New thiopyrazolo[3,4-d]pyrimidine derivatives as anti-mycobacterial agents

Article abstract of DOI:10.1016/j.bmcl.2006.12.066

The multiple parallel synthesis of a series of N,S-bis-alkylated thiopyrazolo[3,4-d]pyrimidines, based on sequential S- then N-alkylation, is reported. These compounds showed significant anti-mycobacterial activity (MICs down to ≤2 μg/ml) and their potent

Full text of DOI:10.1016/j.bmcl.2006.12.066

Bioorganic & Medicinal Chemistry Letters 17 (2007) 1736–1740
New thiopyrazolo[3,4-d]pyrimidine derivatives
as anti-mycobacterial agents
a
c
c
Lluis Ballell,a,b, Robert A. Field, Gavin A. C. Chung and Robert J. Young
*
aCentre for Carbohydrate Chemistry, School of Chemical Sciences and Pharmacy, University of East Anglia, Norwich, NR4 7TJ, UK
bGlaxoSmithKline, MMPD CEDD, Diseases of the Developing World (DDW), Tres Cantos, 28760, Spain
cGlaxoSmithKline, Medicines Research Centre, Gunnels Wood Road, Stevenage, SG1 2NY, UK
Received 27 November 2006; revised 14 December 2006; accepted 16 December 2006
Available online 22 December 2006
Abstract—The multiple parallel synthesis of a series of N,S-bis-alkylated thiopyrazolo[3,4-d]pyrimidines, based on sequential S- then
N-alkylation, is reported. These compounds showed significant anti-mycobacterial activity (MICs down to 62 lg/ml) and their
potential as significant drug-like leads is substantiated through cytotoxicity evaluation and in silico profiling.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Synthetic drugs for treating tuberculosis (TB) have been
available for over half a century, but incidences of the
disease continue to rise world-wide. In 2004, the last
year for which statistics are available, it is estimated that
24,500 people developed active disease and close to 5500
died from TB every day.1 Co-infection with HIV is driv-
ing the increase in incidence2 and the cause of death in
31% of AIDS patients in Africa can be attributed to
TB.3,4 When coupled with the emergence of multi-drug
resistant strains of Mycobacterium tuberculosis (MDR-
TB), the scale of the problem is amplified. It is now more
than a decade since the WHO declared TB ‘a global
health emergency’.1 The need for new drugs to extend
the range of TB treatment options is hence acute; in par-
ticular, new chemical entities5 with novel mechanisms of
action are required.6
drugs (e.g., 6-thiopurine for leukaemia,7 acyclovir as
an anti-viral8 and allopurinol for treatment of gout9).
Recently, the purine core has been exploited in the syn-
thesis of protein kinase inhibitors,10–13 inhibitors of car-
bohydrate14 and estrogen15 sulfotransferases, as well as
in compounds displaying osteogenesis-inducing activity
in stem cells.16 Compounds of this generic nature (1–5)
have also been reported to display anti-mycobacterial
activity (Fig. 1).17–20
O
S
SC12H25
N
HN
N
N
N
N
N
N
N
Cl
N
N
H
N
O S O
NMe2
(1)
MIC 0.78 μg/ml
(2)
(3)
Purine-containing molecules are ubiquitous in nature
and are found as components of nucleosides, nucleo-
tides, co-factors and signalling molecules. A significant
proportion of any genome codes for proteins that recog-
nise purine-containing ligands (e.g., kinases, DNA and
RNA polymerases, ATPases, GTPases, purine recep-
tors). Unsurprisingly, structural analogues of purines
have proved attractive templates for drug discovery pro-
grammes, leading to a number of significant synthetic
MIC 0.78 μg/ml
MIC 0.78 μg/ml
O2N
NO2
S
N
H
N
N
H3C
N
N
S
N
(4)
(5)
Keywords: Anti-mycobacterial; Anti-tubercular; Thiopyrazolopyrimi-
dines; SAR.
