Novel potent 5-HT1F receptor agonists: Structure-activity studies of a series of substituted N-[3-(1-methyl-4-piperidinyl)-1H-pyrrolo[3,2-b]pyridin-5-yl]amides

Article abstract of DOI:10.1021/jm030020m

Compound 1a (LY334370), a selective 5-HT_1F receptor agonist (SSOFRA), inhibited dural inflammation in the neurogenic plasma protein extravasation model of migraine and demonstrated clinical efficacy for the acute treatment of migraine. Although

Full text of DOI:10.1021/jm030020m

3060
J . Med. Chem. 2003, 46, 3060-3071
Novel P oten t 5-HT1F Recep tor Agon ists: Str u ctu r e-Activity Stu d ies of a Ser ies
of Su bstitu ted
N-[3-(1-Meth yl-4-p ip er id in yl)-1H-p yr r olo[3,2-b]p yr id in -5-yl]a m id es§
Sandra A. Filla,*,†,# Brian M. Mathes,† Kirk W. J ohnson,† Lee A. Phebus,† Marlene L. Cohen,† David L. Nelson,†
J ohn M. Zgombick,‡,| J on A. Erickson,† Kathryn W. Schenck,† David B. Wainscott,† Theresa A. Branchek,‡ and
J ohn M. Schaus†
Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana 46285, and Synaptic Pharmaceutical Corporation,
215 College Road, Paramus, New J ersey 07652
Received J anuary 14, 2003
Compound 1a (LY334370), a selective 5-HT1F receptor agonist (SSOFRA), inhibited dural
inflammation in the neurogenic plasma protein extravasation model of migraine and
demonstrated clinical efficacy for the acute treatment of migraine. Although 1a was greater
than 100-fold selective over both the 5-HT1B and 5-HT1D receptors, it exhibited appreciable
5-HT1A receptor affinity. Described here is the synthesis and evaluation of a series of pyrrolo-
[2,3-c]pyridine and pyrrolo[3,2-b]pyridine (2a and 3a ) as well as pyrrolo[3,2-d]pyrimidine (4a )
analogues of 1a , compounds prepared in an effort to identify SSOFRAs with improved selectivity
over other 5-HT1 receptor subtypes. The pyrrolo[3,2-b]pyridine analogue 3a showed high 5-HT1F
receptor affinity but offered no improvement in selectivity compared to 1a . However, the C-5
acetamide derivative, 3b, was greater than 100-fold selective over the 5-HT1A, 5-HT1B, and
5-HT1D receptors. SAR studies of this series determined that alkylamides in particular exhibited
high selectivity for the 5-HT1F receptor. Replacement at C-5 with other substituents decreased
affinity or selectivity. These SAR studies identified SSOFRAs that demonstrated oral activity
in the neurogenic plasma protein extravasation model, a model indicative of antimigraine
activity.
In tr od u ction
extravasation in the dura, causing inflammation and
pain. Inhibition of neuropeptide release and plasma
protein extravasation in the dural membrane blocks
neurogenic inflammation and central pain transmission,
resulting in alleviation of migraine pain.6 In fact, the
triptans were all effective in blocking dural plasma
protein extravasation (PPE) following electrical stimu-
lation of the trigeminal ganglion in guinea pigs,7 a model
that, although controversial,8 is believed to be predictive
of clinical efficacy for migraine treatment.9 This effect
of the triptans is likely mediated through activation of
the 5-HT1F receptor. Phebus et al. reported a distinct
correlation between a compound’s potency in the PPE
model and its affinity for the 5-HT1F receptor but not
other 5-HT1 receptors.9a In addition, 5-HT1F receptor
agonists did not contract the rabbit saphenous vein,10
an effect that correlated well with contractile responses
in the human coronary artery.11 Recent reports dem-
onstrated that the 5-HT1B receptor mediated vascular
contractile responses to triptans12 and supported the
contention that it also mediated vascular responses in
the rabbit saphenous vein,11 suggesting that this recep-
tor subtype was responsible for the triptans’ vasocon-
strictive properties. On the basis of these studies, we
proposed that a compound selective for the 5-HT1F
receptor may provide an effective treatment for mi-
graine without cardiovascular side effects. We have
recently reported on the first of these SSOFRAs (selec-
tive serotonin one F receptor agonists).13 Compound 1a
(LY334370) exhibited high affinity for the 5-HT1F recep-
tor as well as high selectivity over the 5-HT1B and
Sumatriptan, a nonselective 5-HT1B/1D agonist, is an
effective therapeutic for the acute treatment of mi-
graine.1 The vasodilatory hypothesis of migraine sug-
gests that the dilatation of cranial blood vessels is a
primary cause of pain during migraine, and Sumatrip-
tan’s efficacy has been attributed to its ability to
constrict cranial blood vessels.2 Sumatriptan is also a
potent vasoconstrictor of human arteries,3 as are other
5-HT1B/1D agonists that have recently been developed.
It and other agents within this triptan class of anti-
migraine compounds exhibit some effects that include
dizziness, tingling, and chest tightness/pressure.1 In
addition, there have been reported incidences of coro-
nary vasoconstriction, and the use of drugs such as
Sumatriptan is contraindicated in patients with known
heart disease.4
The mechanism of action of 5-HT1 agonists in acute
migraine treatment has been widely debated,5 and the
ability of these compounds to block dural inflammation
mediated by trigeminal afferents might also account for
their antimigraine activity. Stimulation of the trigemi-
nal nerve results in vasodilatation and plasma protein
§ Dedicated to our talented and energetic colleague, Dr. J ohn
Zgombick, 1958-2002.
* To whom correspondence should be addressed. Phone: 781-839-
4732. Fax: 781-839-4500. E-mail: sandra.filla@astrazeneca.com.
† Eli Lilly and Company.
#
Present address: AstraZeneca R&D Boston, 35 Gatehouse Drive,
Waltham, MA 02451.
‡ Synaptic Pharmaceutical Corporation.
| Deceased November 8, 2002.
10.1021/jm030020m CCC: $25.00 © 2003 American Chemical Society
Published on Web 06/03/2003

5-HT1F Receptor Agonists
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 3061
Ta ble 1. Comparison of in Vitro Binding Affinities of
5-Acylaminopyrrolopyridine/Pyrrolopyrimidine Analogues vs
5-Acylaminoindole Analogues
Ch a r t 1
binding affinitya Ki (nM)
compd
R
X
Y
5-HT1F 5-HT1A
5-HT1B
5-HT1D
1a
2a
3a
4a
1b
2b
3b
4b
4-F-phenyl
4-F-phenyl
4-F-phenyl
4-F-phenyl
methyl
methyl
methyl
methyl
C
N
C
N
C
N
C
N
C
C
N
N
C
C
N
N
2.1(0.6) 22(4.4)
13(1.2) 17(3.1)
7.6(1.0) 56(5.1)
240(33)
80(6.4)
1200(150) 730(9)
430(30)
53(6.4)
67(6.5) 200(31) 6000(540) 1900(270)
9.7(2.3) 68(8.0)
390(18)
150(4.0)
39(10)
250(38)
210(48) 2.0(1.0) NDb
5.0(0.5) 620(37) 270(47)
220(37) 780(100) 2900(550) 470(65)
a
Affinities for receptors were determined in vitro by radioligand
binding assays using cell lines expressing the appropriate human
serotonin receptor.19-21 Each value is the mean of at least three
Sch em e 1a
b
determinations. SEM is shown in parentheses. ND ) Ki not
determined. <50% displacement of radioligand at 10-7 M test
ligand.
5-HT1D receptors, although it demonstrated appreciable
affinity for the 5-HT1A receptor (Table 1).14 It did not
contract the rabbit saphenous vein, yet was effective in
the PPE model of migraine.9a,14b Compound 1a also
significantly inhibited activation of second-order neu-
rons in the trigeminal nucleus caudalus produced by
electrical stimulation of the dura mater in rats, another
well-accepted model of migraine.15 In addition, 1a
demonstrated clinical efficacy as a treatment for mi-
graine without the cardiovascular side effects observed
with nonselective 5-HT1 receptor agonists.16
Our subsequent focus in this program turned to
identifying potential SSOFRAs with improved 5-HT
receptor selectivity, particularly against the 5-HT1A
receptor. As a continuation of our structure-activity
relationship (SAR) studies, we proposed to survey
replacement of the indole nucleus of 1a with various
pyrrolopyridine nuclei. Pyrrolopyridines retain the geo-
metric and conformational attributes of indole but have
different physicochemical properties, which may alter
their biological activity. In addition to an additional
heteroatom that itself may participate in a hydrogen-
bonding interaction with the receptor, we anticipated
that replacement of a carbon atom with nitrogen
adjacent to the C-5 amide bond might impact the pKa
of the amide proton as well as potentially affect the
conformation of this amide group (vide infra). We
prepared 4-fluoro-N-[3-(1-methyl-4-piperidinyl)-1H-pyr-
rolo[2,3-c]pyridin-5-yl]benzamide (2a ), 4-fluoro-N-[3-(1-
methyl-4-piperidinyl)-1H-pyrrolo[3,2-b]pyridin-5-yl]benz-
amide (3a ), and 4-fluoro-N-[3-(1-methyl-4-piperidinyl)-
1H-pyrrolo[3,2-d]pyrimidin-5-yl]benzamide (4a ) and
determined the 5-HT receptor binding affinities of these
analogues (Chart 1). From these studies, we identified
a series of novel and highly selective 5-HT1F receptor
agonists. We report here the synthesis of these com-
pounds and their pharmacologic profiles and describe
the SAR studies that led to the identification of potent
and selective 5-HT1F receptor agonists with potential
for acute treatment of migraine.
a
Reagents: (a) DMFDMA, DMF, 110 °C; (b) 10% Pd/C, H2 (40
psi), EtOH, room temp; (c) Na0, MeOH, 1-methyl-4-piperidone, 75
°C; (d) 10% Pd/C, H2 (60 psi), EtOH, 40 °C; (e) RCOCl, pyridine,
room temp.
Ch em istr y
Each of the three 5-substituted pyrrolopyridine and
pyrrolopyrimidine nuclei (7, 11, and 24) was prepared
through a modified Batcho-Leimgruber indole synthe-
sis. The preparation of the pyrrolo[2,3-c]pyridine deriva-
tives is illustrated in Scheme 1. Treatment of 2-amino-
4-methyl-5-nitropyridine 5 with N,N-dimethylformamide
dimethyl acetal (DMFDMA) in DMF provided eneamine
6. Reductive cyclization (H2, Pd/C) yielded the C-5
formamidine pyrrolo[2,3-c]pyridine nucleus 7. Conden-
sation of 7 with 1-methyl-4-piperidone in sodium meth-
oxide/methanol introduced the 1,2,5,6-tetrahydropyridyl
moiety at C-3 with concomitant hydrolysis of the C-5
amidine group, providing amine 8. The desired 5-acyl-
aminopyrrolo[2,3-c]pyridine analogues 2a ,b were sub-
sequently prepared from 8 by hydrogenation of the
olefin to 9 followed by acylation with the requisite acid
chloride in pyridine.

