyl
0.13
0.17
1.8
SMe
H
0.082
>10
2.0
0.11
0.35
0.65
1.7
Cl
a Prepared from 1,2,5,6,7,8-hexahydroquinoline-2,5-dione by the route
shown in Scheme 1.
2-i-Pr phenyl
2-Ph phenyl
2-F phenyl
>10
4.6
1-Naphthyl
Phenyl
5.9
>10
>10
>10
>10
>10
Table 5. C5aR binding assay results
3-Thienyl
3-Pyridyl
Compound
[
125I]-C5a binding Compound
[
125I]-C5a
5-Benzo-1,3-dioxolyl
Cl
Ki (lM)
binding Ki (lM)
15
17
26
27
28
29
0.18
43
48
51
54
55
56
0.68
1.2
0.035
0.0097
0.0089
0.0073
0.18
0.086
3.7
improvement of the potency to 270 nM for the 8-trifluo-
romethyl substituted antagonist 26. Screening a variety
of additional cyclic and acyclic substituted amines (data
not shown) led to the identification of an active 1-naph-
0.37
8.3

J. K. Barbay et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2544–2548
2547
Table 6. Human liver microsome (HLM) stability and rat pharmacokinetic profile of compound 27a
Compound
HLM stability
(t1/2, min)
Cmax, po SD
(lM)
Cl, iv SD
(mL/min/kg)
t1/2, po SD
(h)
F (%)
27
54
1.3 0.3
31
4
1.8 0.2
59
a Dosed po at 10 mg/kg, iv at 2 mg/kg, n = 4.
thylamine subseries (Table 2). Compound 27, bearing an
N-ethyl-1-naphthylamine substituent, had an IC50 of
100 nM. Modest changes to the N-alkyl group (ethyl
to methyl or 2-hydroxyethyl, 28–29) were tolerated,
while replacing this group with an acetyl substituent re-
sulted in lM level activity. Removal of the N-alkyl
group caused a decrease in activity (31, >10 lM). The
naphthalene group was also required; N-ethyl secondary
amine 32 did not display significant antagonist activity.
Attachment of the amine substituent at the 1-position of
the naphthalene ring appeared to be preferred, as 2-
naphthyl analog 33 displayed decreased activity. A small
set of methoxy-substituted 1-naphthylamines (35–37)
generally displayed reduced activity, with 6-methoxy
substitution affording the most active compound (35,
180 nM). 1,2,3,4-Tetrahydronaphthalenes 38–41 (each
tested as a single diastereomer) displayed activities rang-
ing from 820 nM to 8.1 lM, while reduction of the
naphthalene to the 5,6,7,8-tetrahydronaphthalene was
better tolerated (42, 220 nM). Replacement of the naph-
thalene with an unsubstituted phenyl reduced activity
(43, 7.9 lM) as did the homologation to the naphtha-
len-1-ylmethanamine 45 (>10 lM).
The effect of the stereochemistry of 27 on its C5aR
antagonist activity was investigated. The enantiomers
of this compound were resolved by chiral supercritical
fluid chromatography (SFC). The levorotary enantio-
mer (absolute stereochemistry undetermined) was
shown to be primarily responsible for activity ((À)-27
IC50 27 nM; (+)-27 IC50 > 10 lM).
The human liver microsome stability and pharmacoki-
netic profile of racemic 27 in rats were determined
(Table 6). The compound displayed 59% oral bioavail-
ability. It had reasonable human liver microsome stabil-
ity (t1/2 54 min; 88% remaining at 10 min), but displayed
moderate clearance (31 mL/min/kg).
In conclusion, a series of substituted 5,6,7,8-tetrahydro-
quinoline C5aR antagonists was discovered. Structure–
activity relationships at the 2-, 4-, and 5-position of
the tetrahydroquinoline core were studied. The com-
pounds displayed activity in binding (displacement of
[
125I]-C5a) and functional (C5a-induced calcium mobili-
zation) assays in a human cell line. These results add to
the growing body of evidence that C5aR can be potently
antagonized with non-peptidic small molecules. Further
exploration of the series will be reported separately.
