Pyrazolo[1,5-a]pyridine antiherpetics: Effects of the C3 substituent on antiviral activity

Article abstract of DOI:10.1016/j.bmcl.2007.02.058

A recently disclosed series of pyrazolo[1,5-a]pyridine inhibitors of herpes virus replication has been closely examined herein for effects of the C3 substituent on antiviral activity. Significant changes in activity are observed by alterations of the hete

Full text of DOI:10.1016/j.bmcl.2007.02.058

Bioorganic & Medicinal Chemistry Letters 17 (2007) 2858–2862
Pyrazolo[1,5-a]pyridine antiherpetics: Effects of the C3
substituent on antiviral activity
Brian A. Johns,* Kristjan S. Gudmundsson and Scott H. Allen
Department of Medicinal Chemistry, Infectious Diseases CEDD, GlaxoSmithKline Research & Development,
Five Moore Drive, Research Triangle Park, NC 27709-3398, USA
Received 9 January 2007; revised 21 February 2007; accepted 21 February 2007
Available online 25 February 2007
Abstract—A recently disclosed series of pyrazolo[1,5-a]pyridine inhibitors of herpes virus replication has been closely examined
herein for effects of the C3 substituent on antiviral activity. Significant changes in activity are observed by alterations of the hetero-
atom basicity and orientation of the group at C3. These results in combination with previous studies have served to further elaborate
the minimal pharmacophore required for potency of this novel series of antiviral agents. During the course of these studies, several
novel synthetic approaches were developed and are described.
Ó 2007 Elsevier Ltd. All rights reserved.
Herpesviruses are among the most prevalent infectious
F
F
agents known and have the ability to infect nearly every
animal species.1 There are currently eight herpesviruses
that infect humans including herpes simplex 1 and 2
(HSV-1, HSV-2). The severity of the viral infections
can range from the host being asymptomatic due to ade-
quate control of the virus by the immune system to dis-
comfort caused by oral fever blisters to death in certain
immunocompromised individuals or neonates.2 The cur-
rent gold standard therapy, valacyclovir, is well known,
widely used, and exceptionally well tolerated. While this
has been an extremely successful therapy for patients,
there is still a goal to improve treatment in the areas
of time to healing of lesions, attenuation of pain, and
reduction in the frequency of reactivations. A major
goal of our current efforts has been to find novel thera-
peutic agents to control viral replication through alter-
native mechanisms of action (MOA) such that
potential synergy and complimentarity with valacyclovir
type of treatments may exist.
N
N
6'
N
H
N
7
H
N
N
N
R
3
N
2'
N
GW3733 (1)
2
H
Figure 1.
described the synthesis and SAR for the C2, C6 and
the effects resulting from combinations of C20 and C7
substituent alterations.4 While those initial reports kept
the C3 pyrimidine substituent constant, the current
investigation was aimed to examine the importance of
the pyrimidine ring appended at C3 with respect to the
heteroatom requirements within the pyrimidine ring
along with their basicity and orientation for optimal
activity.
We have recently disclosed a series of heterocyclic mol-
ecules containing the pyrazolo[1,5-a]pyridine core which
show potent and selective inhibition of HSV-1 and -2,
and do not act via the traditional herpes polymerase
MOA (Fig. 1).3 Previous reports from our laboratories
Our initial strategy to examine the C3 substituent effects
was to gradually tear down the 20-aminopyrimidine
group of the lead structure GW3733 beginning with
the removal of the individual ring heteroatoms through
a series of pyridine derivatives. This series of analogs
were made starting from the previously reported
3-bromo derivative 33 through a standard Suzuki
cross-coupling sequence as depicted in Scheme 1. The
4-pyridyl (6) and 2-fluoro-4-pyridyl derivatives (7) were
Keywords: Pyrazolo[1,5-a]pyridine; Herpes; HSV.