3062 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Filla et al.
Sch em e 2a
Sch em e 3a
a
Reagents: (a) [(CH3)2N]3CH, DMF, 100 °C; (b) 5% Pd/C, H2
(60 psi), THF/EtOH, room temp; (c) POCl3, 115 °C; (d) 10% Pd/C,
H
2 (40 psi), NaHCO3, EtOH, room temp; (e) Na0, MeOH, 1-methyl-
4-piperidone, 75 °C; (f) 10% Pd/C, H2 (60 psi), EtOH, room temp;
(g) RCOCl, pyridine, 50 °C.
oxopyrimidine 20,18 since attempts to nitrate 2-amino-
6-methylpyrimidine directly proved to be unsuccessful.
This strategy necessitated removal of the undesired
carbonyl functionality at some stage of the synthesis.
Treatment of 20 with DMFDMA or DMFDEA resulted
in formation of the desired C-6 eneamine but with
appreciable amounts of N-3 alkylation. Use of tris-
(dimethylamino)methane, however, circumvented this
problem, and compound 21 was isolated as the sole
product. Reductive cyclization provided the 2-formami-
dino-9-deazaguanine intermediate 22. The C-6 oxo
group was removed in a two-step sequence. Halogena-
tion with phosphorus oxychloride yielded 23 without
loss of the C-2 amidine-protecting group. Hydrogenolysis
then gave the pyrrolo[3,2-d]pyrimidine intermediate 24.
This compound was converted to the desired analogues
4,b using the three-step sequence described previously.
In this case, selective acylation of the C-2 amine of 25
proved to be difficult because of the electron-withdraw-
ing character of the adjacent nitrogen atoms. Under
optimized conditions, treatment of 25 with the requisite
acid chloride in pyridine at 50 °C provided 4a ,b in
modest yields, accompanied by some diacylated product
as well as unreacted starting amine.
a
Reagents: (a) DMFDMA, DMF, 110 °C; (b) 10% Pd/C, H2 (40
psi), EtOH, room temp; (c) KOH, MeOH, 1-R-4-piperidone, 75 °C;
(d) R′COCl, pyridine, 55 °C; (e) 10% Pd/C, H2 (40 psi), THF/EtOH,
room temp; (f) R′SO2Cl, pyridine, 55 °C; (g) R′NCO, THF/DMF,
room temp; (h) TFA, CH2Cl2, room temp.
The 5-substituted pyrrolo[3,2-b]pyridine nucleus 11
was synthesized in an analogous fashion starting with
2-amino-6-methyl-5-nitropyridine 10 (Scheme 2).17 Con-
densation of 11 with an N-alkyl-4-piperidone or with
1-(tert-butoxycarbonyl)-4-piperidone under basic condi-
tions provided 1,2,5,6-tetrahydropyridyl intermediates
12a -g, which were either acylated at the C-5 amine to
prepare derivatives 13a ,b or hydrogenated to yield the
corresponding N-substituted piperidinyl C-5 amines
14a -g. Reaction of the C-5 amine of 14a (R ) CH3) with
an acid chloride, sulfonyl chloride, or isocyanate pro-
vided amides 3a -ff, sulfonamides 15a ,b, and ureas
16a -c. Likewise, acylation of the other N-substituted
piperidinyl C-5 amines 14b-g using the same condi-
tions provided analogues 17a -e and 18a -c (Scheme
2). The piperidine-BOC protecting group of compounds
18a -c was subsequently removed with TFA to generate
19a -c.
We replaced the C-5 amide with representative non-
hydrogen-bond-donating groups to evaluate what role
the N-H functionality at this position played in 5-HT1F
receptor binding. The compounds of interest were
prepared from common intermediate 5-methoxypyrrolo-
[3,2-b]pyridine 27, which was prepared in a manner
analogous to the synthesis of 11. Compound 27 was then
subjected to standard conditions to provide 28. Hydroly-
sis of the methyl ether with HBr/HOAc gave 29, which
was converted to sulfonate ester derivatives 30a ,b
(Scheme 4).
The preparation of the pyrrolo[3,2-d]pyrimidine ana-
logues 4a ,b is described in Scheme 3. In this case, we
opted to start with 2-amino-6-methyl-5-nitro-4(3H)-

5-HT1F Receptor Agonists
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 3063
Sch em e 4a
Ch a r t 2
by comparing the C-5 acetyl analogues (2b, 3b, and 4b)
with the corresponding indole derivative 1b (Table 1).
Again, we observed that the pyrrolo[3,2-b]pyridine
analogue 3b retained high affinity for the 5-HT1F
receptor. Furthermore, this analogue possessed im-
proved selectivity at the 5-HT1A receptor versus 1b and
in fact was more selective than 1a , indicating that this
new series may exhibit an improved pharmacologic
profile over the indole series.
a
We sought to rationalize the effect of the nitrogen
atom in the heterocyclic nuclei of 2a , 3a , and 4a on the
compounds’ 5-HT receptor affinities. Assuming that
each analogue bound to the receptor in a similar fashion
with respect to the basic piperidine nitrogen and the
heterocyclic moiety, the binding affinity differences
could be ascribed to direct van der Waal or hydrogen-
bonding interactions of the pyrrolopyridine or pyrrol-
opyrimidine nucleus with the receptor. This explanation
was not satisfactory, however, in light of the data at
hand. In both acylamino series tested, the pyrrolo[3,2-
b]pyridine analogues (3a and 3b) exhibited higher
affinity at the 5-HT1F receptor than the pyrrolo[2,3-c]-
pyridine derivatives (2a and 2b). While an added
favorable interaction between the nitrogen in the 4-po-
sition of the heterobicycle (Y ) N) and the receptor in
3a ,b could explain the increased affinity, it would not
explain the high affinity observed with the indole
analogues (1a and 1b). Consequently we proposed that
rather than directly influencing receptor binding through
electronic effects, perhaps the ring nitrogen indirectly
affected binding affinity by changing the conformational
preference of the C-5 amide bond. We examined this
possibility by comparing the rotational barriers of each
of the 4-fluorobenzamide analogues. For 1a -4a , the ring
nitrogen (i.e., torsion Y-C-N-C) bond was driven from
0° to 360° at 10° increments using Macromodel (Chart
2).24 The MM3* force field was utilized, and water
solvation effects were modeled using the GB/SA method.25
The resulting relative energy versus torsion angle plots
are shown in Figure 1. Minima existed close to 0° (20°
and 340° for conformer A) and 180° (160° and 200° for
conformer B) for the highest affinity 5-HT1F receptor
analogue 1a , which was consistent with resonance
stabilization of the amide unit by the indole aromatic
system. On the other hand, the two pyrrolopyridine
isomers, 2a and 3a , had significant energy differences
(∼3.5 kcal mol-1) between their 0° and 180° conforma-
tions. The more active isomer 3a favored the 180°
conformation (conformer B), while the less active isomer
2a favored the 0° conformation (conformer A), presum-
ably because of dipole minimization or lone pair repul-
sion between the amide carbonyl and the pyrrolopyri-
Reagents: (a) DMFDMA, Et3N, DMF, 120 °C; (b) 10% Pd/C,
H2 (40 psi), EtOH, room temp; (c) KOH, MeOH, 1-methyl-4-
piperidone, 75 °C; (d) 10% Pd/C, H2 (40 psi), THF/EtOH, room
temp; (e) HBr/HOAc, 105 °C; (f) (RSO2)2O or RSO2Cl, pyridine,
60 °C.
P h a r m a cology
Affinities of compounds for receptors, expressed in Ki
(nM), were determined in vitro by radioligand binding
assays using cell lines expressing the human 5-HT1A
,
5-HT1B, 5-HT1D, and 5-HT1F receptors.19-21 Intrinsic
efficacy at the 5-HT1F receptor was evaluated in a cell-
based measure of receptor activation via stimulation of
[35S]GTP-γ-S binding.22 The vascular contractile effects
of 5-HT1F agonists with high affinity and selectivity for
the 5-HT1F receptor were assessed in vitro using ring
preparations of rabbit saphenous vein.10 Finally, com-
pounds with minimal in vitro vascular contractile activ-
ity were tested in vivo in the guinea pig PPE model of
migraine.7,23 Unilateral stimulation of the trigeminal
ganglia caused the release of inflammatory neuropep-
tides such as CGRP and substance P from the trigemi-
nal nerve terminals, resulting in the leakage (extra-
vasation) of plasma proteins labeled with Evan’s blue
fluorescent dye.23 The inhibition of this extravasation
was quantified by measurement of the ratio of fluores-
cence from the stimulated versus the unstimulated side
of the dural tissue. Data are reported as a ratio of
plasma protein extravasation on the stimulated versus
unstimulated side.
Resu lts a n d Discu ssion
Eva lu a tion of P yr r olop yr id in e a n d P yr r olop y-
r im id in e An a logu es of 1a . Compounds 2a , 3a , and
4a, the 4-fluorobenzamide analogues of 1a, were screened
at a number of 5-HT1 receptor subtypes to compare their
affinity and selectivity for the 5-HT1F receptor (Table
1). The pyrrolo[2,3-c]pyridine and the pyrrolo[3,2-d]-
pyrimidine analogues (2a and 4a ) suffered diminished
affinity and selectivity. On the other hand, the pyrrolo-
[3,2-b]pyridine analogue 3a showed a slight decrease
in 5-HT1F receptor affinity compared with 1a but
retained selectivity. We further explored these nuclei