MIC 1.3 μg/ml
MIC 12.5 μg/ml
*
Corresponding author. Tel.: +34 9180 70412; fax: +34 9180 70550;
e-mail: lluis.p.ballell@gsk.com
Figure 1. Purine-related anti-mycobacterials: (1),17 (2),18 (3),19(4),20 (5).25
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.12.066

L. Ballell et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1736–1740
1737
Table 1. MIC data30 for an initial set of nine 4-thiopyrazolopyrimidine
derivatives against Mycobacterium tuberculosis H37Rv
The isomeric pyrazolo[3,4-d]pyrimidine nucleus has
attracted comparatively little attention, though it has
been exploited in inhibitors of Src protein kinases,21
EGF receptor tyrosine kinases,22 cyclic AMP phosphodi-
esterases23 and Staphylococcus aureus DNA polymerase
III.24 Interestingly, such a compound reported as an
intermediate in the synthesis of prospective HIV replica-
tion inhibitors showed modest anti-mycobacterial activi-
ty (compound 5; MIC 12.5 lg/ml).25 In addition, the
often present alkylsulfanyl moiety has been previously
reported as important for anti-mycobacterial activity.26
This result prompted us to explore the thiopyrazolo[3,4-
d]pyrimidine template (TPP) in our search for new anti-
mycobacterial agents, given an encouraging level of
potency and straightforward syntheses via sequential S-
then N-alkylation of a commercial heterocycle.
R1
S
N
N
N
R1
N
R2
R2
125
32
32
63
63
63
To explore the potential of this observation, we first
constructed a 3 · 3 array around this hit. Synthetically,
the S-alkylation reactions of commercially available
1,5-dihydro-4H-pyrazolo[3,4-d]pyrimidine-4-thione with
the requisite alkyl bromides were readily accomplished
in good yields (60–90%) using hydroxide resin in
DMF.27 The subsequent 1-N-alkylation reactions
(employing alkyl bromides with triethylamine or
K2CO3 in DMF) proved somewhat capricious, but
yields of the desired compounds varying from 25% to
80% were ultimately achieved. In spite of these issues,
the requisite compounds were produced and purified
in parallel on a 100 mg scale.28,29 Compounds from this
first iteration were assayed against M. tuberculosis
H37Rv; MIC data are recorded in Table 1.30
16
32
32
MIC 125 g/ml
μ
125 > MIC 32 g/ml
32 > MIC
μ
8 g/ml
μ
MIC < 8 g/ml
μ
These compounds are based around literature hit 5.25
R1=allyl, benzyl
11 substituents
Whilst in our hands compound
5 (R1 = benzyl;
R2 = propargyl) was marginally less active (MIC 32 lg/
ml vs 12.5), activity was established with all nine com-
pounds. In a second iteration, a set of 11 prospective S/
N-substituents was selected to construct a bigger array;
however, to lessen the synthetic burden, we employed
stepwise logic in our synthetic execution with the
11 · 11 matrix of possible combinations. Thus, of the
121 possible structures this presents, an initial sub-set
of 40 compounds was synthesized by restricting the
choice of either the S- or N-substituent to one of two
groups (allyl or benzyl) and exploiting all 11 options at
the second site (Fig. 2). These initial sets of compounds,
plus the mono-alkylated intermediates, were assayed to
identify preferred S-(R1) and N-(R2) substituents, with
a view to focusing SAR work in subsequent iterations,
without having to make all 121 elements of the matrix.
S
N
S
N
N
N
N
N
N
N
R2=allyl, benzyl
11 substituents
Figure 2. Strategy for development of initial SAR data.
ment from 5% at or below the 8 lg/ml threshold to 33%
in the second iteration. Pleasingly, one com-
pound (R1 = 4-chlorobenzyl; R2 = 3-pyridylmethylene)29
showed an MIC 62 lg/ml. Subsequent re-testing showed
this compound to have an MIC between 0.5 and 1 lg/ml.
In this exercise, we were able to substantiate the origi-
nally reported TPP as a genuine hit and improved on
it, whilst delineating some SAR. It is interesting to com-
pare the presence of electron poor aromatic rings in our
more active molecules with the reported QSAR study
based on electron withdrawing properties of substituents
on S-benzyl thiopyridine derivatives31 and alkylthio-
chloropyrimidines.32 Clearly, any future work should
first focus on S-benzyl analogues in order to explore
electron withdrawing substituents, their preferred posi-
tioning, thus enabling possible QSAR studies.
Compounds from this second iteration were also as-
sayed against M. tuberculosis H37Rv; MIC data are
recorded in Table 2.30 Analysis of the partially filled
rows and columns of this matrix enabled us to select a
focused set of 3 · 3 R1 and R2 groups for the second
iteration, displayed in the inner box of Table 2.
