2-Alkyl-4-aryl-pyrimidine fused heterocycles as selective 5-HT2A antagonists

Article abstract of DOI:10.1016/j.bmcl.2008.01.090

The synthesis and SAR for a novel series of 2-alkyl-4-aryl-tetrahydro-pyrido-pyrimidines and 2-alkyl-4-aryl-tetrahydro-pyrimido-azepines is described. Representative compounds were shown to be subtype selective 5-HT_2A antagonists. Optimal place

Full text of DOI:10.1016/j.bmcl.2008.01.090

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2103–2108
2-Alkyl-4-aryl-pyrimidine fused heterocycles as selective
5-HT2A antagonists
Brock T. Shireman,* Curt A. Dvorak, Dale A. Rudolph, Pascal Bonaventure,
Diane Nepomuceno, Lisa Dvorak, Kirsten L. Miller,
Timothy W. Lovenberg and Nicholas I. Carruthers
Johnson & Johnson Pharmaceutical Research and Development, L.L.C., 3210 Merryfield Row, San Diego, CA 92121, USA
Received 11 December 2007; revised 17 January 2008; accepted 23 January 2008
Available online 30 January 2008
Abstract—The synthesis and SAR for a novel series of 2-alkyl-4-aryl-tetrahydro-pyrido-pyrimidines and 2-alkyl-4-aryl-tetrahydro-
pyrimido-azepines is described. Representative compounds were shown to be subtype selective 5-HT2A antagonists. Optimal place-
ment of a basic nitrogen relative to the pyrimidine and the presence of a 4-fluorophenyl group in the pyrimidine 4-position was
found to have a profound effect on affinity and selectivity.
Ó 2008 Elsevier Ltd. All rights reserved.
Known for almost 50 years, the neurotransmitter sero-
tonin was first isolated and identified in as 5-hydroxy-
tryptamine (5-HT).1,2 Perceptive operational studies
have shown that the monoamine elicits a complex array
of pharmacological and physiological responses by act-
ing at a diversity of 5-HT receptors. Molecular cloning
studies1 confirmed the existence of at least 14 different
subtypes of 5-HT receptors, each encoded by distinct
genes. These receptors have been divided into seven fam-
ilies, designated as 5-HT,1–7 based on pharmacology,
amino acid sequences, gene organization, and second
messenger coupling pathways.3 With the exception of
the 5-HT3 receptor,4 a ligand gated ion-channel, all of
the 5-HT receptor subtypes are coupled to G-proteins.
tor.5 In addition to affinity for presynaptic a2 receptors,
the antidepressant mirtazepine possesses pharmacology
that includes but is not limited to antagonism of the 5-
HT2 receptor subtypes.5 Consequently, the undesirable
side effects of these and other medicines may be a result
of their lack of selectivity. Therefore, the ability to de-
sign selective 5-HT receptor agonists or antagonists rep-
resents an opportunity to discover better tolerated
medicines.
Early pharmacological studies also suggested a role for
5-HT2A antagonists in the treatment of certain sleep dis-
orders.8 In fact, both selective and non-selective 5-HT2A
antagonists have been shown to increase the amount of
time humans spend in slow wave sleep, the most restor-
ative stage of the sleep cycle.9 Shown in Figure 1 are rep-
resentative 5-HT2A antagonists that have been reported.
Introduced in the mid-1980s, ritanserin (1)7–9 is a high
affinity 5-HT2A antagonist that has been shown to in-
crease the amount of time humans spend in slow wave
sleep.7 Eplivanserin (2),10 a highly selective 5-HT2A
antagonist has been shown to improve sleep mainte-
nance, by reducing wake after sleep onset, decreasing
the number of awakenings, increasing the total sleep
time, and improving the quality of sleep. One of the
more highly studied selective 5-HT2A antagonists is
MDL-100907 (3).11 A recent proof of concept study
demonstrated that MDL-100907 increases deep slow
wave sleep in wild-type but not in 5-HT2A knockout
mice.9 Although 3 is selective for the 5-HT2A receptor,
The therapeutic value of many widely prescribed CNS
drugs may be attributed to their action on at least one
of the 5-HT receptor subtypes.5 For example, the 5-
HT2A receptor has been implicated in a variety of behav-
ioral processes and neuropsychiatric disorders.6,7 In
addition, the 5-HT2A receptor appears to be the site of
action of many hallucinogenic compounds. Specifically
LSD, mescaline, and bufotenin are 5-HT2A agonists.