*
Corresponding author. Tel.: +1 919 483 6006; fax: +1 919 483
6053; e-mail: brian.a.johns@gsk.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.02.058

B. A. Johns et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2858–2862
2859
Toward this goal, the 4-pyridazine derivative 23 was
made by exploiting the nucleophilicity of the 3-position
of the pyrazolopyridine ring system (Scheme 3). From
previous work, it was determined that the 7-chloro
derivative 19 had the optimal balance of C3 reactivity
and did not undergo side reactions at the 4- and 6-posi-
tions as is the case if the 7-chloro is replaced at this stage
with an amino substituent.3 Treatment of 19 with pyrid-
azine and ethyl chloroformate at 0 °C resulted in a sep-
arable 2:1 mixture of the 4- and 2-substituted
dihydropyridazine carbamates 20 and 21.8 The desired
4-substituted intermediate 20 was re-aromatized to the
4-pyridazinyl analog 22 by treatment with LiOH in the
presence of air. The 7-chloro substituent was displaced
under Buchwald type coupling conditions in the pres-
ence of cyclopentylamine resulting in the pyridazine
containing analog 23.9
F
F
a,b
N
N
N
NH
N
NH
X
Br
N
3
X
B(OH)2
B(OH)
2 6 X = H
9 X = NHNH2
10 X = N3
7 X = F
N
N
4
55
8 X = NH-c-C5H9
Scheme 1. Reagents and conditions: (a) PdCl2(PPh3)2, 2 M (aq)
Na2CO3, DMF 100 °C, 4 or 5, 18 h (for 6, 46%, for 7, 77%); (b)
starting from 7: for 8, cyclopentylamine, 150 °C neat, 12 h (72%), for 9,
hydrazine (anhyd), EtOH, 100 °C, 13 h (65%); for 10, NaN3, NMP,
129 °C, 4 d (52%). See above-mentioned references for further
information.
constructed using this method. The 2-fluoro-4-pyridyl
derivative proved to be not only interesting for SAR
purposes but also was a useful intermediate for further
displacements with various nucleophiles resulting in
the cyclopentylamino (8), hydrazino (9), and azido (10)
substituted variants.
The isomeric pyrimidine 30 was the next target exam-
ined. This analog effectively moves the 3-position nitro-
gen in the parent GW3733 structure to the 5-position.
The seemingly simple change however required a vastly
different synthetic approach. Our strategy to construct
this analog was to make use of the [3+2] cyclization of
N-aminopyridinium iodide with an alkyne which has
been shown in the past to be a powerful method of con-
structing the pyrazolopyridine ring system (Scheme 4).4c
It was found that conversion of commercially available
4,6-dichloropyrimidine (24) to its corresponding diio-
dide 25 resulted in a substrate that could be selectively
converted to the 4-amino-6-iodopyrimidine 26. The
amine displacement chemistry works as well with the
dichloro starting material but we were unable to convert
the remaining chlorine into iodide 26 once the amine
was introduced. The mono-chloro derivative was also
much slower to undergo the subsequent Sonogashira
coupling thus we used the sequence of steps shown in
Scheme 4.
The unsubstituted 3-pyridyl derivative 14 was made
using analogous Suzuki chemistry using the boronic acid
12 (Scheme 2). In addition, it was desired to construct
the corresponding cyclopentylamine substituted 3-pyr-
idyl derivatives having the amine flank either side of
the pyridine nitrogen. These analogs were made using
a related strategy from the appropriate fluoropyridine
coupling partners through Stille or Suzuki methodology
to couple bromide 3 with stannane 116 and boronic acid
137 to give the fluoro derivatives 15 and 17, respectively.
To complete the synthesis, nucleophilic displacement of
the fluorine was cleanly accomplished in each case by
heating in neat cyclopentylamine in a sealed tube for
5–7 days resulting in the aminopyridine analogs 16 and
18.
With the iodo-pyrimidine 26 in hand, a Sonogashira
coupling with the commercially available 4-fluorophenyl
acetylene proceeded smoothly to give the alkyne 27.
This material served as the key intermediate for [3+2]
cyclization with the aminopyridinium salt 28 to give
In addition to the pyridine replacements at C3, we also
desired to look into additional analogs containing
multiple nitrogens in the 6-membered heterocycle.
SnBu3
F
F
F
F
a,b
N
F
-or-
3
a
11
N
c,d
N
N
N
N
N
NH
B(OH)2
6
Cl
F
N
Cl
F
N
Cl
N
N
EtO2C
X2
N
N
N
X
N
19
20
CO2Et
12 X = H
14 X = H
21
13 X = 2-F
15 X = 2-F
b
16 X = 2-NH-c-C5H9
c
20
17 X = 6-F
N
N
N
N
18 X = 6-NH-c-C5H9
Cl
NH
N
N
N
N
Scheme 2. Reagents and conditions for 15: (a) Pd(PPh3)4, 11, DMF
100 °C, 18 h (28%), for 16; (b) cyclopentylamine, 140 °C neat, 7 d
(87%), for 14 and 16; (c) PdCl2(PPh3)2, 12 or 13, 2 M (aq) Na2CO3,
DMF 100 °C, 18 h (for 14, 46%, for 16, 27%), for 18; (d) cyclopen-
tylamine, 140 °C neat, 5 d (99%).
