Novel benzodifuran analogs as potent 5-HT2A receptor agonists with ocular hypotensive activity

Article abstract of DOI:10.1016/j.bmcl.2007.03.073

A series of 8-substituted benzodifuran analogs was prepared and evaluated for 5-HT_2A receptor binding and activation. Several compounds containing ether and ester functionality were found to be potent agonists. Topical ocular administration of 5, 18, and 25 effectively reduced intra-ocular pressure in the hypertensive cynomolgus monkey eye in the range of 25-37%.

Full text of DOI:10.1016/j.bmcl.2007.03.073

Bioorganic & Medicinal Chemistry Letters 17 (2007) 2998–3002
Novel benzodifuran analogs as potent 5-HT2A receptor
agonists with ocular hypotensive activity
Zixia Feng, Suchismita Mohapatra,* Peter G. Klimko, Mark R. Hellberg, Jesse A. May,
Curtis Kelly, Gary Williams, Marsha A. McLaughlin and Najam A. Sharif
Alcon Research, Ltd, Research Chemistry, 6201 South Freeway, Fort Worth, TX 76134, USA
Received 14 February 2007; revised 20 March 2007; accepted 21 March 2007
Available online 25 march 2007
Abstract—A series of 8-substituted benzodifuran analogs was prepared and evaluated for 5-HT2A receptor binding and activation.
Several compounds containing ether and ester functionality were found to be potent agonists. Topical ocular administration of 5,
18, and 25 effectively reduced intra-ocular pressure in the hypertensive cynomolgus monkey eye in the range of 25–37%.
Ó 2007 Elsevier Ltd. All rights reserved.
CH3
NH2
OCH3
CH3
NH2
Glaucoma is a sight threatening optic neuropathy that is
the second leading cause of blindness in the developed
world. Therapeutic agents currently used for the treat-
ment of glaucoma reduce elevated intra-ocular pressure
(IOP), one of the most important risk factors associated
with the disease. Serotonin (5-hydroxytryptamine, 5-
HT) has been identified in the aqueous humor of
humans and other mammals,1,2 and 5-HT receptors
have been identified in relevant ocular tissues, such as
the iris-ciliary body.3,4 These findings have generated
interest regarding the role that serotonin might have in
controlling IOP. Recently, serotonin 5-HT2 receptor
agonists have been shown to be effective in lowering
IOP in the ocular hypertensive monkeys and represent
a potential new class of topical ocular hypotensive
agents.5–8
O
H3CO
O
I
Br
(R)-DOI
1
for psychotropic side effects compared to the benchmark
5-HT2A agonist (R)-DOI. The general approach to
achieve this goal was to decrease the blood brain barrier
(BBB) permeability by decreasing lipophilicity of the
compounds via substitution of a polar moiety in place
of the C8 halogen. Based on the reports of a series of tet-
rahydrobenzodifuran and benzodifuran analogs of 1 as
potent 5-HT2 agonists,10,11 we used this tricyclic nucleus
of compound ‘1’ as a platform to design molecules antic-
ipated to be potent 5-HT2A agonists with reduced lipo-
philicity. From the synthesized compounds that fit
these criteria, compounds 5, 18, and 25 were further
investigated to assess their ability to lower IOP in
laser-induced ocular hypertensive cynomolgus monkeys.
(R)-4-Iodo-2,5-dimethoxyamphetamine (R-DOI), a po-
tent hallucinogen that activates central 5-HT2A recep-
tors,9 demonstrated a pronounced reduction of IOP in
our conscious cynomolgus monkey model of laser-
induced ocular hypertension.
The goal of the present investigation was to develop no-
vel 5-HT2A receptor agonists with decreased potential
CH3
CH3
NH2
CH3
NH2
NH2
HO
O
O
O
O
O
N
N
O
Keywords: 5-HT2 agonist; Glaucoma; IOP; Benzodifuran; (R)-DOI;
Oxadiazole.
