Synthesis and SAR studies of very potent imidazopyridine antiprotozoal agents

Article abstract of DOI:10.1016/j.bmcl.2006.01.092

Compounds 10a (IC₅₀ 110 pM) and 21 (IC₅₀ 40 pM) are the most potent inhibitors of Eimeria tenella cGMP-dependent protein kinase activity reported to date and are efficacious in the in vivo antiparasitic assay when administered to chi

Full text of DOI:10.1016/j.bmcl.2006.01.092

Bioorganic & Medicinal Chemistry Letters 16 (2006) 2479–2483
Synthesis and SAR studies of very potent imidazopyridine
antiprotozoal agents
Tesfaye Biftu,a,* Dennis Feng,a Michael Fisher,a Gui-Bai Liang,a Xiaoxia Qian,a
Andrew Scribner,c Richard Dennis,c Shuliang Lee,c Paul A. Liberator,b Chris Brown,b
Anne Gurnett,b Penny S. Leavitt,b Donald Thompson,b John Mathew,b Andrew Misura,b
Samantha Samaras,b Tamas Tamas,b Joseph F. Sina,d Kathleen A. McNulty,d
Crystal G. McKnight,d Dennis M. Schmatzb and Matthew Wyvratta
aMerck Research Laboratories, Department of Medicinal Chemistry, Merck and Co., Inc, PO Box 2000, Rahway, NJ 07065, USA
bHuman and Animal Infectious Disease Research, Merck and Co., Inc, PO Box 2000, Rahway, NJ 07065, USA
cScynexis, Inc, PO Box 12878, Research Triangle Park, NC 27709-2878, USA
dMerck Research Laboratories, Department of Genetic and Cellular Toxicology, West Point, PA 19486, USA
Received 29 December 2005; revised 20 January 2006; accepted 20 January 2006
Available online 7 February 2006
Abstract—Compounds 10a (IC50 110 pM) and 21 (IC50 40 pM) are the most potent inhibitors of Eimeria tenella cGMP-dependent
protein kinase activity reported to date and are efficacious in the in vivo antiparasitic assay when administered to chickens at 12.5
and 6.25 ppm levels in the feed. However, both compounds are positive in the Ames microbial mutagenesis assay which precludes
them from further development as antiprotozoal agents in the absence of negative lifetime rodent carcinogenicity studies.
Ó 2006 Elsevier Ltd. All rights reserved.
Human diseases such as malaria and toxoplasmosis are
caused by the Apicomplexan parasites Plasmodium falci-
parum and Toxoplasma gondii, respectively. Coccidiosis
is caused by the related Apicomplexans, the Eimeria spe-
cies of parasites.1 The intestinal pathology due to Eime-
ria infections in chickens is of major importance to the
poultry industry, leading to considerable economic loss-
es through morbidity and mortality. Polyether iono-
phore anticoccidials, discovered over thirty years ago,
have been successfully used as prophylactic agents to
combat coccidiosis. However, resistance to these and
other existing anticoccidials has been observed and
new therapeutic agents with novel mechanisms of action
are needed. Recently, we reported on a novel anticocci-
dial agent2 with potent in vitro and in vivo activity
against Eimeria parasites. It was determined that inhibi-
tion of parasite growth by these compounds was due to
the inhibition of parasite specific cGMP-dependent
protein kinase (PKG), a serine and/or threonine protein
kinase.3,4
In this paper, we describe the synthesis and structure–
activity relationships of very potent imidazopyridine
inhibitors of PKG activity, compounds which are more
active than our previously reported pyrrole analogs.2
Inhibition of native Eimeria tenella PKG enzyme activ-
ity was used as an initial screen for compound evalua-
tion. Second, compounds administered orally in feed
were ranked for anticoccidial activity using a seven
day in vivo efficacy model. A quantitative measure of
E. tenella and Eimeria acervulina oocyst shedding from
infected birds served as a measure of antiparasitic activ-
ity. Details of these procedures and rules of scoring have
been described.2
The synthesis of imidazopyridine analogs with various
substituents is outlined in Scheme 1. The ketone 3 was
made by coupling the anion generated from 4-methyl-2-
methylsulfanylpyrimidine 2 and lithium diisopropyla-
mide with methyl 4-fluorobenzoate in tetrahydrofuran.
The bromo ketone 4 was made by brominating ketone 3
with tetra-n-butylammonium tribromide in methylene
Keywords: Protein kinase G; Antiprotozoan; Anticoccidial;
Imidazopyridine.
*
Corresponding author. Tel.: +1 732 594 6514; e-mail: Tesfaye_
Biftu@merck.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.01.092