3064 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Filla et al.
Ta ble 2. In Vitro Binding Affinities of Substituted
N-[3-(1-Methyl-4-piperidinyl)-1H-pyrrolo[3,2-b]pyridin-
5-yl]amides: Various C-5 Substituents
binding affinitya Ki (nM)
compd
R
5-HT1F
5-HT1B
5-HT1D
14a
28
29
30a
30b
NH2
OCH3
O
OSO2CH3
OSO2C6H4-4-F
100(8.6)
23(4.5)
100(13)
4.7(1.6)
7.5(0.2)
800(30)
110(24)
NDb
420(75)
23(5.3)
NDb
8.0(1.1)
30(2.0)
25(2.2)
NDb
a
Affinities for receptors were determined in vitro by radioligand
binding assays using cell lines expressing the appropriate human
serotonin receptor.19-21 Each value is the mean of at least three
b
determinations. SEM is shown in parentheses. ND ) Ki not
determined. <50% displacement of radioligand at 10-7 M test
ligand.
perhaps its increased basicity accounted for the com-
pound’s diminished activity. Sulfonate esters 30a ,b
exhibited high 5-HT1F receptor affinity, but their affini-
ties for the 5-HT1B and 5-HT1D receptors were also high.
This suggested that perhaps hydrogen bond donation
might not be necessary for 5-HT1F receptor activity but
was advantageous for selectivity over other 5-HT1
receptor subtypes. We concluded that retaining the
hydrogen bond donating/accepting character of the C-5
substituent was critical. Thus, we initiated SAR studies
at the C-5 amine of 3-(1-methyl-4-piperidinyl)-1H-
pyrrolo[3,2-b]pyridin-5-ylamine (14a ), preparing mul-
tiple C-5 amide, sulfonamide, and urea derivatives from
this intermediate.
F igu r e 1. MM3*(GB/SA) relative energy versus torsion angle
for 4-fluorobenzamidoindole analogues 1a -4a .
dine nitrogen atom. Furthermore, analyses of available
crystal structures (Cambridge Structural Database26)
of structurally related 2-acylaminopyridines were con-
sistent with our calculations. Approximately 90% of the
over 270 compounds evaluated adopted the conformer
represented by conformer B in Chart 2 (where X ) C;
Y ) N). This difference in conformational preference
between 2a and 3a may point to conformer B as
important for 5-HT1F receptor binding, possibly direct-
ing either the amide carbonyl oxygen or the nitrogen
proton to a favorable interaction with the receptor.
Conformer A may be relevant for high binding affinity
at the other 5-HT1 receptors, accounting for the de-
creased 5-HT1F selectivity of analogues 1a ,b and 2a ,b.
As a result of the in vitro binding analysis of the
pyrrolopyridine and pyrrolopyrimidine analogues, we
decided to investigate the pyrrolo[3,2-b]pyridine series
in an effort to identify potent and selective 5-HT1F
receptor agonists. We focused primarily on varying
amine-substituted groups at C-5. In addition, we also
evaluated substituent effects on N-1 of the 4-piperidine
moiety at C-3.
SAR Su m m a r y of t h e P yr r olo[3,2-b]p yr id in e
Ser ies. In Vitr o Bin d in g a n d F u n ction a l Activities.
On the basis of SAR studies that identified 1a , we
determined that C-5 substituents bearing a hydrogen
bond donor and acceptor were optimal. This trend
carried over to the pyrrolo[3,2-b]pyridine series, since
compounds devoid of this functionality at C-5 showed
greatly reduced 5-HT1F receptor affinity or selectivity
(Table 2). Amine substitution at C-5 (14a ) reduced
5-HT1F receptor affinity compared with the C-5 amide
analogues. While this functional group could participate
as a hydrogen bond donor, it lacked the corresponding
hydrogen bond accepting carbonyl unit. Alternatively,
In light of the reasonable affinity and selectivity of
3a , we focused initially on substituted benzamides
(Table 3). Small substituents at the meta or para
position were generally preferred. In all cases, the ortho-
substituted benzamides possessed diminished affinity
for the 5-HT1F receptor. Compound 3a , the pyrrolo[3,2-
b]pyridine analogue of 1a , was selective for the 5-HT1F
receptor over the 5-HT1B and 5-HT1D receptors (160- and
96-fold, respectively) and was a full agonist (EC50 ) 4.1
nM, Emax ) 96% maximal response to 5-HT) in the
5-HT1F [35S]GTP-γ-S assay. Other 4-substituted (3m ,n )
and the unsubstituted (3c) benzamides also had high
5-HT1F receptor affinity and were full agonists (data not
shown); however they possessed high 5-HT1A receptor
affinity. In fact, appreciable affinity at the 5-HT1A
receptor accompanied this entire series of compounds.
For example, the 2,4-dichloro derivative 3p had high
affinity for the 5-HT1F receptor but was nonselective
versus the 5-HT1A receptor. In the best case, we ob-
served a 5-HT1A/5-HT1F receptor affinity selectivity ratio
of 17 with 3k ; however, this compound was only a
partial 5-HT1F receptor agonist (EC50 ) 22 nM, Emax
)
60% maximal response to 5-HT). On the other hand,
most compounds within this series were quite selective
for the 5-HT1F receptor over the 5-HT1B receptor, the
receptor that concerned us most regarding potential
cardiovascular liabilities.27

5-HT1F Receptor Agonists
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 3065
Ta ble 3. In Vitro Binding Affinities of Substituted
N-[3-(1-Methyl-4-piperidinyl)-1H-pyrrolo[3,2-b]pyridin-5-
yl]amides: Substituted Benzamides
Representative alternative hydrogen bond donating/
accepting substituents at the C-5 amino position, specif-
ically sulfonamides (15a ,b) and ureas (16a -c), were
investigated (Table 5). The slightly more acidic sulfona-
mide derivatives exhibited reduced 5-HT1F receptor
affinity relative to compounds in the amide series. On
the other hand, the urea compounds showed high
5-HT1F receptor affinity but had high 5-HT1B and
5-HT1D receptor affinities as well. As a result of this
study, we concluded that C-5 amide substituents in the
pyrrolo[3,2-b]pyridine series were preferred, since these
analogues exhibited superior 5-HT1F receptor affinity
and selectivity compared to compounds with other C-5
substituents.
binding affinitya Ki (nM)
compd
R
5-HT1F
5-HT1A
5-HT1B
5-HT1D
3a
3c
3d
3e
3f
3g
3h
3i
4-F
H
2-F
7.6(1.0) 56(5.1)
7.3(0.6) 72(15)
1200(150) 730(9)
A focused SAR study evaluated whether the N-meth-
yl-4-piperidinyl moiety at C-3 was optimal in the pyr-
rolo[3,2-b]pyridine series (Table 6). We evaluated the
effect of varying the N-1 alkyl substituent on the
4-piperidinyl moiety in the C-5 4-fluorobenzamide se-
ries. Increasing the chain length at N-1 decreased
5-HT1F receptor affinity (17a -e), as did chain branching
(17d ). The unsubstituted piperidinyl (19a ) as well as
the N-ethyl-1,2,3,6-tetrahyropyridyl (13a ) analogues
showed reasonable affinity for the 5-HT1F receptor, and
we evaluated some C-5 alkyl amide derivatives in each
case in an attempt to optimize these two series (13b,
19b,c). We were particularly interested in investigating
the latter platform, given the dramatic difference in
binding affinity between piperidinyl compound 17a and
its tetrahydropyridyl analogue 13a . None of the com-
pounds prepared showed increased 5-HT1F receptor
affinity, suggesting that the methyl group was optimal
at N-1 of the 4-piperidinyl moiety at C-3.
820(100)
NDb
1000(180)
19(2.5)
19(3.5)
20(3.7)
12(1.7)
11(2.0)
42(23)
16(2.3)
42(7.0)
64(17)
20(3.5)
106(11)
NDb
2-NO2
2-OCH3
2-CH3
3-F
3-NO2
3-OCH3
3-CH3
3-CN
NDb
170(21)
750(66)
150(57)
NDb
440(100)
135(4.8) 1600(370) 690(91)
130(23)
98(16)
390(61)
330(39)
3j
3k
3l
NDb
NDb
5.8(0.5) 100(16)
1200(200) 490(150)
11(0.9)
100(3.0) 590(170)
NDb
3m
3n
3o
3p
4-CN
4-NO2
4-CH3
7.2(1.6) 31(5.0)
7.7(1.7) 38(4.8)
310(24)
270(24)
NDb
380(100)
340(42)
NDb
12(2.2)
30(3.2)
2,4-di-Cl 4.1(0.9) 5.1(0.9)
NDb
83(22)
a
Affinities for receptors were determined in vitro by radioligand
binding assays using cell lines expressing the appropriate human
serotonin receptor.19-21 Each value is the mean of at least three
b
determinations. SEM is shown in parentheses. ND ) Ki not
determined. <50% displacement of radioligand at 10-7 M test
ligand.
Table 4 summarizes the 5-HT1F receptor affinity of
heteroaromatic and alkyl amides at C-5. Isosteric he-
teoroaryl amide groups such as thienyl and furyl amides
proved to be more selective for the 5-HT1F receptor than
the benzamide derivatives, particularly regarding their
5-HT1A receptor affinities. Both the 3-thienyl (3q) and
the 2-thienyl (3r ) derivatives were >35-fold more selec-
SAR Su m m a r y of t h e P yr r olo[3,2-b]p yr id in e
Ser ies. Effica cy in th e Ra bbit Sa p h en ou s Vein a n d
Migr a in e Mod els. We evaluated compounds that were
potent full agonists at the 5-HT1F receptor and exhibited
greater than 30-fold selectivity over other 5-HT1 recep-
tor subtypes for vasocontractility in the rabbit saphe-
nous vein and then further for in vivo efficacy in the
neurogenic PPE migraine model. Contractile response
of the rabbit saphenous vein to 5-HT receptor agonists
correlated with serotonin agonist induced contractility
of human coronary arteries.27 Sumatriptan, as well as
the other 5-HT1B/5-HT1D receptor agonists, contracted
the rabbit saphenous vein, an effect potentially medi-
ated by the 5-HT1B receptor,11 whereas the SSOFRAs
did not.10,13 To identify compounds that would not have
the cardiovascular liabilities associated with the trip-
tans, we used contractile response in the rabbit saphe-
nous vein as a predictor of vasoconstrictive activity. In
general, these 5-HT1F receptor agonists did not mark-
edly contract the rabbit saphenous vein; however, as a
means to prioritize compounds, we characterized com-
pounds as vasoactive if they contracted the rabbit
saphenous vein at >10% maximal contraction relative
to KCl-induced (67 mM) contraction. Of the benzamides
evaluated (3a , 3c, 3m ), only the 4-fluorobenzamide 3a
contracted the rabbit saphenous vein (Table 7). Two of
the alkyl derivatives, 3x and 3d d , were also vasoactive,
whereas alkylamides 3b, 3w , and 3cc showed minimal
contractile activity.
tive for the 5-HT1F receptor over the 5-HT1A, 5-HT1B
,
and 5-HT1D receptors. Interestingly, 3r was only a
partial agonist in the 5-HT1F receptor functional [35S]-
GTP-γ-S assay, whereas 3q showed better intrinsic
efficacy. The furyl derivatives 3s and 3t were highly
selective and were also full 5-HT1F agonists in the [35S]-
GTP-γ-S functional assay. Two pyridyl derivatives (3u
and 3v) were also prepared. These compounds showed
activity similar to that of the benzamides, possessing
only about a 4-fold difference in 5-HT1A versus 5-HT1F
receptor affinity. Given the high affinity and selectivity
of 3b (Table 1), we also investigated a series of simple
alkylamides (Table 4). Small alkyl and cycloalkyl groups
were tolerated and showed very little affinity for other
5-HT receptors, including the 5-HT1A receptor. As the
chain length or ring size was increased, however, 5-HT1F
receptor affinity correspondingly decreased. Chain
branching (3z) also diminished 5-HT1F receptor affinity.
The acetamide (3b) and propionamide (3w ) were the
most selective compounds identified in this series and
were full agonists at the 5-HT1F receptor. Three other
compounds from this series, 3x, 3cc, and 3d d , also
possessed high 5-HT1F receptor affinity and exhibited
reasonable 5-HT1F receptor intrinsic efficacy in the GTP-
γ-S assay.
Compounds devoid of contractile activity were tested
for in vivo efficacy in the guinea pig neurogenic PPE