22
23
Scheme 3. Reagents and conditions: (a) pyridazine, ClCO2Et, CH2Cl2,
0 °C 19:20 2:1 (99%); (b) 1 M LiOH (aq), MeOH/H2O, air (86%); (c)
Pd(OAc)2, BINAP, Cs2CO3, cyclopentylamine, PhMe, 105 °C (93%).

2860
B. A. Johns et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2858–2862
Eschenmoser’s salt in the presence of trifluoroacetic acid
results in complete regioselective formation of the N,N-
dimethyl-aminomethyl substituted intermediate 41
(Scheme 6). This benzylic amine could be activated as
its quaternary ammonium salt by treatment with methyl
iodide to give 42 which was smoothly converted to the
benzylic nitrile 43 upon heating with KCN in NMP.
The activated methylene of the nitrile was homologated
to enamine 44 with dimethylformamide dimethylacetal
and DBU, and finally condensed with hydrazine to give
the desired amino-substituted pyrazole 45.
c
a
N
N
N
N
N
N
Cl
Cl
b
I
X
I
24
25 X = I
27
26 X = NH-c-C5H9
F
F
F
e,f,g
d
N
N
N
N
N
N
NH2
N+
NH
I-
N
N
28
N
N
H
H
29
30
A final series of analogs were made to examine the
effects of an imidazole, thiazole, and oxazole at the C3
position. Previously known methyl ketone 463 was con-
verted to its silyl enol ether and brominated with NBS in
a one-pot method to give the a-bromoketone 47. This
material was condensed with thiourea to give the corre-
sponding 4-(2-aminothiazole) substituted pyrazolopyri-
dine. A similar condensation was accomplished using
guanidine and amide nucleophiles to give the desired
imidazole and oxazole derivatives 49–51 albeit not unex-
pectedly in significantly lower yields than the thiazole
counterpart.
Scheme 4. Reagents and conditions: (a) NaI in 57% HI (58%);
(b) cyclopentylamine, THF 0 °C to rt (91%); (c) PdCl2(PPh3)2,
4-FC6H4CCH, CuI, Et3N, THF, rt (97%); (d) DBU, MeCN (31%);
(e) LDA, MeSSMe, THF ꢀ78 °C (85%); (f) m-CPBA, CH2Cl2; (g)
cyclopentylamine 145 °C 6 h (20%).
the 3 substituted pyrazolopyridine 29 as a single regio-
isomer. All that remained was to deprotonate the 7-po-
sition and functionalize by treatment of the anion at low
temperature with dimethyl disulfide. The 7-methylthio
intermediate was smoothly oxidized to the sulfoxide
which was displaced under thermal conditions in cyclo-
pentylamine/THF in a sealed tube to give amine 30.
The objective of the current study was to investigate po-
tential replacements for the C3 pyrimidine in GW3733.
The goal was to look for alternative C3 groups that re-
duce MW and lipophilicity. Toward that end the closest
analogs were envisioned to be a set of pyridines that sim-
ply removed one of the pyrimidine ring heteroatoms.
While a pyridine replacement might add increased solu-
bility, it could be imagined that replacement of the
pyrimidine N-3 with a CH might cause unfavorable ste-
ric interactions with the neighboring pyrazolopyridine
ring. Compound 8 clearly shows that the pyridine ring
is tolerated with little to no effect on antiviral activity
resulting from removal of the ‘inner’ pyrimidine
nitrogen (Table 1) (Scheme 7).
The final synthetic goal in the C3 6-membered ring series
of analogs was to remove the ring heteroatoms and
maintain the amine substituent. The simple aniline
derivative 31 seemed to be an appropriate derivative to
answer the question of the importance of removing ring
nitrogens while keeping the exocyclic amine. This was
made via standard Suzuki coupling of bromide 3 and
3-aminophenyl boronic acid. The corresponding pyrmi-
dine containing compound 32 (20-NH2) was also in-
cluded for activity comparison and was made as
previously reported.4 In addition, several 5-membered
ring heterocycles which were devoid of basic nitrogens
(examples 33–37) were also appended at C3 via the
Suzuki or Stille methodology as described above.