In the next phase of the study the N-ethyl-1-naphthalene
substituent was held constant while the aryl group at the
2-position was varied (Table 3). In general, 2,6-disubsti-
tuted analogs (46–49) were more potent C5a receptor
antagonists than 2-monosubstituted compounds (50–
54). Continuing the trend, the unsubstituted phenyl ana-
log 56 did not show significant activity, nor did several
heterocycles unsubstituted at the ortho carbons (57–
59). These results suggest that maintaining a twisted
biaryl conformation is important for activity.
Acknowledgments
We thank the ADME team (J&JPRD, Spring House,
PA) for pharmacokinetic data, Michel Carpentier
(J&JPRD, Beerse, Belgium) for enantiomeric separation
of compound 27, and Dr. Christopher Teleha (J&JPRD,
Spring House, PA) for supplying intermediate 5.
Finally, structure–activity relationships at the 4-position
were explored by varying this group while holding con-
stant the N-ethyl-1-naphthylamine and 2,6-diethylphe-
nyl substituents (Table 4). Alkoxy groups larger than
methoxy groups were found to be capable of improving
activity (ethoxy, 67 nM; cyclopentoxy, 37 nM). Re-
moval of this substituent, or replacement with a chloro,
resulted in lower potency (IC50 > 10 and 2.0 lM, respec-
tively). A thiomethyl replacement for the methoxy group
was tolerated (82 nM).
References and notes
1. (a) Morgan, B. P. Complement. Clinical Aspects and
Relevance to Disease; Academic Press Limited: London,
1990; (b) Gasque, P. Mol. Immunol. 2004, 41, 1089.
2. Gerard, C.; Gerard, N. P. Annu. Rev. Immunol. 1994, 12,
775.
3. (a) Makrides, S. C. Pharmacol. Rev. 1998, 50, 59; (b)
Mizuno, M.; Morgan, B. P. Curr. Drug Targets Inflamm.
Allergy 2004, 3, 87; (c) Ricklin, D.; Lambris, J. D. Nat.
Biotechnol. 2007, 25, 1265.
The binding affinity of selected compounds to the C5a
receptor was assessed by their ability to displace [125I]-
C5a from differentiated U937 cells (Table 5).11 A num-
ber of compounds demonstrated binding to the receptor
with high affinity (for instance, compound 27, Ki
8.9 nM). These results provide evidence that the ob-
served functional antagonism of members of the series
is a consequence of binding to C5aR (as opposed to a
mechanism involving the downstream signaling
cascade).
4. (a) Zareba, K. M. Drugs Today 2007, 43, 539; (b) Rother,
R. P.; Rollins, S. A.; Mojcik, C. F.; Brodsky, R. A.; Bell,
L. Nat. Biotechnol. 2007, 25, 1256.
5. Holland, M. C. H.; Morikis, D.; Lambris, J. D. Curr.
Opin. Investig. Drugs 2004, 5, 1164.
6. Reviews: (a) Wong, A. K.; Taylor, S. M.; Fairlie, D. P.
iDrugs 1999, 2, 686; (b) Pellas, T. C.; Wennogle, L. P.
Curr. Pharm. Des. 1999, 5, 737; (c) Taylor, S. M.; Fairlie,
D. P. Exp. Opin. Ther. Patents 2000, 10, 449; (d)
Sumichika, H. Curr. Opin. Investig. Drugs 2004, 5, 505;
(e) Hutchison, A. J.; Krause, J. E. Annu. Rep. Med. Chem.

2548
J. K. Barbay et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2544–2548
2004, 39, 139; (f) Allegretti, M.; Moriconi, A.; Beccari, A.
and 100 lg/mL streptomycin (all Invitrogen). U937 cells
were treated with 1 mM dibutyryl cAMP (Sigma) for 48–
72 h to induce differentiation prior to assessment of
intracellular Ca2+ mobilization according to a modified
protocol described in Van Lommen et al. (Bioorg. Med.