3066 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Filla et al.
Ta ble 4. In Vitro Binding Affinities (Ki, nM) and in Vitro Functional Activity of Substituted
N-[3-(1-Methyl-4-piperidinyl)-1H-pyrrolo[3,2-b]pyridin-5-yl]amides: Heteroaromatic and Alkyl Amides
binding affinitya Ki (nM)
b
compd
R
5-HT1F
5-HT1A
5-HT1B
5-HT1D
EC50 (nM)
Emax c (%)
3q
3r
3s
3t
3u
3v
3b
3w
3x
3-thienyl
2-thienyl
3-furyl
7.4(1.9)
7.1(1.5)
4.6(0.3)
13(5)
6.3(1.5)
8.4(2.1)
5.0(0.5)
5.5(0.6)
6.9(0.5)
14(5.0)
62(20)
13(3.0)
20(5.2)
9.8(1.4)
8.7(1.6)
11(2.7)
20(5.0)
400(53)
350(65)
840(130)
560(20)
41(3.9)
40(1.7)
620(37)
1000(270)
NDd
1000(120)
390(19)
930(300)
1100(64)
1000(480)
330(24)
270(47)
720(220)
NDd
810(76)
650(130)
1100(250)
930(16)
820(55)
420(42)
250(38)
730(55)
NDd
9.4(0.5)
15(1.1)
11(2.2)
11(1.5)
25(3.2)
11(2.4)
36(1.5)
32(3.8)
13(0.5)
84(0.7)
61(2.3)
88(2.4)
82(2.3)
63(1.6)
97(1.3)
98(1.8)
88(1.1)
94(0.2)
2-furyl
2-pyridyl
4-pyridyl
methyl
ethyl
n-propyl
n-butyl
tert-butyl
n-pentyl
n-heptyl
cyclopropyl
cyclobutyl
cyclopentyl
cyclohexyl
3y
3z
NDd
NDd
NDd
NDd
NDd
NDd
3a a
3bb
3cc
3d d
3ee
3ff
NDd
NDd
NDd
NDd
32(12)
110(28)
1800(270)
900(150)
740(86)
NDd
1400(520)
2000(290)
2000(400)
660(71)
1700(150)
380(52)
640(110)
NDd
34(5.7)
21(3.5)
77(0.7)
81(1.0)
a
Affinities for receptors were determined in vitro by radioligand binding assays using cell lines expressing the appropriate human
serotonin receptor.19-21 Each value is the mean of at least three determinations. SEM is shown in parentheses. EC50 (nM) for stimulation
of [35S]GTP-γ-S binding in mouse LM(tk-) cells expressing the human 5-HT1F receptor.22 Values are the mean of at least three
determinations. SEM is shown in parentheses. c Maximum stimulation of [35S]GTP-γ-S binding expressed relative to the maximal effect
b
of 5-HT. ND ) Ki not determined. <50% displacement of radioligand at 10-7 M test ligand.
d
Ta ble 5. In Vitro Binding Affinities of 5-Substituted
N-[3-(1-Methyl-4-piperidinyl)-1H-pyrrolo[3,2-b]pyridines:
Sulfonamides and Ureas
Ta ble 6. In Vitro 5-HT1F Receptor Binding Affinity of
Substituted N-[3-(1-Substituted
4-piperidinyl)-1H-pyrrolo[3,2-b]pyridin-5-yl]amides: N-1
Substituent Effects
binding affinitya Ki (nM)
binding affinitya Ki (nM)
5-HT1F
compd
R
5-HT1F
5-HT1B
5-HT1D
NDb
NDb
compd
R
R′
ethyl
15a
15b
16a
16b
16c
4-fluorophenyl
methyl
4-fluorophenyl
methyl
27(4.6)
19(5.8)
11(0.6)
2.5(0.6)
3.8(0.4)
610(260)
3.2(0.6)
14(1.4)
NDb
380(57)
3.8(0.7)
8.5(1.3)
40(8.8)
17a
13a
17b
17c
17d
17e
19a
19b
19c
13b
4-F-phenyl
4-F-phenyl
4-F-phenyl
4-F-phenyl
4-F-phenyl
4-F-phenyl
4-F-phenyl
methyl
170(58)
25(5.5)
18(5.7)
14(4.7)
30(8.6)
75(32)
9.5(2.7)
45(22)
35(14)
21(5.4)
ethyl
n-propyl
n-butyl
isopropyl
phenylethyl
H
H
H
ethyl
ethyl
a
Affinities for receptors were determined in vitro by radioligand
binding assays using cell lines expressing the appropriate human
serotonin receptor.19-21 Each value is the mean of at least three
b
determinations. SEM is shown in parentheses. ND ) Ki not
ethyl
methyl
determined. <50% displacement of radioligand at 10-7 M test
ligand.
a
Affinities for receptors were determined in vitro by radioligand
binding assays using cell lines expressing the appropriate human
serotonin receptor.19-21 Each value is the mean of at least three
determinations. SEM is shown in parentheses.
migraine model. A rapid screening procedure using this
model was developed to evaluate the oral activity and
duration of action of potent and selective 5-HT1F recep-
tor agonists in the pyrrolo[3,2-b]pyridine series. Ini-
tially, we were interested in identifying compounds that
demonstrated complete blockade of plasma protein
extravasation upon oral administration with a long
duration of action. Sumatriptan was reported to have a
short plasma/tissue half-life in humans, and frequently
additional doses were required because of recurrence
of migraine pain.28 Since the normal duration of a
migraine headache is 4-72 h,29 a therapeutic agent with

5-HT1F Receptor Agonists
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 3067
Ta ble 7. In Vitro Contractile Response in Rabbit Saphenous
Vein and in Vivo Efficacy in Neurogenic Plasma Protein
Extravasation Screen
rabbit saphenous veina
neurogenic PPE screenb
contractile concn of max
ratio at ratio at
compd response (%) response (M) 1 h postdose 24 h postdose
3a
3b
3c
3m
3q
3w
3x
3cc
3d d
16(2.3)
5.7(1.7)
0
1.1(0.8)
0
3.0(1.5)
12.4(5.0)
2.3(1.2)
17(5.2)
10-4
3 × 10-5
1.0
1.1
1.5
1.0
1.2
1.1
1.6
10-4
3 × 10-5
10-4
1.6
1.0
10-4
10-4
10-4
1.4
10-4
a
Maximal contraction relative to KCl-induced (67 mM) contrac-
tion in the rabbit saphenous vein.10 Values are the mean (SEM in
parentheses) of three to seven determinations. Dural extrava-
b
F igu r e 3. Duration of action curves of 1a , 3b, and 3w in the
PPE model: ([) 1a ; (b) 3b; (3) 3w . Fasted guinea pigs were
given their approximate oral ID100, the minimal dose that
produced a full inhibition of extravasation 1 h postdose, and
the extravasation ratio was determined at various time points.
Data are expressed as the mean ( SEM (n ) 3).
sation ratio (stimulated side/unstimulated side) approximated
using the guinea pig neurogenic PPE assay (1 ng/kg, po).23
Ta ble 8. 5-HT Receptor Binding Profile of 3b and 3w
binding profilea Ki (nM)
receptor
3b
3w
5-HT1A
5-HT1B
5-HT1D
5-HT1E
5-HT1F
5-HT2A
5-HT2B
5-HT2C
5-HT4
620(37)
270(47)
250(38)
660(110)
5.0(0.5)
4800(2000)
>10000
1000(270)
720(220)
720(55)
770(80)
5.5(0.6)
3400(1600)
>10000
310(41)
2400(740)
1120(203)
NDb
1900(1600)
4000(350)
>10000
5-HT6
5-HT7
>5000
F igu r e 2. Dose-response curves of 1a , 3b, and 3w in the
PPE model after oral administration: ([) 1a ; (b) 3b; (4) 3w .
Values are the mean ratio of Evan’s Blue dye extravasation
in the stimulated side of the dura to that in the unstimulated
side ( SEM (n ) 3). A compound was considered fully
efficacious at a ratio of approximately 1 and an ineffective at
a ratio of approximately 1.8.
a
Affinities for receptors were determined in vitro by radioligand
binding assays using cell lines expressing the appropriate human
serotonin receptor.19-21 Each value is the mean of at least three
b
determinations. SEM is shown in parentheses. ND ) Ki not
determined. <50% displacement of radioligand at 10-7 M test
ligand.
travasation ratio at various times postdose (Figure 3).
Both 3b and 3w (1 ng/kg, po) were effective at 1 and 24
h, but not 48 h, postdose. Thus, the long duration of
action of 3b and 3w compared favorably to 1a and
indicated that these compounds may have a long dura-
tion of action in humans. As shown in Table 8, both
compounds displayed greater than 50-fold selectivity
over 10 other 5-HT receptor subtypes. In addition, these
compounds were evaluated for affinity at 40 other
neurotransmitter binding sites (receptors, ion channels,
and transporters; data not shown). Neither compound
demonstrated any notable affinity at any of the receptor
sites tested. On the basis of these studies, we character-
ized 3b and 3w as potent and selective, as well as orally
active and long acting, 5-HT1F receptor agonists (SSOF-
RAs) with the clinical potential to treat acute migraine.
a long duration of action might minimize breakthrough
headache. The compounds were tested at a single oral
dose of 1 ng/kg, and the extravasation ratio was
determined 1 h postdose (Table 7). For two compounds,
3m and 3cc, the stimulated/unstimulated extravasation
ratio was greater than 1, indicating that these com-
pounds had either limited oral activity at this dose or a
short duration of action. Compounds 3b, 3c, 3q, and 3w
were fully efficacious and were studied further. The
arylamides 3c and 3q (dosed at 1 ng/kg, po) possessed
a relatively long duration of action, exhibiting some
effect 24 h postdose (Table 7). However, the acetamide
derivative 3b and propionamide derivative 3w were
both fully efficacious 24 h postdose, indicating that both
compounds displayed excellent oral activity and dura-
tion of action in this model.
Full dose-response curves were generated for both
compounds (Figure 2). While less potent than compari-
tor compound 1a , both 3b and 3w decreased dural
extravasation in a dose-related manner following oral
administration (ID50 ) 0.2 and 0.6 ng/kg, respectively).
The duration of action of these compounds in the PPE
model was compared to 1a by administering an oral
ID100 to fasted guinea pigs and determining the ex-
Con clu sion s
We prepared a series of pyrrolopyridine and pyrrol-
opyrimidine analogues of SSOFRA 1a (LY334370) in an
effort to explore the structural motif required for 5-HT1F
receptor affinity and selectivity and to identify com-
pounds with an improved pharmacologic profile over 1a .
Evaluation of these analogues revealed the pyrrolo[3,2-