The results obtained provided evidence to vindicate our
approach. An examination of the proportion of com-
pounds with an MIC 68 lg/ml generated by the ‘blind
synthesis’ (main entries, Table 1) versus the ‘oriented syn-
thesis’ (central table entries, Table 2) shows an improve-
In order to substantiate these molecules as robust leads,
which might warrant a significant programme of medic-

1738
L. Ballell et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1736–1740
Table 2. MIC data for 4-thiopyrazolopyrimidinez derivatives against Mycobacterium tuberculosis H37Rv
R1
S
N
N
R1
N
N
R2
R2
Cl
CN
OH
N
N
Cl
H
250
125
250 250 125 250 250 125 125
63
250 250
250 125
32
16
8
63
16
250
63
32
125
16
32
16
32
32
32
16
32
250
32
32
16
32
Cl
125
250
CN
Cl
125
63
125 125
OH
Cl
Cl
63
32
16
8
1
63
32
8
CN
250
63
N
63
16
16
32
CN
63
63
8
N
MIC 125 μg/ml
125 > MIC 32 μg/ml
32 > MIC 8 μg/ml
MIC < 8 μg/ml
16
N
The central table entries represent compounds with R1 and R2 groups selected for a subsequent round of optimisation.

L. Ballell et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1736–1740
1739
Table 3. In silico prediction of parameters associated with bioavailability for all compounds possessing MICs <8 lg/ml
Com- H-bond H-bond
pound donors acceptors
Mw
ClogP Lipinski
Rotatable PSA Veber
ClogD at ClogD at Topliss Bioavailability
score
2
(A ) compliant? pH 7.4
˚
compliant? bonds
Yes
Yes
pH 6.5
(Topliss)
6
7
0
0
0
0
0
4
5
5
5
5
316.8 3.72
333.4 2.56
315.7 2.20
230.2 2.45
367.8 3.16
5
68.9 Yes
81.8 Yes
92.7 Yes
92.7 Yes
81.8 Yes
2.2
2.2
4.07
3.98
3.41
4.01
3.82
Class 4
Class 3
Class 3
Class 4
Class 3
5
2.56
3.72
3.16
3.11
2.56
3.72
3.16
3.11
8
Yes
Yes
Yes
5
5
5
9
10
Only 2 of the 18 Topliss assessment criteria calculated are included in the table.
References and notes
Table 4. Cytotoxicity and anti-mycobacterial activity of hit com-
pounds. Isoniazid (INH), Rifampicin (RIF) and Ethambutol (ETH)
were employed as controls in the MIC assays
1. World Health Organization. Global Tuberculosis Control:
Surveillance, Planning, Financing. WHO Report 2006.
Geneva, Switzerland, ISBN 92-4 156314-1.
Cl
Cl
Cl
2. Williams, B. G.; Dye, C. Science 2003, 301, 1535.
3. Corbett, E. L.; Watt, C. J.; Catherine, J.; Walker, N.;
Maher, D.; Williams, B. G.; Raviglione, M. C.; Dye, C.
Arch. Intl. Med. 2003, 163, 1009.
4. Septkowitz, A.; Raffalli, J.; Riley, T.; Kiehn, T. E.;
Armstrong, D. Clin. Microbiol. Rev. 1995, 8, 180.
5. Recent synthetic leads are reviewed in: Ballell, L.; Field,
R. A.; Duncan, K.; Young, R. J. Antimicrob. Agents
Chemother. 2005, 49, 2153.
CN
S
N
S
N
S
N
S
N
S
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
CN
(7)
(9)
(10)
(6)
(8)
N
N
6. Reviewed in: Duncan, K.; Barry, C. E. Curr. Opin.
Microbiol. 2004, 7, 1.
Compound
Tox 50 HepG2 (lg/mL)
MIC H37Rv (lg/mL)
7. Burchenal, J. H.; Murphy, M. L.; Ellison, R. R.; Sykes,
M. P.; Tan, T. C.; Leone, L. A.; Karnofsky, D. A.; Craver,
L. F.; Dargeon, H. W.; Rhoads, C. P. Blood 1953, 8, 965.
8. Elion, G. B. J. Med. Virol. 1993, 2.
9. Rundles, R. W. Arch. Intl. Med. 1985, 145, 1492.
10. Chiosis, G.; Timaul, M. N.; Lucas, B.; Munster, P. N.;
Zheng, F. F.; Sepp-Lorenzino, L.; Rosen, N. Chem. Biol.
2001, 8, 289.
6
>20
>40
2
8
7
8
8
8
9
10
4
8
>40
—
0.5–1
0.25
0.016
2.5
INH*
RIF*
ETH*
—
11. Crews, C. M.; Mohan, R. Curr. Opin. Chem. Biol. 2000, 4,
47.