However, as part of their pharmacological profile the
atypical antipsychotics risperdal, olanzapine, and cloza-
pine act as high affinity antagonists of the 5-HT2A recep-
Keywords: Serotonin; 5-HT2A; 5-HT2B; 5-HT2C; 5-HT.
*
Corresponding author. Tel.: +1 858 320 3434; fax: +1 858 450
2089; e-mail: BShirema@prdus.jnj.com
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.01.090

2104
B. T. Shireman et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2103–2108
the differences in hybridization between nitrogen and
carbon. Here we present the chemistry and SAR associ-
ated with analogs of the 2-alkylpyrimidine series (6).
F
F
N
N
N
O
Synthetically, template 6 was attractive due to the con-
vergent manner with which molecules could be con-
structed as shown retrosynthetically in Figure 3. We
envisioned pyrimidine construction utilizing readily
accessible amidines 7 and b-ketoesters 8. The choice of
b-ketoester 8 would be based on the desired size of ring
(m and n) and the positioning of the basic nitrogen with-
in the ring. Then, Suzuki cross-coupling with readily
available boronic acids 9 and an appropriately activated
pyrimidine would provide the desired compounds (6)
after deprotection.
N
N
S
O
F
2, Eplivanserin
1, Ritanserin
O
OH
O
F
S
N
N
OMe
OMe
OMe
OMe
F
F
3, MDL-100907
4
Figure 1. Reported selective 5-HT2A antagonists.
The tetrahydro-pyrido-pyrimidine analogs were derived
from commercially available b-ketoesters 10 or 11 by
condensation with amidines 7 as shown in Scheme 1.14
This provided heterocycles 12 and 13 in 53–70% yield
after refluxing in t-BuOH in the presence of Et3N. Selec-
tive O-triflate formation gave 14 and 15 in 66–91% yield
after treatment with Tf2O and Et3N in CH2Cl2. The pen-
dant aryl ring was then installed with a Suzuki cross-
coupling reaction taking advantage of conditions previ-
ously described by our group for the Suzuki couplings of
pyrazole triflates.15 Treatment of the pyrimidine triflates
with the appropriately substituted aryl boronic acid,
K3PO4, Pd(dppf)Cl2, and dppf in dioxane at 100 °C gave
the desired compounds, 16 and 17, in 68–85% yield.
Deprotection of 16 using either 4 M HCl in dioxane or
TFA in CH2Cl2 gave the desired analogs in good yields.
The benzyl group of 17 was removed to give 19 using
1,4-cyclohexadiene in refluxing EtOH in the presence
of Pd/C.
a dose-dependent prolongation of the QTc interval was
observed in anesthetized dogs.12a However, the structur-
ally related a-fluorosulfone 412 has been reported to ex-
hibit an improved cardiovascular profile.
The objective at the outset of our program was to iden-
tify a selective 5-HT2A antagonist that would warrant
further in vivo experiments. However, the challenge in
the discovery of subtype selective 5-HT2A antagonists
is that the 5-HT2A–2C receptor subtypes exhibit >70%
homology in the transmembrane domain.5 This homol-
ogy has impeded the discovery of subtype selective 5-
HT2A antagonists with favorable pharmacokinetics.
Determination of the therapeutic value of a selective 5-
HT2A antagonist is dependent upon the discovery of
such molecules.