CH3
Br
O
OCH3
*
Corresponding author. Tel.: +1 817 551 4355; fax: +1 817 615
3391; e-mail: suchismita.mohapatra@alconlabs.com
25
18
5
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.03.073

Z. Feng et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2998–3002
2999
The 5-HT2A receptor binding and functional response
results are summarized in Tables 1 and 2. Incorporation
of a hydroxymethyl group at C8 position (4) resulted in
a near 10-fold decrease in agonist potency at the 5-HT2A
receptor (EC50 = 175 nM) compared to the response of
R-DOI and 1. However, the methoxymethyl ether (5)
was equally potent to R-DOI and 1 as 5-HT2A receptor
agonist and was 17-fold less lipophilic with a distribu-
tion coefficient of 0.16 (octanol–water, pH 7.4, Table
3). Incorporation of esters at C8 position resulted in po-
tent agonists as in compounds 8 and 10 (EC50 = 6.2 and
26 nM, respectively). However, introduction of a more
polar ionizable functionality such as acetic acid (11) or
propionic acid (12) resulted in a loss of 5-HT2A agonist
potency (EC50 = 4970 and >1000 nM, respectively).
There appears to be a trend toward increased affinity
for the 5-HT2A receptor and an increase in the potency
of agonist response for compounds of this series in the
order of acids < alcohols < ethers < esters for these sub-
stituents in the C8 position. Alkyl esters are readily
hydrolyzed to the corresponding acids by corneal ester-
ases following topical ocular administration, which has
Table 1. In vitro binding and functional response data for rat cloned 5-HT2A receptor
CH3
NH2
CH3
NH2
O
O
O
R
N
O
CH3
R
O
N
(1,4-6, 8-12)
(13-15)
SEMc, nM
EC50
SEMc, nM (% response)
a
b
Compound
R
IC50
(R)-DOI
0.46 0.17
0.3 0.16
1.0 0.06
0.28 0.04
0.88 0.18
0.29 0.05
14 5.4
17.8 5.01 (33)
13 4.9 (47)
175 13.9 (43)
12 1.5 (54)
71 32 (50)
1d
4
–Br
–CH2–OH
–CH2–OCH3
5
6
–CH2–O–CH2–CH2–OCH3
–CH2–O–CH2CO2C2H5
–CO2CH3
8
6.2 2.0 (39)
880 120 (48)
26 2.7 (40)
4970 867 (45)
>1000 (15)
9
10
11
12
13
14
15
–CH2–CH2CO2CH3
–CH2–CO2H
0.86 0.37
104 35
–CH2–CH2CO2H
No atom
–CH2
4700 1300
910 78
1.6 0.5
8192 1103 (27)
49.2 6.1 (50)
>10000 (7)
–CH2–CH2
1.0 0.46
a Rat cerebral cortex binding with [125I]DOI.
b Intracellular calcium mobilization in rat vascular smooth muscle cells (A7r5), agonist response relative to that of 5-HT.
c Values are means of at least three experiments.
d Compound 1 is racemic.
Table 2. In vitro binding and functional response data for rat cloned 5-HT2A receptor
CH
3
R
R
NH
2
NH2
O
NH2
O
O
O
O
O
Br
Br
R
(18-20)
(25)
(23)
SEMc , nM
EC50
SEMc, nM (% response)
a
b
Compound
R
IC50
H2C
N
CH3
18
0.38 0.054
16.9 3.5 (60)
O
N
19
20
CH2–OH
0.31 0.19
0.11 0.05
2.6 0.4
33 11.5 (38)
3.6 0.57 (72)
87.8 0.2 (56)
32.7 0.4 (96)
CH2–OCH3
–OH
–OH
23 (S,R)
25 (R,R)
0.7 0.22
a Rat cerebral cortex binding with [125I]DOI.
b Intracellular calcium mobilization in rat vascular smooth muscle cells (A7r5), agonist response relative to that of 5-HT.
c Values are means of at least three experiments.