2480
T. Biftu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2479–2483
Scheme 1.
chloride. The imidazopyridine 5 was made by cyclization
of bromo ketones 4 with 2-amino-4-hydroxymethylpyri-
dine in refluxing ethanol. The primary alcohol 5 was
easily converted to the mesylate 6 and displaced with
dimethylamine to yield tertiary amine 7. Removal of the
methylthio group with Raney-nickel gave the pyrimidine
9. However, oxidation of 7 with hydrogen peroxide to the
sulfone 8 and displacement of the sulfone with ammonia
gave the amino pyrimidine 10a. Compounds 10b–f and
11, were made in a similar manner.

T. Biftu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2479–2483
2481
A different method of preparing imidazopyridines is
shown in Scheme 2. Oxime 14, made from a-chloro ke-
tone 12, was cyclized with a pyridine analog 15 in the
presence of potassium tert-butoxide to afford imidazo-
pyridine 16. Acylation at the 3-position of 16 with acetic
acid–sulfuric acid yielded 17, which was treated with
DMF–DMA to form enone 18. Finally 18 was reacted
with guanidine and sodium methoxide to afford the core
structure 19. When R0 is 4-CBz protected piperidine,
deprotection of the CBZ group by hydrogenolysis gave
20, which was then alkylated with formaldehyde and
sodium cyanoborohydride to give the desired product
21. Analogs 22–24 were made in a similar manner. Con-
version of alcohol 24 to the corresponding mesylate and
displacement with ammonia or methyl amine gave 25
and 26, respectively.
The mutagenesis assays were conducted using a plate
incorporation protocol.5,6 The mixture containing bac-
terial cells (108) in 2 mL molten agar, 0.1 mL of test arti-
cle in dimethylsulfoxide, and 0.5 mL cofactor/with or
without S9 mix was poured in triplicate onto plates con-
taining 25 mL of hardened agar. The liver S9 fraction
prepared from phenobarbital and b-naphthoflavone-
Cl
Cl
Br
NH2OMe.HCl
O
O
O
LiBr
N
N
F
F
F
12
13
14
R'
1. 14
2. t-BuOK
N
R'
N
16
N
F
Ac2O,
H2SO4
15
O
O
R'
R'
N
N
N
N
DMF / DMA
N
F
17
18
F
Guanidine HCl,
NaOMe
Acetamidine
CH3
NH2
N
N
N
N
R'
N
R'
N
N
N
22
19
F
F
Pd/C
H2
NCBz
R =
NH2
NH2
formaldehyde,
NaBH3CN,
AcOH
N
N
N
N
N
NH
N
N
N
N
F
F
20
21
Scheme 2.