3068 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Filla et al.
tallization from EtOAc provided the title compound as a white
solid (1.81 g, 80%): mp ) 196-199 °C; 1H NMR (DMSO-d6) δ
10.61 (s, 1H), 8.11 (s, 1H), 7.11 (s, 1H), 6.54 (s, 1H), 4.92 (br
s, 2H), 2.87-2.83 (m, 2H), 2.51-2.53 (m, 1H), 2.20 (s, 3H),
2.00-1.64 (m, 6H) ppm; MS (m/e) 231 (M+). Anal. (C13H18N4‚
0.1C4H8O2) C, H, N.
b]pyridine derivatives as a novel series of 5-HT1F
receptor agonists. From the SAR studies of this series
coupled with in vivo pharmacologic evaluation, we have
identified methyl-N-[3-(1-methyl-4-piperidinyl)-1H-pyr-
rolo[3,2-b]pyridin-5-yl]amide (3b) and ethyl-N-[3-(1-
methyl-4-piperidinyl)-1H-pyrrolo[3,2-b]pyridin-5-yl]-
amide (3w ) as potent and selective 5-HT1F receptor
agonists.
4-F lu or o-N-[3-(1-m eth ylp ip er id in -4-yl)-1H-p yr r olo[2,3-
c]p yr id in -5-yl]ben za m id e (2a ). To a solution of 9 (0.14 g,
0.60 mmol) dissolved in pyridine (5 mL) was added 4-fluoro-
benzoyl chloride (0.09 mL, 0.71 mmol). The reaction mixture
was stirred at room temperature for 4 h, concentrated in vacuo,
and partitioned between CH2Cl2 and 1 N aqueous NaOH. The
aqueous layer was extracted with CH2Cl2, and the combined
organics were washed with saturated aqueous NaCl, dried
(Na2SO4), and concentrated in vacuo. Column chromatography
(5-20% 2 M NH3-MeOH/CH2Cl2) provided the title compound
(0.16 g, 77%) as a white solid: 1H NMR (CD3OD) δ 8.53 (s,
1H), 8.26 (s, 1H), 8.06 (dd, J ) 5.3, 8.9 Hz, 2H), 7.47 (s, 1H),
7.26 (t, J ) 8.7 Hz, 2H), 3.70-3.50 (m, 2H), 3.30-3.25 (m, 3H),
2.93 (s, 3H), 2.40-2.00 (m, 4H) ppm; MS (m/e) 252 (M+). Anal.
(C20H21FN4O) C, H, N.
N,N-Dim eth yl-N′-(1H-p yr r olo[3,2-b]p yr id in -5-yl)for m -
a m id in e (11). To a solution of 1017 (7.5 g, 49.0 mmol) dissolved
in 35 mL of DMF was added in one portion DMFDMA (32.5
mL, 244.7 mmol). The deep-red reaction mixture was heated
at 110 °C for 24 h, cooled to room temperature, and concen-
trated in vacuo to provide 12.5 g (97%) of the title compound
as a red solid. An analytical sample was purified by recrys-
tallization from benzene: mp ) 159 °C; 1H NMR (CDCl3) δ
8.43 (s, 1H), 8.17 (d, J ) 8.7 Hz, 1H), 8.00 (d, J ) 12.5 Hz,
1H), 6.40 (d, J ) 8.7 Hz, 1H), 6.39 (d, J ) 12.5 Hz, 1H), 3.13
(s, 3H), 3.10 (s, 3H), 2.99 (s, 6H) ppm.
Exp er im en ta l Section
Gen er a l Ch em ica l Meth od s. All commercially available
solvents and reagents were used as received. Reactions were
conducted under a nitrogen atmosphere with magnetic stir-
ring. Column chromatography was performed on E. Merck
Kieselgel 230-400 mesh silica gel 60 using the solvent system
described in parentheses. Melting points were obtained with
a Gallenkamp digital melting point apparatus and are uncor-
rected. 1H NMR spectra were recorded at 300 MHz in the
solvent noted in parentheses. Chemical shifts are reported in
ppm (δ) relative to solvent. Mass spectrometry was performed
using a Varian MAT 731 instrument. Elemental analyses were
carried out by the Physical Chemistry Department at Lilly
Research Laboratories.
N,N-Dim eth yl-N′-[4-(2-(d im eth yla m in o)vin yl)-5-n itr o-
p yr id in -2-yl]for m a m id in e (6). To a solution of 2-amino-4-
methyl-5-nitropyridine 5 (10.0 g, 65.3 mmol) dissolved in 90
mL of DMF was added in one portion N,N-dimethylformamide
dimethyl acetal (87.0 mL, 653.0 mmol). The deep-red reaction
mixture was heated at 110 °C overnight and then cooled to
room temperature and concentrated in vacuo to provide 17.7
g (100%) of the title compound as a red solid. An analytical
sample was purified by recrystallization from benzene: mp )
149-151 °C; 1H NMR (CDCl3) δ 8.84 (s, 1H), 8.57 (s, 1H), 7.28
(d, J ) 13.2 Hz, 1H), 6.82 (s, 1H), 6.03 (d, J ) 13.6 Hz, 1H),
3.10 (s, 3H), 3.08 (s, 3H), 2.93 (s, 6H) ppm; MS (m/e) 263 (M+).
Anal. (C12H17N5O2) C, H, N.
A mixture of this intermediate (7.2 g, 27.3 mmol) and 10%
palladium on carbon (1.1 g) in 90 mL of EtOH was hydroge-
nated at room temperature under 40 psi of hydrogen pressure
for 4 h, filtered through Celite, and concentrated in vacuo. The
resulting dark-brown foam was chromatographed (5-20% 2
M NH3-MeOH/CH2Cl2) to provide 3.8 g (74%) of a brown solid
that was slurried in EtOAc (150 mL), filtered, and rinsed with
cold EtOAc to provide the title compound as an ivory solid:
N,N-Dim eth yl-N′-(1H-p yr r olo[2,3-c]p yr id in -5-yl)for m -
a m id in e (7). A mixture of 6 (8.0 g, 30.4 mmol) and 10%
palladium on carbon (2.0 g) in 80 mL of EtOH was hydroge-
nated at room temperature under 40 psi of hydrogen pressure
for 24 h, filtered through Celite, and concentrated in vacuo.
The resulting dark-brown foam was chromatographed (10-
20% 2 M NH3-MeOH/CH2Cl2) to provide 3.4 g (60%) of a
brown solid that was slurried in EtOAc (50 mL), filtered, and
rinsed with cold EtOAc to provide the title compound as an
1
mp ) 187-189 °C; H NMR (DMSO-d6) δ 10.98 (s, 1H), 8.42
(s, 1H), 7.58 (d, J ) 8.5 Hz, H), 7.43 (m, 1H), 6.64 (d, J ) 8.5
Hz, 1H), 6.32 (m, 1H), 3.06 (s, 3H), 2.96 (s, 3H) ppm. Anal.
(C10H12N4) C, H, N.
Rep r esen ta tive P r oced u r e for th e Con d en sa tion of
1-Su bstitu ted 4-P ip er id on es w ith 11 (12a -g). P r ep a r a -
tion of 3-[1-Meth yl-(1,2,3,6-tetr a h yd r op yr id in -4-yl)]-1H-
p yr r olo[3,2-b]p yr id in -5-yla m in e (12a ). To a solution of
KOH (4.0 g, 70.7 mmol) and 11 (3.8 g, 20.2 mmol) dissolved
in MeOH (41 mL) was added 1-methyl-4-piperidone (3.2 mL,
26.3 mmol). The reaction mixture was heated at 75 °C for 24
h, cooled to room temperature, and concentrated in vacuo. The
residue was partitioned between 3:1 CHCl3/IPA and ice/water,
and the aqueous phase was extracted with CHCl3. The
combined organics were washed with saturated aqueous NaCl,
dried (MgSO4), and concentrated in vacuo. Column chroma-
tography (10-20% 2 M NH3-MeOH/CH2Cl2) followed by
recrystallization (CH3CN) provided 3.76 g (82%) of the title
compound as an orange solid: mp ) 199-202 °C; 1H NMR
(DMSO-d6) δ 10.69 (s, 1H), 7.40 (d, J ) 8.6 Hz, 1H), 7.23 (m,
1H), 7.04 (m, 1H), 6.32 (d, J ) 8.6 Hz, 1H), 5.38 (br s, 2H),
3.00 (m, 2H), 2.55-2.45 (m, 4H), 2.26 (s, 3H) ppm; MS (m/e)
229 (M+). Anal. (C13H16N4) C, H, N.
Rep r esen ta tive P r oced u r e for P r ep a r a tion of N-Su b-
stitu ted P ip er id in yl In ter m ed ia tes (14a -g). P r ep a r a tion
of 3-(1-Met h ylp ip er id in -4-yl)-1H -p yr r olo[3,2-b]p yr id in -
5-yla m in e (14a ). A solution of 12a (R ) CH3) (3.09 g, 13.9
mmol) and 10% palladium on carbon (0.46 g) in 130 mL of 1:1
THF/EtOH was hydrogenated at 40 psi for 24 h. The reaction
mixture was filtered through Celite, the catalyst was washed
with EtOAc, and the solution was concentrated in vacuo. The
crude material was slurried in 30 mL of EtOAc, cooled, filtered,
and rinsed with cold EtOAc, providing the title compound as
a white solid (2.74 g, 88%): 1H NMR (DMSO-d6) δ 10.45 (s,
1
ivory solid: mp ) 162-166 °C; H NMR (DMSO-d6) δ 11.25
(s, 1H), 8.45 (s, 1H), 8.42 (s, 1H), 7.46 (d, J ) 2.6 Hz, 1 H),
6.94 (s, 1H), 6.31 (d, J ) 2.6 Hz, 1H), 2.98 (s, 6H) ppm; MS
(m/e) 188 (M+). Anal. (C10H12N4) C, H, N.
3-[1-Meth yl-(1,2,3,6-tetr ah ydr opyr idin -4-yl)]-1H-pyr r olo-
[2,3-c]p yr id in -5-yla m in e (8). Sodium metal (1.20 g, 52.6
mmol) was added in portions to MeOH (50 mL) cooled to 0 °C.
When the sodium completely dissolved, 7 (2.48 g, 13.2 mmol)
followed by 1-methyl-4-piperidone (2.4 mL, 19.8 mmol) was
added. The reaction mixture was heated at 75 °C overnight,
and then 1 mL of H2O was added. The reaction mixture was
heated at the same temperature for an additional 24 h, cooled
to room temperature, and concentrated in vacuo. The residue
was partitioned between 3:1 CHCl3/IPA and ice/water, and the
aqueous phase was extracted with CHCl3. The combined
organics were washed with saturated aqueous NaCl, dried
(MgSO4), and concentrated in vacuo. Column chromatography
(10-20% 2 M NH3-MeOH/CH2Cl2) gave 2.59 g (86%) of the
title compound as a white solid: 1H NMR (DMSO-d6) δ 10.98
(s, 1H), 8.15 (s, 1H), 7.38 (s, 1H), 6.81 (s, 1H), 5.97 (br s, 1H),
5.04 (br s, 2H), 3.03 (m, 2H), 2.55-2.50 (m, 4H), 2.27 (s, 3H)
ppm; MS (m/e) 229 (M+). Anal. (C13H16N4) C, H, N.
3-(1-Meth ylp ip er id in -4-yl)-1H-p yr r olo[2,3-c]p yr id in -5-
yla m in e (9). A mixture of 8 (2.25 g, 5.0 mmol) and 10%
palladium on carbon (0.56 g) in 200 mL of EtOH was
hydrogenated at 60 psi at 40 °C for 24 h. The reaction mixture
was filtered through Celite, the catalyst was washed with
EtOAc, and the solution was concentrated in vacuo. Recrys-