A high-throughput screen of our internal compound
collection identified aminopyrimidine 513 (5-HT2A
Ki = 20 nM) as a high affinity ligand for the 5-HT2A
receptor, Figure 2. However, this compound also
showed significant affinity for the 5-HT2B and 5-HT2C
receptors. As discussed above, for our program we de-
sired a molecule with increased selectivity versus the 5-
HT2B and 5-HT2C receptors. While evaluating this lead,
one of our approaches was to replace the amino group
on the pyrimidine with an alkyl group, represented gen-
erally as 6. This type of substitution was envisioned to
decrease the basicity and overall electronics of the het-
erocyclic ring. The replacement would also change the
spatial relationship of this hydrophobic group due to
Evaluation of an azepine in place of the piperidine ring,
required the synthesis of b-ketoesters 21, 23, and 24,
Scheme 2. Beginning with 4-N-Boc-piperidinone (20),
treatment with ethyl diazoacetate in Et2O at 0 °C gave
the b-ketoester 21 in 77% yield.16 The preparation of
isomeric azepines 23 and 24 required the use of 3-N-
Boc-piperidinone (22). Utilizing identical conditions
for the conversion of 20 to 21, piperidine 22 was ring ex-
R1
N
N
n
R2
m
R1
N
R3
N
N
N
n
6
N
N
R2
m
O
O
n
N
H
OH
B
R1
N
H
OEt
m
+
+
HO
H2N
NH
R2
6, R1=alkyl
N
5
R3
5-HT2A Ki=20 nM
5-HT2B Ki=70 nM
5-HT2C Ki=300 nM
7
8
9
Figure 3. Retrosynthesis of the fused 2-alkyl-4-aryl-pyrimidine
template.
Figure 2. HTS hit.

B. T. Shireman et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2103–2108
2105
R1
R1
O
O
O
Y
O
R1
R1
N
N
N
Y
N
a
a
+
OEt
+
OEt
H2N
NH
OH
X
H2N
NH
OH
X
X
Y
Z
X
Y
Z
7
23, X=NBoc, Y=Z=CH2
21, X=CH2, Y=NBoc, Z=CH2
24, X=Y=CH2, Z=NBoc
25, X=NBoc, Y=Z=CH2
26, X=CH2, Y=NBoc, Z=CH2
27, X=Y=CH2, Z=NBoc
7, R1=alkyl
10, X=CH2, Y=NBoc
11·HCl, X=NBn, Y=CH2
12, X=CH2, Y=NBoc
13, X=NBn, Y=CH2
R1
R1
R1
R1
N
N
N
N
N
Y
N
N
Y
N
c
b
c
b
OTf
OTf
R2
R2
X
X
X
X
Y
Y
Z
Z
28, X=NBoc, Y=Z=CH2
29, X=CH2, Y=NBoc, Z=CH2
30, X=Y=CH2, Z=NBoc
31, X=NBoc, Y=Z=CH2
32, X=CH2, Y=NBoc, Z=CH2
33, X=Y=CH2, Z=NBoc
14, X=CH2, Y=NBoc
15, X=NBn, Y=CH2
16, X=CH2, Y=NBoc
17, X=NBn, Y=CH2
d
e
d
18, X=CH2, Y=NH
19, X=NH, Y=CH2
34, X=NH, Y=Z=CH2
35, X=CH2, Y=NH, Z=CH2
36, X=Y=CH2, Z=NH
Scheme 1. Reagents, conditions, and typical yields: (a) Et3N, t-BuOH,
reflux, 53–70% yield; (b) Tf2O, Et3N, THF, 0 °C to rt, 66–91% yield;
(c) ArB(OH)2, Pd(dppf)Cl2, dppf, K3PO4, dioxane, 100 °C, 68–85%
yield; (d) 4 M HCl in dioxane/EtOAc or TFA/CH2Cl2; (e) 1,4-
cyclohexadiene, Pd/C, EtOH, reflux.
Scheme 3. Reagents, conditions, and typical yields: (a) Et3N, t-BuOH,
reflux, 26–52% yield; (b) Tf2O, Et3N, THF, 0 °C to rt, 73–89% yield;
(c) ArB(OH)2, Pd(dppf)Cl2, dppf, K3PO4, dioxane, 100 °C, 57–91%
yield; (d) 4 M HCl in dioxane/EtOAc or TFA/CH2Cl2.
