Development of an Orally Available and Central Nervous System (CNS) Penetrant Toxoplasma gondii Calcium-Dependent Protein Kinase 1 (TgCDPK1) Inhibitor with Minimal Human Ether-a-go-go-Related Gene (hERG) Activity for the Treatment of Toxoplasmosis

Article abstract of DOI:10.1021/acs.jmedchem.6b00760

New therapies are needed for the treatment of toxoplasmosis, which is a disease caused by the protozoan parasite Toxoplasma gondii. To this end, we previously developed a potent and selective inhibitor (compound 1) of Toxoplasma gondii calcium-dependent protein kinase 1 (TgCDPK1) that possesses antitoxoplasmosis activity in vitro and in vivo. Unfortunately, 1 has potent human ether-a-go-go-related gene (hERG) inhibitory activity, associated with long Q-T syndrome, and consequently presents a cardiotoxicity risk. Here, we describe the identification of an optimized TgCDPK1 inhibitor 32, which does not have a hERG liability and possesses a favorable pharmacokinetic profile in small and large animals. 32 is CNS-penetrant and highly effective in acute and latent mouse models of T. gondii infection, significantly reducing the amount of parasite in the brain, spleen, and peritoneal fluid and reducing brain cysts by >85%. These properties make 32 a promising lead for the development of a new antitoxoplasmosis therapy.

Full text of DOI:10.1021/acs.jmedchem.6b00760

Subscriber access provided by UNIV OF CAMBRIDGE
Article
Development of an Orally Available and Central Nervous System
CNS)-Penetrant Toxoplasma gondii calcium-dependent protein
(
kinase 1 (TgCDPK1) Inhibitor with Minimal Human Ether-à-go-go-
Related Gene (hERG) Activity for the Treatment of Toxoplasmosis
Rama Subba Rao Vidadala, Kasey L. Rivas, Kayode K Ojo, Matthew A. Hulverson, Jennifer
A. Zambriski, Igor Bruzual, Tracey L. Schultz, Wenlin Huang, Zhongsheng Zhang, Suzanne
Scheele, Amy E. DeRocher, Ryan Choi, Lynn K Barrett, Latha Kallur Siddaramaiah, Wim G.J. Hol,
Erkang Fan, Ethan A Merritt, Marilyn Parsons, Gail Freiberg, Kennan Marsh, Dale Kempf, Vern
B. Carruthers, Nina Isoherranen, J. Stone Doggett, Wesley C. Van Voorhis, and Dustin J. Maly
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00760 • Publication Date (Web): 16 Jun 2016
Downloaded from http://pubs.acs.org on June 17, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Development of an Orally Available and Central Nervous System
(
CNS)-Penetrant Toxoplasma gondii calcium-dependent protein
kinase 1 (TgCDPK1) Inhibitor with Minimal Human Ether-à-go-go-
Related Gene (hERG) Activity for the Treatment of Toxoplasmosis
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
§
‡
‡
‡
Rama Subba Rao Vidadala, Kasey L. Rivas, Kayode K. Ojo, Matthew A. Hulverson,
€
¶
&
†
Jennifer A. Zambriski, Igor Bruzual, Tracey L. Schultz, Wenlin Huang, Zhongsheng
†
#
#
‡
‡
Zhang, Suzanne Scheele, Amy E. DeRocher, Ryan Choi, Lynn K. Barrett, Latha
†
†
†
†
Kallur Siddaramaiah, Wim G. J. Hol, Erkang Fan, Ethan A. Merritt, Marilyn
#
,^
λ
λ
λ
&
Parsons, Gail Freiberg, Kennan Marsh, Dale Kempf, Vern B. Carruthers, Nina
¶
‡,^*
§,†*
∥
Isoherranen, J. Stone Doggett, Wesley C. Van Voorhis, and Dustin J. Maly
†
Department of Biochemistry, University of Washington, Seattle, WA 98195, United
States
‡
Department of Medicine, Division of Allergy and Infectious Diseases, and the Center for
Emerging and Reꢀemerging Infectious Diseases (CERID), University of Washington,
Seattle, WA 98109, United States
§
Department of Chemistry, University of Washington, Seattle, WA 98195, United States
#
Center for Infectious Disease Research (formerly Seattle Biomedical Research Institute),
Seattle, WA 98109, United States
^
Department of Global Health, University of Washington, Seattle, WA 98195 United
States
ACS Paragon Plus Environment