Pyrazole and isoxazole derivatives 39 and 40 were made
by treating vinylogous amide 383 with hydrazine or
hydroxylamine in ethanol, respectively (Scheme 5).
F
F
a,b
c,d
N
Substituted pyrazoles were not accessible through the
above methods of simple condensation with earlier
intermediates or via cross-coupling strategies so an alter-
nate route was developed. Treatment of 19 with
N
N
NH
19
N
Cl
NC
X
43
41 X = NMe2
F
F
42 X = N+Me3I-
f
e
F
F
N
N
N
N
Me2N
HN
N
NH
NH
45
a
NC
N
N
NH2
N
N
O
44
39 X = NH
NH
40 X = O
NH
N X
Scheme 6. Reagents and conditions: (a) Eschenmosher’s salt, TFA,
CH2Cl2 (99%); (b) MeI, MeOH (96%); (c) KCN, NMP, 100 °C (68%);
(d) Pd(OAc)2, BINAP, Cs2CO3, cyclopentylamine, PhMe, 105 °C
(89%); (e) DMF–DMA, DBU, 115 °C (35%); (f) anhyd H2NNH2,
EtOH, AcOH (90%).
Me2N
38
Scheme 5. Reagents and conditions: (a) for 39 anhyd H2NNH2, EtOH,
reflux, 2 h (86%), for 40, H2NOH HCl, K2CO3, EtOH (93%).

B. A. Johns et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2858–2862
Table 1. HSV-1 antiviral activity and cytotoxicity of C3 analogs
2861
F
N
N
H
N
R
TIc
a
b
Compound
R
EC50 (lM)
CC50 (lM)
1
6
4-(2-c-C5H9NH-pyrimidyl)–
4-Pyridyl–
4-(2-F-pyridyl)–
0.26
2.5
>160
ꢁ20
n.d.
>40
40
>615
80
7
>40
0.37
5.7
n.d.
>108
7
8
4-(2-c-C5H9NH-pyridyl)–
4-(2-Hydrazino-pyridyl)–
4-(2-N3-pyridyl)–
3-Pyridyl–
9
10
14
15
16
17
18
23
30
31
32
33
34
35
36
37
39
40
45
48
49
50
51
>40
3.4
n.d.
>40
n.d.
n.d.
n.d.
n.d.
n.d.
>40
n.d.
46
n.d.
>11
n.d.
n.d.
n.d.
n.d.
n.d.
>46
n.d.
11
3-(2-F-pyridyl)–
3-(2-c-C5H9NH-pyridyl)–
3-(6-F-pyridyl)–
n.d.
>40
>40
>40
>10
0.85
>10
4.1
3-(6-c-C5H9NH-pyridyl)–
4-Pyridazinyl–
4-(6-c-C5H9NH-pyrimidyl)–
3-NH2–C6H4–
4-(2-NH2-pyrimidyl)–
5-Indolyl–
3-Thiopheneyl–
3.9
20
5
>10
>10
>10
4.7
n.d.
n.d.
n.d.
10
n.d.
n.d.
n.d.
2
3-Furanyl–
2-Furanyl–
4-Pyrazolo–
3-Pyrazolo–
5-Isoxazolo–
2.5
14
6
>10
4.9
n.d.
20
n.d.
4
4-(3-NH2-pyrazolo)–
4-(2-NH2-thiazolo)–
4-(2-NH2-imidazolo)–
4-(2-c-C5H9NH-imidazolo)–
4-(2-CH3-oxazolo)–
8.6
n.d
n.d.
40
n.d.
n.d.
5
>10
7.9
>40
n.d.
n.d.
n.d., not determined.
a Vero cells, SC-16 strain. EC50 is the concentration at which 50% efficacy in the antiviral assay is observed using a capture hydrid method.
b CC50 is the concentration at which 50% cytotoxicity is observed.
c Therapeutic index (CC50/EC50).
city of the pyridine nitrogen as is the case in 7 resulted in
F
F
a dramatic loss in activity. The 3-pyridyl series (14)
showed comparable activity to the 4-pyridyl derivative
6. However in this case, the potency could not be im-
proved by the addition of hydrophobic amines at either
the 2- or 6-positions (16 and 18). The pyrimidine nitro-
gen in the 3-position could also be ‘moved’ across the
ring to give formally the 6-cyclopentylamine substituted
pyrimidine 30 with little loss in activity. As expected, the
complete removal of the pyridine or pyrimidine nitrogen
as in examples 31 and 33–36 results in no appreciable
antiviral activity.