O
O
O
5-HT2A receptor. However, replacement of the 4-fluoro-
phenyl with a 4-methylphenyl (35d, 5-HT2A Ki =
2.5 nM) restored 5HT2A affinity at the expense of selec-
tivity. The 4-methoxyphenyl (35e, 5-HT2A Ki = 11 nM)
and 4-cyanophenyl (35f, 5-HT2A Ki = 34 nM) substi-
tuted analogs possessed reduced affinity as compared
to 35c and 35d. None of these analogs demonstrated
the desired selectivity profile over the 5-HT2B and 5-
HT2C receptors. Gratifyingly, a 4-fluorobenzyl (35g, 5-
HT2A Ki = 0.7 nM) or benzyl (35h, 5-HT2A Ki = 0.6 nM)
in the pyrimidine 2-position resulted in compounds with
picomolar affinity for the 5-HT2A receptor with im-
proved selectivity over 5-HT2B and 5-HT2C. The 4-meth-
a
OEt
N
BocN
O
Boc
20
21
O
O
O
O
b
OEt
OEt
+
BocN
BocN
NBoc
22
23
24
Scheme 2. Reagents and conditions: (a) ethyl diazoacetate, Et2O, 0 °C,
77% yield; (b) ethyl diazoacetate, Et2O, 0 °C, 1:1 23 (34% yield):24
(32% yield).
ylphenyl
(35i,
5-HT2A
Ki = 0.4 nM,
5-HT2B
Ki = 2.5 nM, 5-HT2C Ki = 15 nM) and 4-trifluorometh-
ylphenyl (35j, 5-HT2A Ki = 0.8 nM, 5-HT2B Ki = 22 nM,
5-HT2C Ki = 38 nM) analogs were also evaluated. As be-
fore, replacement of the fluorine with a methyl group re-
sulted in a decrease in selectivity. Moderate selectivity
was restored with the trifluoromethyl substituted ana-
log, suggesting that an electron withdrawing group in
this position is important for selectivity.
panded to give 23 and 24 as a 1:1 mixture.17 Chromato-
graphic separation gave a 34% yield of 23 and a 32%
yield of 24. Using b-ketoesters 21, 23, and 24, the desired
tetrahydroazepine containing analogs 34–36 were syn-
thesized in a manner analogous to the preparation of
18 and 19 as shown in Scheme 3.
The SAR associated with pyrimidino-tetrahydroaze-
pines 35 is shown in Table 1.18,19 In general, medium
to large hydrophobic groups are required at the 2-posi-
tion (R1) of the pyrimidine for the desired 5-HT2A affin-
ity. For example, a small hydrophobic substituent such
as a CH3 at R1 resulted in poor 5-HT2A affinity (35a,
Ki = 1200 nM). Increasing the size of this group to an
isopropyl and replacing the 4-chlorophenyl with a 4-
fluorophenyl (35b, 5-HT2A Ki = 4.0 nM) provided a
300-fold increase in 5-HT2A affinity relative to 35a.
However this compound showed poor selectivity versus
the remaining 5-HT2 receptors (5-HT2B Ki = 20 nM, 5-
HT2C Ki = 80 nM). A further increase in steric size to
a cyclopentyl (35c, 5-HT2A Ki = 13.0 nM) resulted in a
compound with slightly decreased affinity for the
We were also intrigued by the selectivity profile offered
by piperidine 18a. Therefore, an evaluation of the alkyl
group in the 2-position as it related to piperidine 18 is
also shown in Table 1 with the 4-F-phenyl substituent
held constant. In general, diverse hydrophobic substitu-
ents were tolerated in the 2-position of the pyrimidine in
terms of 5-HT2A affinity. However, the selectivity profile
of these analogs varied significantly. For example, the
cyclopentyl (18b, 5-HT2A Ki = 2.2 nM) and tert-butyl
(18c, 5-HT2A Ki = 5.0 nM) analogs possessed strong
affinity for the 5-HT2A receptor. However, the cyclopen-
tyl (18b, 5-HT2B Ki = 220 nM) analog was slightly more
selective than 18c (5-HT2B Ki = 70 nM) over the 5-HT2B
receptor. The sec-butyl substituted analog (18d, 5-HT2A

2106
B. T. Shireman et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2103–2108