3000
Z. Feng et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2998–3002
Table 3. Functional response data for rat 5-HT2A,2B,2C receptors for
the selected compounds
CH3
NHTFA
CH3
NHTFA
Compound
EC50 (nM)a, Emax (%)b
DC7.4
O
d
b, c
a
O
5
c
O
O
5-HT2A
5-HT2B
5-HT2C
CH3I
CHO
OH
c, d
(R)-DOI
5
17.8 (33)
12 (54)
71 (50)
11.9
—
30.2 (84)
12.6 (99)
8.6 (93)
4.5 (76)
689 (69)
82 (95)
2.72
0.16
0.096
0.31
0.22
2.92
1.30
OCH3
CH3
NH2
CH3
NHCbz
Br
2
3
6
5.8
59
b, c
O
O
BrCH CO Et
2
2
8
6.2 (39)
49.2 (50)
16.9 (60)
32.7 (96)
c
8
O
14
18
25
130
23
O
b, e
OCH
3
O
OH
4
6.2
39 (100)
6
7
a Intracellular calcium mobilization in rat vascular smooth muscle cells
(A7r5).
b Relative to maximal 5-HT induced response.
c From rat fundus.
d DC7.4, distribution coefficient at pH 7.4.
Scheme 1. Reagents and conditions: (a) NaCNBH4, AcOH, 0 °C, 2 h,
92%, (b) Ag2O, KI, CH3CN:CH2Cl2 (1:1), 23 °C, 12 h, 72%, (c) 5 N aq
NaOH, MeOH/H2O (4:1), 12 h, 23 °C, 93%, (d) Cbz-Cl, aq NaOH,
THF, 23 °C, 4 h, 94%, (e) 10% Pd/C, H2, MeOH, 23 °C, 12 h, 92%.
been used as an effective prodrug approach when the
carboxylic acid has the desired activity, for example,
for prostaglandin analogs (travoprost, latanoprost).
However, as mentioned above, in the present case incor-
poration of a carboxylic acid moiety into the molecule
dramatically reduced agonist activity. Therefore, it was
of interest to explore the incorporation of a bioisostere
surrogate for the carboxylic ester moiety; the 3-methyl-
1,2,4-oxadiazol-5-yl functionality was selected for this
purpose and was incorporated into the C8 position
either by direct substitution (13) or through a methyl
or ethyl link, 14 and 15, respectively. The 8-[2-(3-
methyl-1,2,4-oxadiazol-5-yl)-ethyl] substituent (14) dis-
played an agonist response at the 5-HT2A receptor
(EC50 = 49.2 nM).
CH3
NH2
CH3
NHTFA
CH3
NHTFA
O
a, b, c
d, e
O
O
O
O
O
N
N
CH3
CO2C2H5
O
16
17
18
Scheme 2. Reagents and conditions: (a) Ethyl chlorooxoacetate,
AlCl3, CH2Cl2, 25 °C, 12 h, 68%, (b) Et3SiH, TFA, 80 °C, 5 h, 91%,
(c) DDQ, dioxane, reflux, 3 h, 82%, (d) Acetamide oxime, NaH, THF,
Molecular sieves 4 °A, 60 °C, 3 h, 67%, (e) K2CO3, MeOH, reflux,
12 h, 94%.
The synthesis of oxadiazoles 13–15 was achieved by
reacting the corresponding methyl esters with acetamide
oxime and NaH following the reported procedure.14 For
the synthesis of oxadiazole 18, ethyl ester 17 was treated
with acetamide oxime and NaH followed by hydrolysis
with K2CO3 (Scheme 2). Compound 18 was well charac-
terized from its spectral data.15
Further SAR was done around the benzodifuran ring as
in Table 2. Oxidation of the tetrahydrobenzodifuran
ring to the benzodifuran ring increased 5-HT2A agonist
potency. As a comparison, alcohol 19 is fivefold more
potent than 4. A similar trend has been noticed with
the benzodifuran methyloxadiazole 18 and methyl ether
20, and they are 2.9 and 3.3 times more potent at the
5-HT2A receptor than 14 and 5, respectively. To further
explore the SAR, substitution of hydroxyl group in the
beta carbon to the amine was done. The (R,R)-b-hydro-
xy compound 25 was found to be a more potent 5-HT2A
agonist (EC50 = 32 nM) than the (S,R) isomer 23
(EC50 = 88 nM) and less lipophilic having DC7.4 to be
1.30 as compared to (R)-DOI (DC7.4 = 2.72).