2482
T. Biftu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2479–2483
Table 1.
Table 2.
R
NH2
N
N
N
N
R
N
N
N
N
N
F
F
Compound7
R
H
Ten_K Dose
IC50
(nM)
Et
Ea
(ppm) score score
Compound7
R
Ten_K
IC50 (nM) (ppm) score score
Dose
Et
Ea
20
21
N
H
0.05
25
0
0
9
10a
1.9
25
12.5
0
3
2
3
0
3
3
3
0
0
3
2
3
0
3
3
3
0
NH2
6.25
2.5
3
2
3
2
0.11
0.17
2.5
6.25
N
0.04
2.7
10b
10c
10d
10e
10f
11
NHMe
NMe2
NH–nPr
25
25
22
23
24
25
26
Cl
CN
25
0
0
0
0
0
0
0
0
0
0
0.13
12.5
12.5
25
11.3
2.7
25
25
25
25
NH–neopentyl 0.15
NH–Bn
OMe
CH2OH
CH2NH2
CH2NHMe
0.081
2.6
0.18
0.15
25
induced male Sprague–Dawley rats was present at
50 lL/plate. The plates were incubated at 37 °C for
48 h prior to counting colonies in an automated colony
counter. The number of revertant colonies is given as
mean standard deviation (n = 3).
which provide P80% reduction in oocyst production
are scored with a ‘3’, 50–79% reduction are scored a
‘2’, and those treatments with <50% reduction are rated
‘0’. In general, primary amines 10a and secondary
amines 10b in this position retain PKG activity (110
and 170 pM). By way of comparison, the tertiary amine
10c is an order of magnitude less potent. Since the neo-
pentyl 10e and benzyl 10f amines retain activity, steric
interaction is not very important in this position. How-
ever, hydrogen bonding could be quite important since
the hydrogen analog 9 and the methoxy analog 11 are
over an order of a magnitude less active as PKG inhib-
itors than the primary and secondary amines. As shown
in Table 2, a basic substituent in the 7-position of the
imidazopyridine is very important for PKG activity.
The data presented in Table 1 summarize the effect of
substituents in the 2-position of the pyrimidine ring on
PKG activity and in vivo efficacy. Inhibition of PKG en-
zyme activity (the Ten_K assay) is reported as the
amount of compound required to inhibit activity by
50% (IC50—nM). The antiparasitic activity of com-
pounds administered to chickens (parts per million in
feed) against E. tenella (Et) and E. acervulina (Ea) is
graded based on percent oocyst reduction. Treatments
Table 3.
Compound
lg/plate
Number of revertants/plate Salmonella typhimurium
E. coli WP2 uvrA
pKM101
TA100
TA1535
TA97a
TA98
10a
0
30
188.0 19.7
223.3 18.6
241.7 33.5
362.3 25.1
473.7 34.4
19.2 4.1
18.7 3.5
19.7 5.1
18.7 2.9
18.3 5.1
315.8 29.5
350.7 11.7
401.7 10.5
455.3 8.7
495.0 17.4
50.2 8.2
224.0 19.2
233.3 12.6
265.3 23.7
254.3 23.5
241.7 12.9
124.7 9.3
204.3 4.0
353.0 42.0
573.7 42.9
100
300
1000
2-Aminoanthracene
1
2
634.3 53.3
1428.0 78.0
86.7 7.1
188.0 7.2
750.7 40.3
1435.0 51.3
364.0 65.8
1068.7 108.2
449.3 7.4
1450.0 78.6
21
0
30
172.3 14.4
215.0 22.3
235.0 11.8
363.0 33.8
562.0 42.8
26.0 5.5
23.0 2.0
21.7 10.6
23.7 3.5
23.0 4.0
252.1 27.9
294.7 7.4
289.3 14.3
298.0 19.7
333.0 38.2
37.4 6.6
54.0 6.0
61.3 20.5
67.7 7.8
141.7 38.1
152.8 12.0
156.7 4.6
168.7 26.0
162.7 10.6
145.0 10.6
100
300
1000
2-Aminoanthracene
1
2
383.7 11.6
732.7 42.3
62.0 6.6
112.0 20.1
528.0 53.2
827.7 27.5
278.0 27.7
563.7 43.7
411.0 51.2
1122.7 37.9
Compounds 10a and 21 are positive in two of the bacterial strains (TA98 and TA100) that comprise the Ames microbial mutagenesis assay. A
twofold or greater increase in the number of revertants relative to the solvent control is considered positive (positive data points are highlighted in
bold font). All assays described in this table were conducted with an S9 protein fraction. Experimental compound 10a (as well as the S9-dependent
positive control 2-aminoanthracene) is not positive in the assay without metabolic activation provided by the S9 fraction, while compound 21 did give
a 2.0- to 2.1-fold increase in the number of revertants at 10–1000 mg/plate in the TA98 strain without S9 (data not included). Both compounds have
antibacterial activity at levels of 3000 and 6000 lg/plate (data not included).