5-HT1F Receptor Agonists
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 3069
1H), 7.38 (d, J ) 8.5 Hz, 1H), 7.02 (s, 1H), 6.28 (d, J ) 8.5 Hz,
1H), 5.33 (br s, 2H), 2.90-2.86 (m, 2H), 2.70 (m, 1H), 2.24 (s,
3H), 2.10-1.95 (m, 4H), 1.67-1.64 (m, 2H) ppm; MS (m/e) 230
(M+). Anal. (C13H18N4) C, H, N.
tioned between CHCl3 and 1 N aqueous NaOH. The aqueous
layer was extracted with CHCl3, and the combined extracts
were washed with saturated aqueous NaCl, dried (MgSO4),
and concentrated in vacuo. Column chromatography (10%
MeOH/CH2Cl2-MeOH) provided the title compound (73 mg,
90%) as a white solid: 1H NMR (DMSO-d6) δ 10.99 (s, 1H),
10.52 (s, 1H), 8.24 (m, 1H), 7.84 (m, 1H), 7.81-7.72 (m, 3H),
7.34 (m, 1H), 7.20 (t, J ) 4.4 Hz, 1H), 3.33 (br s, 1H), 3.03-
2.90 (m, 3H), 2.65-2.58 (m, 2H), 2.00-1.96 (m, 2H), 1.57-
1.52 (m, 2H) ppm; MS (m/e) 339 (M+). Anal. (C19H19FN4O) C,
H, N.
N-[4-(2-Dim et h yla m in ovin yl)-5-n it r o-6-oxo-1,6-d ih y-
d r op yr im id in -2-yl]-N,N-for m a m id in e (21). To 2018 (7.0 g,
41.1 mmol) dissolved in 100 mL of DMF was added tris-
(dimethylamino)methane (12.5 mL, 123.4 mmol). The reaction
mixture was heated at 100 °C overnight and then concentrated
in vacuo. Column chromatography (10% MeOH/CH2Cl2-
MeOH) provided the title compound (6.5 g, 57%) as a yellow
solid: 1H NMR (DMSO-d6) δ 11.23 (s, 1H), 8.82 (s, 1H), 8.17
(d, J ) 12.1 Hz, 1H), 5.48 (d, J ) 12.1 Hz, 1H), 3.22 (s, 3H),
3.05 (s, 3H), 3.20-2.80 (m, 6H) ppm; MS (m/e) 281 (M+). Anal.
(C11H16N6O3) C, H, N.
N,N-Dim eth yl-N′-(4-oxo-4,5-d ih yd r o-3H-p yr r olo[3,2-d ]-
p yr im id in -2-yl)for m a m id in e (22). To a solution of the
enamine 21 (12.4 g, 44.1 mmol) dissolved in 1:1 THF/EtOH
(140 mL) was added a slurry of 5% palladium on carbon (3.1
g) in EtOH. The reaction mixture was hydrogenated at 60 psi
for 2 h, filtered through Celite, and concentrated in vacuo.
Purification by column chromatography (CH2Cl2-10% 2 M
NH3-MeOH/CH2Cl2) gave 4.8 g (53%) of an ivory solid that
was recrystallized (MeOH): mp ) 269-271 °C; 1H NMR
(DMSO-d6) δ 11.58 (s, 1H), 10.96 (s, 1H), 8.54 (s, 1H), 7.18
(m, 1H), 6.09 (m, 1H), 3.11 (s, 3H), 2.99 (s, 3H) ppm; MS (m/e)
206 (M+). Anal. (C9H11N5O) C, H, N.
N,N-Dim eth yl-N′-(4-ch lor o-5H-pyr r olo[3,2-d]pyr im idin -
2-yl)for m a m id in e (23). A mixture of 22 (2.2 g, 10.7 mmol)
and phosphorus oxychloride (12.0 mL, 128.8 mmol) was stirred
vigorously while being heated at 115 °C. Within 0.25 h, the
reaction mixture became homogeneous. It was heated for an
additional 0.25 h, then quickly cooled to 0 °C in an ice/water
bath, basified with 5 N NaOH, and extracted with CHCl3. The
organic layer was washed with saturated aqueous NaCl, dried
(Na2SO4), and evaporated. The residue was chromatographed
(5% 2 M NH3-MeOH/CH2Cl2) to give 1.4 g (59%) of an ivory
solid: mp ) 174 °C (dec); 1H NMR (DMSO-d6) δ 11.90 (s, 1H),
8.53 (s, 1H), 7.75 (m, 1H), 6.42 (m, 1H), 3.11 (s, 3H), 2.99 (s,
3H), ppm; MS (m/e) 224 (M+). Anal. (C9H10ClN5) C, H, N.
Rep r esen ta tive P r oced u r e for th e Syn th esis of Su b-
stitu ted N-[3-(1-Su bstitu ted p ip er id in -4-yl)-1H-p yr r olo-
[3,2-b]p yr id in -5-yl]a m id es (3a -ff, 13a ,b, 17a -e a n d 18a -
c). P r ep a r a tion of 4-F lu or o-N-[3-(1-m eth ylp ip er id in -4-
yl)-1H-p yr r olo[3,2-b]p yr id in -5-yl]ben za m id e (3a ). To a
solution of 14a (R ) CH3) (1.00 g, 4.30 mmol) dissolved in
pyridine (85 mL) was added 4-fluorobenzoyl chloride (0.57 mL,
4.80 mmol). The reaction mixture was heated at 55 °C for 0.5
h, concentrated in vacuo, and partitioned between CH2Cl2 and
1 N aqueous NaOH. The aqueous layer was extracted with
CH2Cl2, and the combined organics were washed with satu-
rated aqueous NaCl, dried (Na2SO4), and concentrated in
vacuo. Column chromatography (10-20% 2 M NH3-MeOH/
CH2Cl2) followed by recrystallization (EtOH/H2O) provided the
title compound (1.32 mg, 87%) as a white solid: mp ) 118-
121 °C; 1H NMR (DMSO-d6) δ 10.99 (s, 1H), 10.47 (s, 1H), 8.14
(dd, J ) 5.7, 9.0 Hz, 2H), 7.80 (d, J ) 8.7 Hz, 1H), 7.75 (d, J
) 8.7 Hz, 1H), 7.35 (m, 1H), 7.32 (t, J ) 9.0 Hz, 2H), 2.86-
2.79 (m, 3H), 2.19 (s, 3H), 2.03-1.94 (m, 4H), 1.79-1.67 (m,
2H) ppm; MS (m/e) 352 (M+). Anal. (C20H21FN4O) C, H, N.
Rep r esen ta tive P r oced u r e for th e Syn th esis of Su b-
stitu ted N-[3-(1-Meth ylp ip er id in -4-yl)-1H-p yr r olo[3,2-b]-
p yr id in -5-yl]su lfon a m id es (15a ,b). P r ep a r a tion of 4-F lu -
or o-N -[3-(1-m e t h ylp ip e r id in -4-yl)-1H -p yr r olo[3,2-b]-
p yr id in -5-yl]ben zen esu lfon a m id e (15a ). To a solution of
14a (R ) CH3) (150 mg, 0.65 mmol) dissolved in pyridine (10
mL) was added 4-fluorobenzenesulfonyl chloride (152 mg, 0.78
mmol). The reaction mixture was heated at 55 °C for 2 h,
concentrated in vacuo, and partitioned between 3:1 CHCl3/IPA
and 1 N aqueous NaOH. The pH of the aqueous layer was
adjusted to ca. 11 with 1 N aqueous HCl and then extracted
with 3:1 CHCl3/IPA. The combined extracts were washed with
saturated aqueous NaCl, dried (MgSO4), and concentrated in
vacuo. Column chromatography (20% 2 M NH3-MeOH/CH2-
Cl2) followed by recrystallization (EtOH/H2O) provided the title
1
compound (134 mg, 53%) as a white solid: mp ) 152 °C; H
NMR (DMSO-d6) δ 11.20 (br s, 1H), 11.03 (s, 1H), 7.97 (dd, J
) 5.3, 8.7 Hz, 2H), 7.68 (d, J ) 8.7 Hz, 1H), 7.33 (t, J ) 8.7
Hz, 2H), 7.26 (m, 1H), 7.83 (dd, J ) 9.0 Hz, 1H), 2.84-2.81
(m, 2H), 2.64 (m, 1H), 2.22 (s, 3H), 2.01-1.94 (m, 2H), 1.81-
1.78 (m, 2H), 1.66-1.55 (m, 2H) ppm; MS (m/e) 388 (M+). Anal.
(C19H21FN4O2S‚1H2O) C, H, N.
Rep r esen ta tive P r oced u r e for th e Syn th esis of 1-Su b-
stitu ted 3-[3-(1-Meth ylp ip er id in -4-yl)-1H-p yr r olo[3,2-b]-
p yr id in -5-yl]u r ea s (16a -c). P r ep a r a tion of 1-(4-F lu or o)-
3-[3-(1-m eth ylp ip er id in -4-yl)-1H-p yr r olo[3,2-b]p yr id in -5-
yl]u r ea (16a ). To a solution of 14a (R ) CH3) (150 mg, 0.65
mmol) dissolved in 5:1 THF/DMF (6 mL) was added 4-fluoro-
phenyl isocyanate (0.10 mL, 0.85 mmol). The reaction mixture
was stirred at room temperature overnight, concentrated in
vacuo, and partitioned between 3:1 CHCl3/IPA and 1 N
aqueous NaOH. The aqueous layer was extracted with 3:1
CHCl3/IPA, and the combined extracts were washed with
saturated aqueous NaCl, dried (MgSO4), and concentrated in
vacuo. Column chromatography (10% 2 M NH3-MeOH/CH2-
Cl2) followed by recrystallization (MeOH) provided the title
compound (176 mg, 74%) as a white solid: mp ) 246 °C (dec);
1H NMR (DMSO-d6) δ 11.99 (br s, 1H), 11.01 (s, 1H), 9.54 (s,
1H), 7.74-7.71 (m, 3H), 7.34 (m, 1H), 7.15 (t, J ) 8.7 Hz, 1H),
6.90 (d, J ) 8.7 Hz, 2H), 2.90-2.75 (m, 3H), 2.25 (s, 3H), 2.03-
1.82 (m, 6H) ppm; MS (m/e) 369 (M+). Anal. (C20H22FN5O) C,
H, N.
N,N-Dim eth yl-N′-(5H-p yr r olo[3,2-d ]p yr im id in -2-yl)for -
m a m id in e (24). A mixture of 23 (1.34 g, 6.0 mmol), NaHCO3
(1.00 g, 12.0 mmol), and 10% palladium on carbon (0.60 g) in
EtOH (20 mL) was hydrogenated at 40 psi for 2 h. The reaction
mixture was filtered through Celite, concentrated in vacuo,
chromatographed (5% 2 M NH3-MeOH/CH2Cl2), and recrys-
tallized (MeOH/EtOAc), providing the title compound (0.72 g,
64%) as a white solid: 1H NMR (DMSO-d6) δ 11.43 (br s, 1H),
8.64 (s, 1H), 8.58 (s, 1H), 7.70 (m, 1H), 6.33 (m, 1H), 3.09 (s,
3H), 3.00 (s, 3H) ppm; MS (m/e) 190 (M+). Anal. (C9H11N5‚
0.2H2O) C, H, N.
7-(1-Meth ylpiper idin -4-yl)-5H-pyr r olo[3,2-d]pyr im idin -
2-yla m in e (25). Sodium metal (0.20 g, 8.46 mmol) was added
in portions to MeOH (10 mL) cooled to 0 °C. When the sodium
completely dissolved, 24 (0.40 g, 2.12 mmol) followed by
1-methyl-4-piperidone (0.39 mL, 3.18 mmol) was added. The
reaction mixture was heated at 75 °C for 48 h, then cooled to
room temperature, and concentrated in vacuo. The residue was
partitioned between 3:1 CHCl3/IPA and ice/water, and the
aqueous phase was extracted with CHCl3. The combined
organics were washed with saturated aqueous NaCl, dried
(MgSO4), and concentrated in vacuo. Column chromatography
(10-20% 2 M NH3-MeOH/CH2Cl2) followed by recrystalliza-
tion (MeOH) gave 0.33 g (68%) of the tetrahydropyridyl
Rep r esen ta tive P r oced u r e for th e Syn th esis of
Su bstitu ted N-(3-P ip er id in -4-yl-1H-p yr r olo[3,2-b]p yr i-
d in -5-yl]ben za m id es (19a -c). P r ep a r a tion of 4-F lu or o-
N-(3-p ip er id in -4-yl-1H -p yr r olo[3,2-b]p yr id in -5-yl)b en z-
a m id e (19a ). To 18a (R′ ) 4-fluorophenyl; R ) BOC) (107
mg, 0.24 mmol) dissolved in 2 mL of CH2Cl2 was added an
excess (0.5 mL) of TFA. The reaction mixture was stirred at
room temperature for 1 h, concentrated in vacuo, and parti-
1
intermediate as yellow crystals: mp ) 230-234 °C; H NMR
(DMSO-d6) δ 11.06 (s, 1H), 8.39 (s, 1H), 7.49 (s, 1H), 6.93 (s,
1H), 5.83 (br s, 2H), 3.00 (m, 2H), 2.59-2.38 (m, 4H), 2.27 (s,