Table 1. 5-HT2 binding data for analogs 35a–j and 18a–i
Compound
Kia (nM)
R1
2
N
N
4
R2
n
N
H
n
R1
R2
Cl
5HT2A
5HT2B
5HT2C
35a
35b
35c
2
1200 ( 420)
270 ( 310)
20 ( 0)
30b
5000 ( 1320)
2
2
F
F
4.0 ( 1.0)
80 ( 20)
13.0 ( 0.3)
200 ( 28)
35d
35e
2
2
CH3
2.5 ( 0.5)
11 ( 4)
1.0 ( 0)
23 ( 3)
OCH3
4.8 ( 1.5)
120 ( 33)
35f
2
2
CN
F
34 ( 7)
20 ( 3)
40 ( 0)
460 ( 39)
50 ( 20)
35g
0.7 ( 0.3)
F
35h
35i
2
2
2
1
1
1
1
1
1
1
1
1
F
0.6 ( 0.2)
0.4 ( 0.1)
0.8 ( 0.5)
20 ( 7)
30 ( 0)
20 ( 0)
Ph
CH3
CF3
F
2.5 ( 0.5)
22 ( 7)
15 ( 2)
Ph
Ph
Ph
35j
38 ( 8)
18a
18b
18c
18d
18e
18f
18g
18h
18i
2500 ( 500)
220 ( 17)
70b
400 ( 160)
110 ( 18)
100 ( 48)
1020 ( 480)
1000 ( 500)
410 ( 75)
5000 ( 1370)
1000 ( 330)
4000 ( 590)
F
2.2 ( 0.6)
5.0 ( 1.2)
11 ( 4)
F
F
2250 ( 1250)
2200 ( 1800)
210 ( 75)
9000b
F
13 ( 8)
F
F
11 ( 1)
F
40 ( 10)
29 ( 5)
F
400b
F
50 ( 7)
1000b
a Values are means of two or three experiments in triplicate unless indicated, SEM is in parentheses.
b Values are single experiments in triplicate.
Ki = 11 nM) was found to be a high affinity ligand for
the 5-HT2A receptor and considerably more selective
for 5-HT2A versus 5-HT2B/2C. The 4-F benzyl (18e, 5-
HT2A = 13 nM) substituted analog possessed similar
affinity for the 5-HT2A receptor and maintained good
selectivity over the 5-HT2B and 5-HT2C receptors. The
cyclobutyl analog (18f, 5-HT2A Ki = 11 nM) maintained
affinity with reduced selectivity. The cyclopropyl (18g, 5-
HT2A Ki = 40 nM), iso-butyl (18h, 5-HT2A Ki = 29 nM),
and phenyl (18i, 5-HT2A Ki = 50.0 nM) substituted ana-
logs all possessed decreased affinity for the 5-HT2A
receptor.
For the azepine and piperidine analogs discussed above,
we determined that a benzyl or 4-fluorobenzyl in the
pyrimidine 2-position provided compounds with an
appropriate balance between selectivity and potency
for 5-HT2A. In addition, a 4-fluorophenyl in the 4-posi-

B. T. Shireman et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2103–2108
2107
tion of the pyrimidine maintained high affinity for 5-
HT2A while improving selectivity versus the remaining
5-HT2 receptors. In order to further evaluate the appro-
priate placement of the basic nitrogen in a piperidine or
azepine ring, we examined the effect that positioning of
this nitrogen had on affinity and selectivity for the 5-
HT2A receptor, Table 2. Positioning the nitrogen in
the 8-position of the 2-alkyl-4-aryl-tetrahydro-pyrimi-
do-azepine (34a, 5-HT2A Ki = 8.8 nM) gave a high affin-
ity compound with excellent selectivity for 5-HT2B
(Ki = 1300 nM). Poor selectivity over the 5-HT2C
(Ki = 150 nM) was observed for this compound. The
corresponding piperidine analog 19a (5-HT2A
Ki = 9 nM) possessed similar affinity for 5-HT2A with re-
duced selectivity for the 5-HT2B (Ki = 310 nM) and 5-
HT2C (Ki = 75 nM) receptors. Azepine analog 36a (5-
HT2A Ki = 130 nM), with the nitrogen in 6-position of
the tetrahydroazepine, showed decreased affinity for
the 5-HT2A receptor.
ing of a basic nitrogen relative to the 2,4-substituted
pyrimidine core. In addition, selectivity was largely af-
fected by the presence of a 4-fluorophenyl moiety in
the pyrimidine 4-position. In summary we have identi-
fied a promising series of potent and selective 5-HT2A
antagonists that merit further in vivo evaluation, details
of which will be reported in due course.
References and notes
1. Hoyer, D.; Hannon, J. P.; Martin, G. R. Pharmacol.
Biochem. Behav. 2002, 71, 533.