T. Biftu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2479–2483
2483
Substituents such as chlorine 22, cyano 23, or alcohol 24
are much less active as inhibitors of PKG activity than
the amines 20, 21, 25, and 26.
also fully active in in vivo assays as anticoccidial agents
at 6.25 to 12.5 ppm level in feed. However, the potential
genotoxicity of these compounds precludes them from
further development.
For active PKG inhibitors, the in vivo efficacy is deter-
mined by many factors (including absorption in both
host and parasites) that could be affected by the basicity
and lipophilicity of the compounds. In general, for com-
pounds listed in Tables 1 and 2, the in vivo activity
tracks with the PKG activity fairly well. Potent PKG
inhibitors 10a,b,d–f, and 21 are active in in vivo assays
when administered to birds in feed. The most potent
analogs are 10a and 21. These compounds are active
in feed at a level of 12.5 and 6.25 ppm, respectively.
Commercial coccidiostats such as salinomycin and am-
prol are active at 66 and 130 ppm, respectively. These
imidazopyridines are also active against other protozoan
parasites, including the causative agents of malaria and
toxoplasmosis (manuscripts in preparation). Due to
their potent in vivo anticoccidial activity, both 10a and
21 have been advanced for further evaluation. Unfortu-
nately both compounds are positive in the Ames micro-
bial mutagenesis assay (Table 3). A twofold or greater
increase in the number of revertants is reproducibly seen
for both compounds in two (TA98 and TA100) of the
five standard bacterial strains that comprise the full
Ames assay. Of note, both compounds are positive after
metabolic activation, whereas compound 21 was weakly
positive without S9 activation in the TA98 strain (data
not included). Due to the potential for genotoxicity pre-
dicted by the Ames assay, these compounds will not be
developed for human or animal use in the absence of
negative lifetime rodent carcinogenicity studies.
Acknowledgment
The authors thank Dr. Phil Eskola, Dr. Gerard
Kieczykowshi, Ms. Amanda Makarewicz, Mr. Bob
Frankshun, and Mr. Glen Reynolds for scale-up of
key intermediates, Drs. Ann Weber, Sheila Galloway,
and Vijay Reddy for review of the manuscript, and
Merck Research Laboratory Basic Chemistry
Analytical Support Group for open access service
on LC/MS.
References and notes
1. Williams, R. B. Int. J. Parasitol. 1999, 29, 1209.
2. Biftu, T.; Feng, D.; Liang, G.-B.; Qian, X.; Gurnett, A. M.;
Liberator, P. A.; Dulski, P. M.; Donald, R. G. K.; Diaz, C.
A.; Darkin-Rattray, S. J.; Nare, B.; Crumley, T.; Blum, P.
S.; Misura, A. S.; Tamas, T.; Wyvratt, M.; Fisher, M. H.
Bioorg. Med. Chem. Lett. 2005, 15, 3296.
3. Gurnett, A. M.; Liberator, P. A.; Dulski, P. M.; Salowe, S.
P.; Donald, R. G. K.; Anderson, J. W.; Wiltsie, J.; Diaz, C.
A.; Harris, G.; Chang, B.; Darkin-Rattray, S. J.; Nare, B.;
Crumley, T.; Blum, P. S.; Misura, A. S.; Tamas, T.;
Sardana, M. K.; Yuan, J.; Biftu, T.; Schmatz, D. M. J. Biol.
Chem. 2002, 277, 15913.
4. Donald, R. G. K.; Liberator, P. A. Mol. Biochem.
Parasitol. 2002, 180, 136.
5. Matshushima, T.; Takamoto, Y.; Shirai, A.; Sawamura,
M.; Sugimura, T. Prog. Mutat. Res. 1981, 1, 387.
6. Maron, D. M.; Ames, B. N. Mutat. Res. 1983, 113, 173.
7. All new compounds were characterized by 1H NMR
and LC–MS prior to submission for biological
evaluation.
Novel imidazopyridine analogs were found to be potent
inhibitors of parasite PKG activity. The most potent
compounds are the dimethyl amine 10a and the
N-methyl piperidine 21 analogs. These compounds were