Chem Lett. 2005, 15, 497). Briefly, Fluo-3-loaded cells
(200,000 cells/well) were dissolved in HBSS medium
(Invitrogen) supplemented with 10 mM Hepes, 2.5 mM
probenecid, and 0.1 % BSA (pH = 7.4) and pre-incubated
for 20 min with test compound (1% DMSO) before
recombinant human C5a (1.5 nM, Sigma) was added.
R.; Di Bitondo, R.; Bizzarri, C.; Bertini, R.; Colotta, F.
Curr. Med. Chem. 2005, 12, 217; (g) Mizuno, M.; Cole, D.
S. Expert Opin. Investig. Drugs 2005, 14, 807; (h) Proctor,
L. M.; Woodruff, T. M.; Taylor, S. M. Expert Opin. Ther.
Patents 2006, 16, 445; (i) Monk, P. N.; Scola, A.-M.;
Madala, P.; Fairlie, D. P. Br. J. Pharmacol. 2007, 152, 429;
Recent examples: (j) Sumichika, H.; Sakata, K.; Sato, N.;
Takeshita, S.; Ishibuchi, S.; Nakamura, M.; Kamahori, T.;
Ehara, S.; Itoh, K.; Ohtsuka, T.; Ohbora, T.; Mishina, T.;
Komatsu, H.; Naka, Y. J. Biol. Chem. 2002, 51, 49403; (k)
Taylor, S. M.; Fairlie, D. P. In Structural Biology of the
Complement System; Lambris, J. D., Morikis, D., Eds.;
CRC Press, Taylor & Francis Group: Boca Raton, FL,
2005; pp 341–362.
Changes in intracellular free Ca2+ concentration (Fmax
)
were measured using FLIPR technology (Molecular
Devices). IC50 values were calculated using non-linear
regression in Graphpad Prism.
7. Hutchison, A.; Yuan, J.; Lee, K.; Maynard, G.; Chenard,
B. L.; Liu, N.; Guo, Q.; Guo, Z.; Hrnciar, P. PCT Int.
Appl. WO 2004043925.
8. C5aR homology models: (a) Gerber, B. O.; Meng, E. C.;
Do¨tsch, V.; Baranski, T. J.; Bourne, H. R. J. Biol. Chem.
2001, 276, 3394; (b) Higginbottom, A.; Cain, S. A.;
Woodruff, T. M.; Proctor, L. M.; Madala, P. K.; Tyndall,
J. D. A.; Taylor, S. M.; Fairlie, D. P.; Monk, P. N. J. Biol.
Chem. 2005, 280, 17831.
9. All compounds shown in Table 3 were prepared by the
route depicted in Scheme 2a. Except as specifically noted
in the text, all other compounds were prepared using the
route of Scheme 1.
10. The human monocytic cell line U937 (ATCC CRL-1593)
was maintained in RPMI-1640 medium supplemented
with 10% fetal bovine serum, 10 mM Hepes, 1 mM
pyruvate, 2 mM L-glutamine, 100 IU/mL penicillin G,
11. In C5aR competition binding assays, dibutyryl cAMP-
differentiated U937 cells were dissolved in binding buffer
consisting of 50 mM Hepes, 5 mM MgCl2, 1 mM CaCl2,
0.1% NaN3, and 0.02% protease-free BSA (pH 7.4). After
pre-incubating the cells (2 · 105 cells/well) with test com-
pound (1% DMSO) for 30 min at 25 ꢁC, [125I]-C5a (Bolton
& Hunter labeled, Perkin-Elmer, 2200 Ci/mmol, 0.05 nM)
was added for 60 min while the cells were kept at 4 ꢁC.
Non-specific binding was defined in wells containing
100 nM recombinant C5a. Cells were harvested on GF/B
filter plates (presoaked in 0.5% polyethyleneimine) using a
Packard cell harvester. The filter plates were washed with
binding buffer supplemented with 500 mM NaCl and filter
bound radioactivity was determined by liquid scintillation
counting on a TopCount NXTTM (Packard). Ki values were
calculated using the equation of Cheng and Prusoff
(Biochem. Pharmacol. 1973, 22, 3099) in Graphpad Prism.