2. Hoyer, D.; Martin, G. R. Behav. Brain Res. 1996, 73, 263.
3. Raymond, J. R.; Mukhin, Y. V.; Gelasco, A.; Turner, J.;
Collinsworth, G.; Gettys, T. W.; Grewal, J. S.; Gar-
novskayam, M. N. Pharmacol. Ther. 2001, 92, 179.
4. Thompson, A. J.; Lummis, S. C. R. Curr. Pharm. Des.
2006, 12, 3615.
5. Roth, B. L.; Lopez, E.; Patil, S.; Kroeze, W. K. Neuro-
scientist 2000, 6, 252.
6. For a recent review on the 5HT2A receptor, see: Leyson, J.
E. Curr. Drug Targets CNS Neurol. Disord. 2004, 3, 11.
7. Stefanski, R.; Goldberg, S. R. CNS Drugs 1997, 7, 388.
8. Dugovic, C. J. Sleep. Res. 1992, 1, 163.
In an assay to determine functional activity18,20 a repre-
sentative set of compounds, 35b (pKb = 8.3), 35g
(pKb = 8.2), 18b (pKb = 8.1), and 18e (pKb = 7.1), were
determined to be high affinity antagonists of the 5-
HT2A receptor.
9. Sharpley, A. L.; Elliott, J. M.; Attenburrow, M.-J.;
Cowen, P. J. Neuropharmacology 1994, 33, 467.
10. Rinaldi-Carmona, M.; Congy, C.; Santucci, V.; Simiano,
J.; Gautret, B.; Neliat, G.; Labbeeuw, B.; Le Fur, G.;
Soubrie, P.; Breliere, J. C. J. Pharmacol. Exp. Ther. 1992,
262, 759.
In conclusion, replacement of an aminopyrimidine in an
initial HTS hit with an alkylpyrimidine was found to
provide a novel series of subtype selective 5-HT2A antag-
onists. Utilizing a convergent synthesis, the synthetic
accessibility made it possible to rapidly evaluate this
series. Selectivity and affinity for the 5-HT2A receptor
was found to be highly dependent upon proper position-
11. Kehne, J. H.; Baron, B. M.; Carr, A. A.; Chaney, S. F.;
Elands, J.; Feldman, D. J.; Frank, R. A.; Van Giersber-
gen, P. L. M.; McCloskey, T. C.; Johnson, M. P.;
McCarthy, D. R.; Poirot, M.; Senyah, Y.; Siegel, B. W.;
Widmaier, C. J. Pharmacol. Exp. Ther. 1992, 277, 968.
12. (a) Fish, L. R.; Gilligan, M. T.; Humphries, A. C.;
Ivarsson, M.; Ladduwahetty, T.; Merchant, K. J.; O’Con-
nor, D.; Patel, S.; Philipps, E.; Vargas, H. M.; Hutson, P.
H.; MacLeod, A. M. Bioorg. Med. Chem. Lett. 2005, 15,
3665; (b) Fletcher, S. R.; Burkamp, F.; Blurton, P.; Cheng,
S. K. F.; Clarkson, R.; O’Connor, D.; Spinks, D.; Tudge,
M.; van Niel, M. B.; Patel, S.; Chapman, K.; Marwood,
R.; Shepheard, S.; Bentley, G.; Cook, G. P.; Bristow, L. J.;
Castro, J. L.; Hutson, P. H.; MacLeod, A. M. J. Med.
Chem. 2002, 45, 492.
13. Sanfilippo, P. J.; Urbanski, M.; Williams, L.; Press, J. B.;
Katz, L. B.; Shriver, D. A.; Fernandez, J. A.; Shatynski,
D.; Offord, S. J. Eur. J. Med. Chem. 1992, 27, 655.