a
N
N
O
O
N
NH
F
N
NH
Br
47
46
48 X = S, R = NH2
b
49 X = NH, R = NH2
50 X = NH, R = NH-c-C5H9
51 X = O, R = Me
N
N
X
NH
N
R
The remaining question was to determine if any 5-mem-
bered ring heterocycles could retain the activity of the
pyrimidine/pyridine series. Interestingly, after a fairly
extensive amount of work to make imidazole, thiazole,
oxazole, and pyrazole examples only modest activity
was observed and typically was not well separated from
cytotoxicity.
Scheme 7. Reagents and conditions: (a) TBSOTf, i-PrNEt, CH2Cl2,
0 °C 30 min, then add NBS in THF (75%); (b) for 48, H2NC(@S)NH2,
DMF, 1 h, 90 °C (56%); for 49, H2NC(@NH)NH2, K2CO3, DMF,
80 °C (18%); for 50, H2NC(@NH)NH–c-C5H9, K2CO3, DMF, 80 °C
(17%); for 51, H2NCOMe, DMF, 90 °C (20%).
It is also consistent with our previous work that removal
of the cyclopentylamine group as is the case for analog 6
has a deleterious effect on potency. Decreasing the basi-
The above SAR study has served to map out the
C3 pharmacophore requirements in detail. A clear

2862
B. A. Johns et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2858–2862
Med. Chem. 2005, 13, 5346; (b) Allen, S. H.; Johns, B. A.;
Gudmundson, K. S.; Freeman, G. A.; Boyd, F. L., Jr.;
Sexton, C. J.; Selleseth, D. W.; Creech, K. L.; Moniri, K. R.
Bioorg. Med. Chem. 2006, 14, 944; (c) Johns, B. A.;
Gudmundsson, K. S.; Turner, E. M.; Allen, S.; Samano, V.
A.; Ray, J. A.; Freeman, G. A.; Boyd, F. L., Jr.; Sexton, C.
J.; Selleseth, D. W.; Creech, K. L.; Moniri, K. R. Bioorg.
Med. Chem. 2005, 13, 2397.
understanding that a 4-substituted, Lewis basic 6-mem-
bered ring heterocycle was essential to the potent antivi-
ral activity of the series was a key outcome of these
investigations. In addition several new strategies for
the construction of novel 3-substituted pyrazolopyridine
ring systems have been developed.
5. Boyd, F. L.; Chamberlain, S. D.; Cheung, M.; Gudmunds-
son, K.; Harris, P. A.; Johns, B. A.; Jung, D. K.; Peel, M.
R.; Stanford, J. B.; Sexton, C. J. WO 02/48148, 2002; Chem.
Abstr. 2002, 137, 33325.
6. Stannane 11 was prepared from 2-F-pyridine and LDA in
THF followed by quenching with Bu3SnCl.
References and notes
1. Roizman, B.; Pellett, P. E. In Fields Virology; Knipe, D. M.,
Howley, P. M., Eds.; Lippincott Williams and Wilkins:
Philadelphia, PA, 2001; Vol. 2, p 2381.
7. Boronic acid 13 was prepared from 2-F,5-Br-pyridine and
BuLi in THF followed by treatment with B(OMe)3 and
acidic workup.
8. The structure of 21 was confirmed by X-ray.
9. Interestingly, 2-substituted dihydro intermediate 21 did
not undergo the aromatization step when subjected to
the same conditions as its 4-substituted counterpart,
20.
2. Vinh, D. C.; Aoki, F. Y. Expert Opin. Pharmacother. 2006,
7, 2271.
3. Johns, B. A.; Gudmundsson, K. S.; Turner, E. M.; Allen, S.
H.; Jung, D. K.; Sexton, C. J.; Boyd, F. L., Jr.; Peel, M. R.
Tetrahedron 2003, 59, 9001.
4. (a) Gudmundsson, K. S.; Johns, B. A.; Wang, Z.; turner, E.
M.; Allen, S. H.; Freemen, G. A.; Boyd, F. L., Jr.; Sexton,
C. J.; Selleseth, D. W.; Moniri, K. R.; Creech, K. L. Bioorg.