The aldehyde 2 was oxidized with 2 equiv of DDQ to
provide the benzodifuran nucleus and was used for the
synthesis of alcohol 19 and methyl ether 20 using the
same methods as described for 4 and 5, respectively.
CH3
NHCOCF3
O
O
Tetrahydrobenzodifuran analogs 4, 5,12 6, and 8 (Table
1) were prepared from aldehyde 2 (Scheme 1). Aldehyde
2 was synthesized from its precursor 1610 using
Cl2CHOCH3 and SnCl4 following the literature proce-
dure.11 For the synthesis of ester 9, aldehyde 2 was
oxidized with NaClO2/KH2PO4/2-methyl-2-butene13
followed by hydrolysis to afford the corresponding ami-
no acid. The acid was esterified with HCl in methanol to
afford 9. Acid 12 was made by Wittig olefination of 2
with PPh3@CHCO2Me, hydrogenation, and basic
hydrolysis. Ester 10 was prepared from 12 by esterifica-
tion with 2,20-DMP and cat. HCl. Similarly, acid 11 was
prepared from 16 by Friedel–Crafts acylation with ethyl
chlorooxoacetate and AlCl3, reduction with Et3SiH/
TFA followed by basic hydrolysis.
O
c, d
a, b
23
O
O
O
Cl
c, e
Br
21
NHCOCF3
F3C
22
N
O
CH3
O
f, d
25
O
Br
24
Scheme 3. Reagents and conditions: (a) Br2, AcOH, 0 °C, 2 h, 72%, (b)
AlCl3, CH2Cl2, 12 h, 23 °C, 32%, (c) PhMe2SiH, TFA, 0 °C, 2 h, 47%,
(d) 5 N aq NaOH, MeOH, H2O (4:1), 12 h, 89% (e) MsCl, Et3N,
CH2Cl2, 0 °C, 30 min, (f) 0.5 M HCl, THF/ether (1:3), 47%.

Z. Feng et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2998–3002
3001
5
0
IOP response in hypertensive eye of the cynomolgus
monkeys
Compounds 5, 18 and 25
-5
-10
-15
-20
-25
-30
-35
-40
-45
5 (300 µg)
18 (300 µg)
25 (300 µg)
(R)-DOI (300 µg)
Mean vehicle response
0
1
2
3
4
5
6
Time (hr) post dose
Figure 1. Monkey IOP response for 5, 18, and 25.
The chiral b-hydroxy analogs 23 and 25 were prepared
from 2110 (Scheme 3). The (S,R)-isomer 23 was prepared
by stereoselective reduction of the ketone 22 using
PhMe2SiH in TFA8 followed by basic hydrolysis. Simi-
larly, the (R,R)-isomer 25 was made via a five-membered
cyclic oxime intermediate 24 and was opened by 0.5 M
HCl.16 Hydrolysis by aq NaOH afforded the free base
25.17
for each dose are means of at least eight animals SEM.
Values for the vehicle are means of at least five ani-
mals SEM. The plotted vehicle curve represents the
mean vehicle response for the indicated studies.
Compounds 5, 18, and 25 exhibited potent and effica-
cious ocular hypotensive activity in this model, affording
maximum IOP reductions of 36.8%, 27%, and 24%,
respectively, at 300 lg doses.