3070 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Filla et al.
3H) ppm. Anal. Calcd for C12H15N5: C, 62.86; H, 6.59; N, 30.54.
Found: C, 63.09; H, 6.69; N, 30.69.
Hz, 1H), 7.22 (m, 1H), 6.49 (d, J ) 8.8 Hz, 1H), 3.83 (s, 3H),
2.88-2.85 (m, 2H), 2.75 (m, 1H), 2.22 (s, 3H), 2.09-1.98 (m,
3H), 1.83-1.75 (m, 3H) ppm; MS (m/e) 245 (M+). Anal.
(C14H19N3O) C, H, N.
A mixture of this material (0.50 g, 2.18 mmol) and 10%
palladium on carbon (0.13 g) in 35 mL of EtOH was hydroge-
nated at 60 psi overnight. The reaction mixture was filtered
through Celite, the catalyst was washed with EtOAc, and the
mixture was concentrated in vacuo to provide the title com-
pound as a white solid (0.35 g, 69%) that was recrystallized
from MeOH: 1H NMR (DMSO-d6) δ 10.79 (s, 1H), 8.34 (s, 1H),
7.30 (s, 1H), 5.73 (s, 2H), 2.80 (m, 2H), 2.62 (m, 1H), 2.50 (m,
2H), 2.18 (s, 3H), 1.96-1.90 (m, 4H), 1.68 (m, 2H) ppm; MS
(m/e) 232 (M+). Anal. (C12H17N5) C, H, N.
4-F lu or o-N-[7-(1-m eth ylp ip er id in -4-yl)-5H-p yr r olo[3,2-
d ]p yr im id in -2-yl]ben za m id e (4a ). To a solution of 25 (0.36
g, 1.56 mmol) dissolved in 30 mL of pyridine was added
4-fluorobenzoyl chloride (0.37 mL, 3.11 mmol). The cloudy
orange mixture was stirred at 50 °C overnight, then concen-
trated in vacuo, redissolved in MeOH, and placed on a 5 g
Mega Bond Elut SCX column (Varian Sample Preparation
Products). The column was washed with three volumes of
MeOH, and the product was removed from the column with
two volumes of 2 M NH3-MeOH. The resulting organics were
concentrated in vacuo. Radial chromatography (4 mm thick-
ness, 5-25% 2 M NH3-MeOH/CH2Cl2) provided the title
compound as an ivory solid (0.34 g, 62%): 1H NMR (DMSO-
d6) δ 11.50 (br s, 1H), 10.71 (s, 1H), 8.77 (s, 1H), 8.04 (dd, J )
3.3, 8.8 Hz, 2H), 7.65 (d, J ) 1.5 Hz, 1H), 7.32 (t, J ) 8.8, 2H),
2.84-2.81 (m, 2H), 2.75 (m, 1H), 2.19 (s, 3H), 2.01-1.94 (m,
4H), 1.79-1.67 (m, 2H) ppm; MS (m/e) 354 (M + 1). Anal.
(C19H20FN5O) C, H, N.
3-(1-Met h ylp ip er id in -4-yl)-1,4-d ih yd r op yr r olo[3,2-b]-
p yr id in -5-on e (29). A solution of 28 (2.30 g, 9.4 mmol) in 30
mL of 30% HBr/HOAc was heated in a sealed tube at 105 °C
for 72 h. The reaction mixture was cooled to room temperature
and concentrated in vacuo. The residue was dissolved in H2O,
and the pH was adjusted to ca. 13 with 5 N aqueous NaOH.
The mixture was concentrated in vacuo, and the residue was
chromatographed (20% 2 M NH3-MeOH/CH2Cl2). After con-
centration in vacuo, the product was dissolved in MeOH,
charged with Dowex 50X8 (100-200) ion-exchange resin (25
g, prewashed with 200 mL of H2O followed by 100 mL of
MeOH), and stirred overnight at room temperature. The
mixture was filtered, and the resin was washed with water
and MeOH. The resin was then stirred overnight in 100 mL
of 2 M NH3-MeOH and filtered. The filtrate was concentrated
in vacuo to provide 1.84 g (85%) of the desired material, which
was used without further purification: 1H NMR (DMSO-d6) δ
11.39 (br s, 1H), 10.85 (br s, 1H), 7.50 (d, J ) 9.4 Hz, 1H),
6.96 (m, 1H), 5.97 (d, J ) 9.4 Hz, 1H), 2.82-2.72 (m, 2H), 2.66
(m, 1H), 2.18 (s, 3H), 2.01-1.93 (m, 2H), 1.90-1.80 (m, 2H),
1.55-1.43 (m, 2H) ppm. Anal. (C13H17N3O) C, H, N.
Meth a n esu lfon ic Acid 3-(1-Meth ylp ip er id in -4-yl)-1H-
p yr r olo[3,2-b]p yr id in -5-yl Ester (30a ). To a solution of 29
(0.25 g, 1.08 mmol) cooled to 0 °C in pyridine (25 mL) was
added methanesulfonic anhydride (0.49 g, 2.81 mmol). The
reaction mixture was warmed to room temperature over 2 h
and then heated at 60 °C for an additional 2 h. The pyridine
was removed, and the residue was partitioned between 3:1
CHCl3/IPA and saturated aqueous NaHCO3 and then extracted
with 3:1 CHCl3/IPA. The combined organics were washed with
saturated aqueous NaCl, dried with Na2SO4, and concentrated
in vacuo. Chromatography (20% 2 M NH3-MeOH/CH2Cl2)
provided 0.18 g (53%) of the title compound: mp ) 142-144
5-Meth oxy-1H-p yr r olo[3,2-b]p yr id in e (27). To 2630 (29.4
g, 175 mmol) dissolved in 300 mL of DMF was added
DMFDMA (120 mL, 896 mmol) and Et3N (1 mL). The bright-
red reaction mixture was heated at 120 °C for 2 h and then
concentrated in vacuo to provide 38.9 g of a red solid, which
was used in the subsequent reaction without purification: 1H
NMR (CDCl3) δ 8.23 (d, J ) 9.0 Hz, 1H), 8.10 (d, J ) 9.0 Hz,
1H), 6.46 (d, J ) 12.4 Hz, 1H), 6.20 (d, J ) 12.4 Hz, 1H), 3.96
(s, 3H), 3.06 (s, 6H) ppm.
1
°C; H NMR (DMSO-d6) δ 11.32 (s, 1H), 7.88 (d, J ) 8.3 Hz,
1H), 7.52 (d, J ) 2.3 Hz, 1H), 6.92 (d, J ) 8.7 Hz, 1H), 3.61 (s,
3H), 2.87-2.84 (m, 2H), 2.81-2.73 (m, 1H), 2.20 (s, 3H), 2.06-
1.94 (m, 4H), 1.80 (dt, J ) 3.0, 12.1 Hz, 2H) ppm; MS (m/e)
309 (M+). Anal. (C14H19N3O3S) C, H, N.
The crude enamine (38.8 g, 174 mmol) was dissolved in 1.2
L of EtOH and charged with 10% palladium on carbon (5.0 g).
The mixture was hydrogenated at room temperature under
40 psi of hydrogen pressure for 4 h. After filtration through
Celite and chromatography (50% EtOAc/hexane), the material
was recrystallized from EtOAc/hexane to provide 19.6 g of an
Refer en ces
1
ivory solid (76%): mp ) 113-114 °C (dec); H NMR (DMSO-
(1) Plosker, G. L.; McTavish, D. Sumatriptan: A Reappraisal of Its
Pharmacology and Therapeutic Efficacy in the Acute Treatment
of Migraine and Cluster Headache. Drugs 1994, 47, 622-651.
(2) (a) Saxena, P. R.; Ferrari, M. D. 5-HT1-like Receptor Agonists
and the Pathophysiology of Migraine. Trends Pharmacol. Sci.
1989, 10, 200-204. (b) Humphrey, P. P. A.; Feniuk, W.; Perren,
M. J .; Connor, H. E.; Oxford, A. W.; Coates, I. H.; Butina, D.
GR43175, a Selective Agonist for the 5-HT1-like Receptor in Dog
Isolated Saphenous Vein. Br. J . Pharmacol. 1988, 94, 1123-
1132.
(3) MacIntyre, P. D.; Bhargava, B.; Hogg, K. J .; Gemmill, J . D.;
Hillis, W. S. Effect of Subcutaneous Sumatriptan, a Selective
5HT1 Agonist, on the Systemic, Pulmonary, and Coronary
Circulation. Circulation 1993, 87, 401-405.
(4) (a) Ottervanger, J . P.; Stricker, B. H. Cardiovascular Adverse
Reactions to Sumatriptan: Cause for Concern? CNS Drugs 1995,
3, 90-98. (b) Hillis, W. S.; MacIntyre, P. D. Sumatriptan and
chest pain. Lancet 1993, 341, 1564-1565. (c) Luman, W.; Gray,
R. S. Adverse reactions associated with sumatriptan. Lancet
1993, 341, 1091-1092.
(5) (a) Humphrey, P. P. A.; Goadsby, P. J . The mode of action of
sumatriptan is vascular? A debate. Cephalalgia 1994, 14, 401-
410. (b) Moskowitz, M. A. Neurogenic versus vascular mecha-
nisms of sumatriptan and ergot alkaloids in migraine. Trends
Pharmacol. Sci. 1992, 13, 307-316.
(6) Buzzi, M. G.; Moskowitz, M. A. The Trigemino-Vascular System
and Migraine. Pathol. Biol. 1992, 40, 313-317.
(7) (a) Buzzi, M. G.; Bonamini, M.; Moskowitz, M. A. Neurogenic
model of migraine. Cephalalgia 1995, 15, 277-280. (b) Mosk-
owitz, M. A.; Cutrer, F. M.; Lee, W. S.; Waeber, C.; Limmroth,
V.; Yu, X. Drug Discovery in Migraine Circa 1995. In Experi-
mental Headache Models; Olesen, J ., Moskowitz, M. A., Eds.;
Lippincott-Raven Publishers: Philadelphia, PA, 1995; pp 63-
73.
d6) δ 11.15 (br s, 1H), 7.69 (d, J ) 9.0 Hz, 1H), 7.47 (m, 1H),
6.54 (d, J ) 9.0 Hz, 1H), 6.39 (m, 1H), 3.84 (s, 3H) ppm; MS
(m/e) 149 (M+). Anal. (C8H8N2O) C, H, N.
5-Meth oxy-3-(1-m eth ylp ip er id in -4-yl)-1H-p yr r olo[3,2-
b]p yr id in e (28). To a mixture of 27 (7.0 g, 47 mmol) and KOH
(9.2 g, 165 mmol) in 350 mL of MeOH was added 1-methyl-
4-piperidone (9.9 mL, 80 mmol) in one portion. The reaction
mixture was heated at reflux temperature overnight and then
cooled to room temperature. The resulting precipitate was
collected, and the filtrate was concentrated to a minimal
volume. A second crop of crystals was collected, washed with
cold MeOH, and combined with the previous crop to afford 9.0
g (79%) of the tetrahydropyridyl intermediate as an ivory
1
solid: mp ) 203 °C (dec); H NMR (DMSO-d6) δ 11.10 (br s,
1H), 7.68 (d, J ) 8.8 Hz, 1H), 7.44 (m, 1H), 7.02 (m, 1H), 6.56
(d, J ) 8.8 Hz, 1H), 3.86 (s, 3H), 3.03 (m, 2H), 2.50-2.27 (m,
4H), 2.27 (s, 3H) ppm; MS (m/e) 243 (M+). Anal. Calcd for
C
14H17N3O: C, 69.11; H, 7.04; N, 17.27. Found: C, 69.22; H,
7.13; N, 17.47.
The intermediate described above (9.0 g, 37 mmol) was
dissolved in 190 mL of EtOH/THF/MeOH (10:10:1). A slurry
of 10% palladium on carbon (2.2 g) in EtOH was added, and
the reaction mixture was hydrogenated at 40 psi for 96 h. The
mixture was filtered through Celite, the catalyst was washed
with EtOH, and the filtrate was concentrated in vacuo. The
residue was chromatographed on silica gel (5-10% 2 M NH3-
MeOH/CH2Cl2) to provide 8.79 g (97%) of the desired mate-
rial: 1H NMR (DMSO-d6) δ 10.82 (br s, 1H), 7.61 (d, J ) 8.8