12. Knockaert, M.; Gray, N.; Damiens, E.; Chang, Y.-T.;
Grellier, P.; Grant, K.; Fergusson, D.; Mottram, J.; Soete,
M.; Dubremetz, J.-F.; Le Roch, K.; Doerig, C.; Schultz, P.
G.; Meijer, L. Chem. Biol. 2000, 7, 411.
13. Laufer, S. A.; Domeyer, D. M.; Scior, T. R.; Albrecht, W.;
Hauser, D. R. J. Med. Chem. 2005, 48, 710.
14. Armstrong, J. I.; Portley, A. R.; Chang, Y.-T.; Nieren-
garten, D. M.; Cook, B. N.; Bowman, K. G.; Bishop, A.;
Gray, N. S.; Shokat, K. M.; Schultz, P. G.; Bertozzi, C. R.
Angew. Chem. Int. Ed. 2000, 39, 1303.
15. Verdugo, D. E.; Cancilla, M. T.; Ge, X.; Gray, N. S.;
Chang, Y.-T.; Schultz, P. G.; Negishi, M.; Leary, J. A.;
Bertozzi, C. R. J. Med. Chem. 2001, 44, 2683.
16. Wu, X.; Ding, S.; Ding, Q.; Gray, N. S.; Schultz, P. G.
J. Am. Chem. Soc. 2002, 124, 14521.
inal chemistry, we generated cytotoxicity data (Tox 50)
on the most potent compounds in a HepG2 cell line
and generated in silico predictions of the likely oral
pharmacokinetic profile. Encouragingly, three of the five
hit molecules were shown non-cytotoxic at the highest
concentration assayed (Table 4). The in silico predictors
also indicated a strong likelihood of good oral bioavail-
ability;35 through the generation of predictors based on
the calculated parameters described by Lipinski,33
Veber34 and Topliss35 (Table 3).
In summary, N-,S-di-alkyl 4-thio-1H-pyrazolo[3,4-d]py-
rimidines have been investigated as anti-mycobacterial
agents. The best compound exhibited activity in vitro
that is comparable to clinically successful drugs (e.g.,
ethambutol MIC 2 lg/ml) coupled with no cytotoxicity
against a HepG2 cell line. The SAR developed herein,
coupled with favourable pharmacokinetic predictions,
substantiate these molecules as a significant lead series
in the search for new anti-tubercular agents.
17. Gundersen, L. L.; Nissen-Meyer, J.; Spilsberg, B. J. Med.
Chem. 2002, 45, 1383.
18. Scozzafava, A.; Mastrolorenzo, A.; Supuran, C. T. Bioorg.
Med. Chem. Lett. 2001, 11, 1675.
ˇ
´
ˇ´
´
19. Klimesova, V.; Kocı, J.; Waisser, K.; Kaustova, J. Il
Farmaco 2002, 57, 259.
20. Pathak, A. K.; Pathak, V.; Seitz, L. E.; Suling, W. J.;
Reynolds, R. C. J. Med. Chem. 2004, 47, 273.
21. Hanke, J. H.; Gardner, J. P.; Dow, R. L.; Changelian, P.
S.; Brissette, W. H.; Weringer, E. J.; Pollok, B. A.;
Connelly, P. A. J. Biol. Chem. 1996, 271, 695.
22. Traxler, P.; Furet, P. Pharmacol. Ther. 1999, 82, 195.
23. Bergman, M. R.; Holycross, B. J. J. Pharmacol. Exp. Ther.
1996, 279, 247.
Acknowledgments
We thank the GlaxoSmithKline Action TB Initiative for
´
´
financial support and Marıa Jesus Almela for generating
the cytotoxicity data.
24. Ali, A.; Taylor, G. E.; Ellsworth, K.; Harris, G.; Painter,
R.; Silver, L. L.; Young, K. J. Med. Chem. 2003, 46, 1824.

1740
L. Ballell et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1736–1740
25. Moukha-Chafiq, O.; Taha, M. L.; Lazrek, H. B.; Bara-
scut, J.-L.; Imbach, J.-L. CR. Acad. Sci. Paris, Serie IIc,
Chimie 2000, 3, 639.
substituents. Such signals are not evident for the N-1-
substituted isomer.
29. All compounds showed at least P95% purity by HPLC
and had satisfactory MS and NMR data, for example,
ˆ
´
26. Adamec, J.; Waisser, K.; Kunes, J.; Kaustova, J.
Arch. Pharm. 2005, 338, 385, and references
herein.