The procedures used for the in vitro binding and in vivo
functional response were previously described.5–8 In
brief, the relative affinities of compounds at the 5-
HT2A receptor were determined by measuring their abil-
ity to compete for the binding of the agonist radioligand
In conclusion, incorporation of different polar groups in
place of bromine in 1 lowered the overall lipophilicity of
the compounds and led to analogs with high binding
affinity and high agonist potency at 5-HT2A receptor.
Representative examples of both the terahydrobenzo-
difuran and benzodifuran series (5, 18, and 25) effec-
tively reduce IOP in the laser-induced hypertensive
eyes of conscious cynomolgus monkey.
[
125I] DOI to rat brain 5-HT2A receptors. To measure
the functional activity at the 5-HT2A receptor, recep-
tor-mediated mobilization of intracellular calcium
([Ca2+]i) was studied using the FLIPR instrument with
rat vascular smooth muscle cells, A7r5, expressing na-
tive 5-HT2 receptors. The 5-HT2B receptor functional
response was determined using rat isolated stomach
fundus and the assay was conducted by MDS Pharma
services, Bothell, WA, using methods previously
described.18 The 5-HT2C receptor functional activity
was determined as for the 5-HT2A receptor above,
except that SR3T3 cells expressing the recombinant rat
5-HT2C receptor were utilized.
Further studies are required to establish the CNS
side effect profile of these compounds, and to establish
the 5-HT2 subtype receptors involved in IOP
lowering.
References and notes
1. Veglio, F.; De Santis, U.; Schiavone, D.; Cavallone, S.;
Mulatero, P.; Grignolo, F. M.; Chiandussi, L. Ophthal-
mologica 1998, 212, 160.
2. Boerrigter, R. M. M.; Siertsema, J. V.; Kema, I. P. Doc.
Ophthalmol. 1992, 82, 41.
3. Chidlow, G.; Le Core, S.; Osborne, N. N. Neuroscience
1998, 87, 675.
4. Martin, X. D.; Malina, H. Z.; Brennman, M. C.; Hend-
rickson, P. H.; Lichter, P. R. Eur. J. Ophthalmol. 1992, 2,
67.
Based on their 5-HT2 receptor binding and functional
activation data (vide supra), benzodifurans 5, 18, and
25 were evaluated for their ability to lower IOP in con-
scious cynomolgus monkeys with laser-induced ocular
hypertension in the right eyes following the detail proce-
dure as described before.7 Prior assessment of each com-
pound’s safety was performed by bilateral instillation of
one drop of a 1% ophthalmic formulation of the test
compound in five New Zealand Albino rabbits. Gross
observations were made for local or systemic effects dur-
ing the 2-h interval after dosing. For Figure 1, the values
5. May, J. A.; Dantanarayana, A. P.; Zinke, P. W.;
McLaughlin, M. A.; Sharif, N. A. J. Med. Chem. 2006,
49, 318.

3002
Z. Feng et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2998–3002
6. May, J. A.; Chen, H. H.; Rusinko, A.; Vincent, M. L.;
Sharif, N. A.; McLaughlin, M. A. J. Med. Chem. 2003, 46,
4188.
7. May, J. A.; McLaughlin, M. A.; Sharif, N. A.; Hellberg,
M. R.; Dean, T. R. J. Pharmacol. Exp. Ther. 2003, 306,
301.
8. Glennon, R. A.; Bondarev, M. L.; Khorana, N.; Young,
R.; May, J. A.; Hellberg, M. R.; McLaughlin, M. A.;
Sharif, N. A. J. Med. Chem. 2004, 47, 6034.
9. Nichols, D. E. Pharmacol. Ther. 2004, 101, 131.
10. Chambers, J. J.; Kurrasch-Orbaugh, D. M.; Parker, M.
A.; Nichols, D. E. J. Med. Chem. 2001, 44, 1003.
11. Monte, A. P.; Marona-Lewicka, D.; Parker, M. A.;
Wainscott, D. B.; Nelson, D. L.; Nichols, D. E. J. Med.