14. Representative procedures for Schemes 1 and 3: To a t-
BuOH (0.4 M) solution of the appropriate b-ketoester (1.1
equiv), and amidine hydrochloride (1.0 equiv) was added
Et3N (3.0 equiv). The reaction solution was heated at reflux
for 48 h, cooled to rt, and concentrated. The resulting solid
was dissolved in CH2Cl2 and washed with water. The
aqueous layer was extracted with CH2Cl2. The combined
organic layers were dried and concentrated to give a solid
that was triturated with Et2O to give 12, 13, or 25–27
typically as white solids. To a 0 °C solution of 12, 13, or 25–
27 (1.0 equiv) in CH2Cl2 (0.2 M) was added Et3N (1.1
equiv) followed by trifluoromethanesulfonic anhydride
(1.1 equiv) dropwise over 10 min. After 2 h at 0 °C, the
mixture was diluted with CH2Cl2 and washed with water.
The aqueous layer was extracted with CH2Cl2. The
combined organic layers were dried and concentrated.
The resulting residue was purified via SiO2 chromatogra-
phy (10–30% EtOAc/hexanes) to give 14, 15, or 28–30
typically as clear oils or white solids. To 14, 15, or 28–30
Table 2. 5-HT2 binding data for analogs 34a, 19a, and 36a
Compound Structure
Kia (nM)
5-HT2A 5-HT2B
5-HT2C
Ph
2
N
N
34a
8.8
( 0.7)
1300b
150
( 32.0)
9
4
HN
5
F
Ph
2
N
N
19a
9.0
( 1.0)
310 ( 85) 75
( 25)
8
4
HN
5
F
Ph
2
N
N
36a
9
130
10,000b
5000
4
( 3)
( 3000)
5
F
N
H
a Values are means of two or three experiments in triplicate unless
indicated, SEM is in parentheses.
b Values are single experiments in triplicate.

2108
B. T. Shireman et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2103–2108
(1.0 equiv) was added 4-fluorophenylboronic acid (1.5
19. Receptor binding was performed using the human
recombinant 5-HT2A (GB: X57830), 5-HT2B (GB:
Z36748), and 5-HT2C (GB: M81778) receptors. The
affinity of the described compounds for the three different
human 5-HT2 receptor subtypes was evaluated by com-
petitive radioligand binding assays using [3H]ketanserin
(h5-HT2A) or [3H]mesulergine (h5-HT2B and h5-HT2C).
The assays were performed on membranes prepared from
NIH3T3 stably transfected with h5-HT2A or CHO stably
equiv), K3PO4 (1.5 equiv), Pd(Cl)2dppfÆCH2Cl2 (5.6 mol%)
and dppf (3.6 mol%). The mixture was evacuated with N2,
dioxane (0.1 M) was added and the mixture was heated at
100 °C for 2–15 h. After cooling to room temperature, the
mixture was diluted with Et2O, filtered through a small
SiO2 plug, and the filtrate was concentrated. The resulting
residue was purified via SiO2 chromatography (5–30%
EtOAc/hexanes) to give 16, 17, or 31–33.
15. Dvorak, C. A.; Rudolph, D. A.; Ma, S.; Carruthers, N. I.
J. Org. Chem. 2005, 70, 4188.
16. (a) Moriya, T.; Oki, T.; Yamaguchi, S.; Morosawa, S.;
Yokoo, A. Bull. Chem. Soc. JpN. 1968, 41, 230; (b)
Roglans, A.; Marquet, J.; Moreno-Manas, M. Syn.
Comm. 1992, 22, 1249.
17. Lotte, B.; Krogsgaard-Larsen, P.; Schaumburg, K.;
Johansen, J. S.; Falch, E.; Curtis, D. R. J. Med. Chem.
1986, 29, 224.
18. Bonaventure, P.; Nepomuceno, D.; Miller, K.; Chen, J.;
Kuei, C.; Kamme, F.; Tran, D.; Lovenberg, T. W.; Liu, C.
Eur. J. Pharmacol. 2005, 513, 181.
transfected with h5-HT2B and h5-HT2C.
20. In vitro functional properties on the different 5-HT2
receptor subtypes were determined using fluorometric
imaging plate reader (FLIPR) based calcium assay. The
same cell lines as described18 were used for the FLIPR
experiments. Responses to 5-HT in all three cell lines
were typified by a rapid peak in fluorescence corre-
sponding to an increase in [Ca2+]. Compounds 35b, 35g,
18b, 18e, ritanserin, MDL-100907, eplivanserin, and
risperidone failed to show agonist activity (increase in
[Ca2+]) in any of the cell types (i.e., 5-HT2A, 5-HT2B
and 5-HT2C).
,