Compound
10
4-{[(4-chlorophenyl)methyl]thio}-
1-(3-pyridinylmethyl)-1H-pyrazolo[3,4-d]pyrimidine: dH
(DMSO-d6) 8.86 (1H, s), 8.56 (1H, d, J = 1.8 Hz), 8.50
(1H, dd, J = 4.8, 1.8 Hz), 8.38 (1H, s), 7.63 (1H, m), 7.51
(2H, d, J = 8.1 Hz), 7.38 (2H, d, J = 8.1 Hz), 7.34 (1H, m),
5.69 (2H, s), 4.69 (2 H, s). dC (DMSO-d6) 164.6, 155.0,
149.8, 149.7, 137.0, 136.2, 132.9, 131.8, 129.2, 124.4, 48.4,
31.8. Mass spectrum, expected: C18H15ClN5S+ 368.0737,
found 368.0736.
27. Typical procedure for thiol alkylation: TPP (1 equiv) was
dissolved in dry DMF (10 ml/g of substrate), ion exchange
resin (Dowex AG 1-X8, OHꢀ form; 1 g per 100 mg of
substrate) was added and the mixture was stirred gently
for 5 min. The alkylating agent (1.2 mol equiv) was then
added and the mixture was stirred until LC–MS showed
the reaction to be complete.26 The reaction was quenched
by the addition of aqueous NH3 (5 mol equiv), filtered to
remove the resin and concentrated in vacuo. The crude
product obtained was preabsorbed onto silica (5 g) and
then eluted in series with a prepacked silica column, eluted
with a gradient of 0 ! 70% EtOAc in cyclohexane, to give
the S-alkylated product. Typical procedure for N-alkyl-
ation: The substrate (1 equiv) was dissolved in dry DMF
(10 ml/g of substrate), base (Et3N, 3 mol equiv, or K2CO3,
5 mol equiv) was added and the mixture was stirred for
5 min. The alkylating agent (3 mol equiv) was then added
and the mixture was heated at reflux under N2 for 16 h.
The reaction was quenched by the addition of aqueous
NH3 (5 mol equiv), filtered to remove the resin and
concentrated in vacuo. The bis-alkylated product was
purified as described above.
30. MIC (defined as the lowest concentration of compound
producing a >90% reduction in fluorescence observed in a
whole cell assay of M. tuberculosis) data were obtained in
Middlebrook’s 7H11 broth using a microtitre plate Ala-
mar Blue Assay (MABA) as described in: Collins, L.;
Franzblau, S. G. Antimicrob. Agents Chemother. 1997, 41,
1004.
ˇ
´
´
ˇ
31. Klimesova, V.; Palat, K.; Waisser, K.; Klimes, J. Int.
J. Pharm. 2000, 207, 1.
32. Agarwal, N.; Srivastava, P.; Raghuwanshi, S. K.; Upad-
hyay, D. N.; Sinha, S.; Shukla, P. K.; Ram, V. J. Bioorg.
Med. Chem. 2002, 10, 869.
33. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney,
P. J. Adv. Drug Del. Rev. 1997, 23, 3, The Lipinski ‘rule
of 5’ requires that no more than one of the following
criteria are met or exceeded: 5 Hydrogen-bond donors,
10 Hydrogen-bond acceptors, molecular weight 500,
ClogP 5.
28. The isomeric 2-alkyl-4-(alkylthio)-2H-pyrazolo[3,4-d]py-
rimidines were sometimes significant by-products in these
syntheses.
34. Veber, D. F.; Johnson, S. R.; Cheng, H.-Y.; Smith, B. R.;
Ward, K. W.; Kopple, K. D. J. Med. Chem. 2002, 45,
2615, This analysis suggests that molecules with high oral
bioavailability should have <10 rotatable bonds and a
polar surface area (PSA) of <140 A or <12 H-bond
donors and acceptors.
35. Yoshida, F.; Topliss, J. G. J. Med. Chem. 2000, 43, 2575,
The output from this model suggests four predicted classes
of F%, class 1, 620%; class 2, 20–49%; class 3, 50–79%;
class 4, P80%.
R1
S
N
2
˚
N
R2
N
N
2-alkyl-4-(alkylthio)-2H-pyrazolo[3,4-d]pyrimidine
The regioisomers are easily distinguishable by virtue of
NOE signals between the heterocycle H-3 proton and N-2