Chem. 1996, 39, 2953.
3.40 (m, 2H), 3.70–3.72 (m, 1H), 4.66–4.68 (m, 2H), 6.97
(d, 2H, J = 13.6 Hz), 7.76–7.77 (m, 2H); 3C NMR
(CD3OD, 100 MHz): d 11.19, 18.72, 25.20, 33.37, 49.28,
106.35, 106.52, 108.86, 111.38, 126.84, 127.08, 147.70,
147.89, 151.60, 151.97, 168.51, 179.04; LC/MS (m/z) =
312 (M+1)+; Anal. Calcd for C17H17N3O3Æ2.0HClÆ0.2
H2O: C, 52.64; H, 5.04; N, 10.83. Found: C, 52.53; H,
4.83; O, 10.60; (b) The compounds are converted to their
HCl salt by dissolving the sample in minimum amount of
methanol followed by addition of 1 M HCl in ether until
pH 4. Ether was added to precipitate the HCl salt,
filtering the solids and drying afforded the corresponding
HCl salts.
16. Somanathan, R.; Aguilar, H. R.; Rivero, I. A.; Aguirre,
G.; Hellberg, L. H.; Yu, Z.; Thomas, J. A. J. Chem. Res.;
Synop. 2001, 92, 348.
12. Spectral data of compound 5: 1H NMR (CD3OD,
400 MHz): d 1.30 (d, 3H, J = 6.4 Hz), 2.81–2.76 (m, 1H),
2.91–2.86 (m, 1H), 3.24–3.14 (m, 4H), 3.34–3.32 (m, 3H),
3.64–3.60 (m, 1H), 4.42 (s, 2H), 4.60–4.55 (m, 4H); 13C
NMR (CD3OD, 100 MHz): d 18.57, 30.0, 30.10, 34.16,
48.78, 58.30, 67.97, 72.56, 72.86, 115.39, 116.38, 127.14,
127.68, 153.77, 154.23; LC/MS (m/z) = 264 (M+1)+; Anal.
Calcd for C15H21NO3Æ1.0HCl: C, 60.10; H, 7.40; N, 4.67.
Found: C, 60.09; H, 7.39; N, 4.61.
17. (a) Spectral data of compound 25: 1H NMR (CD3OD,
400 MHz): d 1.15 (d, 3H, J = 6.4 Hz), 3.15–3.13 (m, 2H),
3.31–3.30 (m, 1H), 3.45–3.43 (m, 1H), 3.61–3.59 (m, 1H),
4.61–4.56 (m, 4H), 4.70 (d, 1H, J = 8 Hz); 13C NMR
(CD3OD, 100 MHz): d 15.52, 31.82, 32.84, 52.54, 71.22,
72.75, 73.17, 100.67, 119.10, 127.20, 128.90, 153.11,
153.43; LC/MS (m/z) = 315 (M+1)+; Anal. Calcd for
C13H16BrNO3Æ1.0HClÆ0.7H2O: C, 42.99; H, 5.10; N,
3.85. Found: C, 42.92; H, 4.92; N, 3.79; (b) The relative
stereochemistry of both isomers 23 and 25 was assigned on
13. Kraus, G. A.; Taschner, M. J. J. Org. Chem. 1980, 45,
1175.
1
14. Showell, G. A.; Gibbons, T. L.; Kneen, C. O.; MacLeod,
A. M.; Merchant, K.; Saunders, J.; Freedman, S. B.; Patel,
S.; Baker, R. J. Med. Chem. 1991, 34, 1086.
the basis of H NMR shift of –CH–OH protons and was
compared with the analogous structures as reported in
Ref. 8.
18. Cohen, M. L.; Fludinski, L. A. J. Pharmacol. Exp. Ther.
1987, 243, 264.
15. (a) Spectral data of compound 18: 1H NMR (CD3OD,
400 MHz): d 1.21 (d, 3H, J = 6.4 Hz), 2.18 (s, 3H), 3.21–