5-HT1F Receptor Agonists
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 3071
(8) (a) Hargreaves, R. J .; Williamson, D. J .; Shepheard, S. L.
Neurogenic inflammation: relation to novel antimigraine drugs.
In Migraine and Headache Pathophysiology; Edvinsson, L., Ed.;
Martin Dunitz: London, 1999; pp 95-103. (b) May, A.; Shep-
heard, S. L.; Knorr, M.; Effert, R.; Wessing, A.; Hargreaves, R.
J .; Goadsby, P. J .; Diener, H. C. Retinal plasma extravasation
in animals but not humans: implications for the pathophysiology
of migraine. Brain 1998, 121, 1231-1237. (c) May, A.; Gijsman,
H. J .; Wallno¨fer, A.; J ones, R.; Diener, H. C.; Ferrari, M. D.
Endothelian antagonist bosentan blocks neurogenic inflamma-
tion, but is not effective in aborting migraine attacks. Pain 1996,
67, 375-378.
(9) Compounds with efficacy in the clinic for the treatment of
migraine pain, such as sumatriptan, zolmitriptan, rizatriptan,
LY334370, and classical migraine abortives such as dihydroer-
gotamine, inhibit neurogenic dural inflammation following
electrical stimulation. (a) J ohnson, K. W.; Schaus, J . M.; Durkin,
M. M.; Audia, J . E.; Kaldor, S. W.; Flaugh, M. E.; Adham, N.;
Zgombick, J . M.; Cohen, M. L.; Branchek, T. A.; Phebus, L. A.
5-HT1F receptor agonists inhibit neurogenic dural inflammation
in guinea pigs. NeuroReport 1997, 8, 2237-2240. (b) Buzzi, M.
G.; Moskowitz, M. A. The antimigraine drug, sumatriptan
(GR43175), selectively blocks neurogenic plasma extravasation
from blood vessels in dura mater. Br. J . Pharmacol. 1990, 99,
202-206.
(10) Cohen, M. L.; Schenck, K. W. 5-Hydroxytryptamine1F Receptors
Do Not Participate in Vasoconstriction: Lack of Vasoconstriction
to LY344864, a Selective Serotonin1F Receptor Agonist in Rabbit
Saphenous Vein. J . Pharmacol. Exp. Ther. 1999, 290, 935-939.
(11) Cohen, M. L.; J ohnson, K. W.; Schenck, K. W.; Phebus, L. A.
Migraine therapy: relationship between serotonergic contractile
receptors in canine and rabbit saphenous veins to human
cerebral and coronary arteries. Cephalalgia 1997, 17, 631-638.
(12) (a) Bouchelet, I.; Cohen, Z.; Case, B.; Se´gue´la, P.; Hamel, E.
Differential Expression of Sumatriptan-Sensitive 5-Hydrox-
ytryptamine Receptors in Human Trigeminal Ganglia and
Cerebral Blood Vessels. Mol. Pharmacol. 1996, 50, 219-223. (b)
Verheggen, R.; Hundeshagen, A. G.; Brown, A. M.; Schindler,
M.; Kaumann, A. J . 5-HT1B receptor-mediated contractions in
human temporal artery: evidence from selective antagonists and
5-HT receptor mRNA expression. Br. J . Pharmacol. 1998, 124,
1345-1354. (c) De Vries, P.; Willems, E. W.; Heiligers, J . P. C.;
Villalo´n, C. M.; Saxena, P. R. Investigation of the role of 5-HT1B
and 5-HT1D receptors in the sumatriptan-induced constriction
of porcine carotid arteriovenous anastomoses. Br. J . Pharmacol.
1999, 127, 405-412.
(15) Shepheard, S.; Edvinsson, L.; Cumberbatch, M.; Williamson, D.;
Mason, G.; Webb, J .; Boyce, S.; Hill, R. Possible antimigraine
mechanisms of action of the 5-HT1F receptor agonist LY334370.
Cephalalgia 1999, 19, 851-858.
(16) (a) Goldstein, D. J .; Roon, K. I.; Offen, W. W.; Phebus, L. A.;
J ohnson, K. W.; Schaus, J . M.; VanLaar, T.; Ferrari, M. D.
Migraine treatment with selective 5-HT1F receptor agonist
(SSOFRA) LY334370. Celphalalgia 1999, 19, 318. (b) Goldstein,
D. J .; Roon, K. I.; Offen, W. W.; Ramadan, N. M.; Phebus, L. A.;
J ohnson, K. W.; Schaus, J . M.; Ferrari, M. D. Selective serotonin
1F (5-HT1F) receptor agonist LY334370 for acute migraine:
a
randomized controlled trial. Lancet 2001, 358, 1230-1234.
(17) Parker, E. D.; Shive, W. Substituted 2-Picolines Derived from
6-Amino-2-picoline. J . Am. Chem. Soc. 1947, 69, 63-67.
(18) Taylor, E. C.; Young, W. B. Pyrrolo[3,2-d]pyrimidine Folate
Analogues: “Inverted” Analogues of the Cytotoxic Agent
LY231514. J . Org. Chem. 1995, 60, 7947-7952.
(19) Adham, N.; Kao, H.-T.; Schechter, L. E.; Bard, J .; Olsen, M.;
Urquhart, D.; Durkin, M.; Hartig, P. R.; Weinshank, R. L.;
Branchek, T. A. Cloning of another human serotonin receptor
(5-HT1F): a fifth 5-HT1 receptor subtype coupled to the inhibition
of adenylate-cyclase. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 408-
412.
(20) Weinshank, R. L.; Zgombick, J . M.; Macchi, M. J .; Branchek, T.
A.; Hartig, P. R. Human serotonin1D receptor is encoded by a
subfamily of two distinct genes, 5-HT1DR and 5-HT1Dâ. Proc. Natl.
Acad. Sci. U.S.A. 1992, 89, 3630-3634.
(21) Zgombick, J . M.; Weinshank, R. L.; Macchi, M.; Schechter, L.
E.; Branchek, T. A.; Hartig, P. R. Expression and Pharmacologi-
cal Characterization of a Canine 5-Hydroxytryptamine1D Recep-
tor Subtype. Mol. Pharmacol. 1991, 40, 1036-1042.
(22) Wainscott, D. B.; J ohnson, K. W.; Phebus, L. A.; Schaus, J . M.;
Nelson, D. L. Human 5-HT1F receptor-stimulated [35S]GTPγS
binding: correlation with inhibition of guinea pig dural plasma
protein extravasation. Eur. J . Pharmacol. 1998, 352, 117-124.
(23) J ohnson, K. W.; Phebus, L. A. A fluorescence-based method for
assessing dural protein extravasation induced by trigeminal
ganglion stimulation. J . Neurosci. Methods 1998, 81, 19-24.
(24) MacroModel V6.5. Mohamadi, F.; Richards, N. G. J .; Guida, W.
C.; Liskamp, R.; Lipton, M.; Caulfield, C.; Chang, G.; Hendrick-
son, T.; Still, W. C. MacroModelsAn Integrated Software System
for Modeling Organic and Bioorganic Molecules Using Molecular
Mechanics. J . Comput. Chem. 1990, 11, 440-467.
(13) (a) Xu, Y.-C.; J ohnson, K. W.; Phebus, L. A.; Cohen, M.; Nelson,
D. L.; Schenck, K.; Walker, C. D.; Fritz, J . E.; Kaldor, S. W.;
LeTourneau, M. E.; Murff, R. E.; Zgombick, J . M.; Calligaro, D.
O.; Audia, J . E.; Schaus, J . M. N-[3-(2-Dimethylaminoethyl)-2-
methyl-1H-indol-5-yl]-4-fluorobenzamide: A Potent, Selective,
and Orally Active 5-HT1F Receptor Agonist Potentially Useful
for Migraine Therapy. J . Med. Chem. 2001, 44, 4031-4034. (b)
Schaus, J . M.; Audia, J . E.; Dressman, B. A.; Kaldor, S. W.;
Krushinski, J . H.; Adham, N.; Benvenga, M. J .; Brancheck, T.
A.; Calligaro, D. O.; Fuller, R. W.; Hemrick-Luecke, S. K.;
J ohnson, K. W.; Leander, J . D.; Lucaites, V. L.; Nelson, D. L.
G.; Overshiner, C. C.; Phebus, L. A.; Roush, M. E.; Wainscott,
D. B.; Wolff, M. C.; Zgombick, J . M. LY334370 is a High Affinity,
Selective 5-HT1F Receptor Agonist. Presented at the 8th Con-
gress of the International Headache Society, Amsterdam, J une
10-14, 1997.
(25) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T.
Semianalytical Treatment of Solvation for Molecular Mechanics
and Dynamics. J . Am. Chem. Soc. 1990, 112, 6127-6129.
(26) Structures of ligands studied were obtained from the Cambridge
Structural Database (CSD): Cambridge Crystallographic Data
Center (CCDC), 12 Union Road, Cambridge CBZ 1EZ, U.K.
(27) (a) Verheggen, R.; Freudenthaler, S.; Meyer-Dulheuer, F.; Kau-
mann, A. J . Participation of 5-HT1-like and 5-HT2A receptors in
the contraction of human temporal artery by 5-hydroxytryptamine
and related drugs. Br. J . Pharmacol. 1996, 117, 283-289. (b)
Razzaque, Z.; Heald, M. A.; Pickard, J . D.; Maskell, L.; Beer, M.
S.; Hill, R. G.; Longmore, J . Vasoconstriction in human isolated
middle meningeal arteries: determining the contribution of
5-HT1B- and 5-HT1F-receptor activation. Br. J . Clin. Pharmacol.
1999, 47, 75-82.
(14) (a) Dupuis, D. S.; Colpaert, C.; Pauwels, P. J . G-protein activa-
tion at 5-HT1A receptors by the 5-HT1F ligand LY334370 in
guinea-pig brain sections and recombinant cell lines. Br. J .
Pharmacol. 1998, 124, 283-290. (b) Overshiner, C. D.; Adham,
N.; Zgombick, J . M.; Branchek, T. A.; Calligaro, D. O.; Phebus,
L. A.; Roush, M. E.; J ohnson, K. W.; Hemrick-Luecke, S. K.;
Fuller, R. W.; Lucaites, V. L.; Wainscott, D. B.; Nelson, D. L.;
Wikffm, M. C.; Benvenga, M. J .; Audia, J . E.; Schaus, J . M.;
Krushinski, J . H.; Kaldor, S. W.; Dressman, B. A.; Leander, J .
D. LY334370 is Selective for the Cloned 5-HT1F Receptor.
Presented at the 26th Annual Meeting of the Society for
Neuroscience, Washington, DC, November 1996; Poster 52,812.
(28) Dahlo¨f, C. Would any acute treatment for migraine demonstrate
recurrence? Cephalalgia 1997, 17 (Suppl. 17), 17-20.
(29) Headache Classification Committee of the International Head-
ache Society. Classification and diagnostic criteria for headache
disorders, cranial neuralgias and facial pain. Cephalalgia 1988,
8 (Suppl. 7), 1-6.
(30) Baumgarten, H. E.; Chien-Fan Su, H. Synthesis of 3- and
5-Nitro-2-picoline and Derivatives. J . Am. Chem. Soc. 1952, 74,
3828-3831.
J M030020M