Page 3 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
¶
Portland VA Medical Center, Portland, OR 97239, United States
€
Paul G. Allen School for Global Animal Health, College of Veterinary Medicine,
Washington State University, Pullman, WA 99164, United States
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
&
Department of Microbiology and Immunology, University of Michigan Medical
School, Ann Arbor, MI 48109 United States
∥Department of Pharmaceutics, University of Washington, Seattle, WA 98195, United
States
λAbbVie, N. Chicago, Illinois 60064, United States
Correspondence should be addressed to: DJM (Tel: 206ꢀ543ꢀ1653. Fax 206ꢀ685 7002. Eꢀ
mail: maly@chem.washington.edu) or WCVV (Tel: 206ꢀ543ꢀ2447. Fax: 206ꢀ616ꢀ4898.
Eꢀmail: wesley@uw.edu)
Abstract: New therapies are needed for the treatment of toxoplasmosis, which is a
disease caused by the protozoan parasite Toxoplasma gondii. To this end, we previously
developed a potent and selective inhibitor (compound 1) of Toxoplasma gondii calciumꢀ
dependent protein kinase 1 (TgCDPK1) that possesses antiꢀtoxoplasmosis activity in vitro
and in vivo. Unfortunately, 1 has potent human EtherꢀàꢀgoꢀgoꢀRelated Gene (hERG)
inhibitory activity, associated with long QꢀT syndrome, and, consequently, presents a
cardiotoxicity risk. Here, we describe the identification of an optimized TgCDPK1
inhibitor 32, which does not have a hERG liability and possesses a favorable
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
pharmacokinetic profile in small and large animals. 32 is CNSꢀpenetrant and highly
effective in acute and latent mouse models of T. gondii infection, significantly reducing
the amount of parasite in the brain, spleen, and peritoneal fluid and reducing brain cysts
by >85%. These properties make 32 a promising lead for the development of a new antiꢀ
toxoplasmosis therapy.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Introduction
Toxoplasmosis is an infectious disease that results from the infection of the protozoan
1
parasite Toxoplasma gondii. While cats are the definitive host of T. gondii, this zoonotic
parasite is estimated to be one of the most prevalent in humans, with approximately 30ꢀ
2
5
0% of the world’s population being seropositive. T. gondii infections of humans can be
caused by the consumption of undercooked meat containing tissue cysts or by the
1
,3
accidental ingestion or inhalation of oocysts from cat feces. T. gondii parasites rapidly
proliferate during acute infection before being controlled by the immune system in
4
immunocompetent individuals. However, despite this robust immune response, the T.
gondii parasite is not eliminated. Instead, the parasite transforms into a slowly dividing
cellular stage called the bradyzoite, which cluster into tissue cysts mainly located in the
5
brain, eyes, and striated muscle. Tissue cysts can persist for the lifetime of the host;
periodically rupturing and releasing tachyzoites that can again rapidly replicate and
reform cysts. While in immunocompetent individuals toxoplasmosis is usually
asymptomatic and recovery occurs in the absence of antibiotic treatment, T. gondii
infection can sometimes trigger severe retinal diseases, which has been recently
6,7
recognized to be particularly prevalent in regions of South America. When T. gondii
infection occurs during pregnancy, vertical transmission can ensue, resulting in birth
ACS Paragon Plus Environment

Page 5 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
8
defects or miscarriage. For individuals with compromised immune systems, active T.
gondii infection primarily leads to encephalitis or chorioretinitis, but may cause disease
9
in other organs as well.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Current firstꢀline therapy for toxoplasmosis involves treatment with a
combination of pyramethamine and sulfadiazine, which inhibit the folate pathway in T.
10
gondii. However, this therapeutic regimen requires long dosing periods and can be
1
1
limited by rash, leukopenia, nephrotoxicity, teratogenicity, and kernicterus in newborns.
There is a longstanding need for better toxoplasmosis drugs, especially therapeutic agents
that are nonꢀtoxic, nonteratogenic, and with simple dosing profiles.
In order to develop new therapies for the treatment of toxoplasmosis, we and
others have explored strategies that involve inhibiting enzymatic components of calciumꢀ
dependent cellular processes in the parasite, which are necessary for invasion into and
egress from host cells. T. gondii calciumꢀdependent protein kinase 1 (TgCDPK1) is a
1
2
particularly promising target due to its role in gliding and motility. Gliding and motility
are necessary for T. gondii invasion and egress, and as T. gondii is an obligate
intracellular parasite that requires these processes to proliferate, inhibition of TgCDPK1
represents a promising new antiꢀparasitic strategy. Indeed, genetic or pharmacological
disruption of TgCDPK1 function blocks parasite growth in vitro and is able to prevent
13ꢀ20
and treat established T. gondii infections in mice.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
B
Figure 1. Inhibitor 1. (A) The chemical structure of compound 1. Pyrazolopyrimidine scaffold
numbering and sites of inhibitor derivatization (R and R ) are shown. (B) Schematic of how 1
interacts with the ATPꢀbinding site of TgCDPK1.
1
2
We previously developed a number of highly promising pyrazolopyrimidine (PP)ꢀ
based ATPꢀcompetitive inhibitors that are able to potently block the enzymatic activity of
TgCDPK1. These selective TgCDPK1 inhibitors are able to block the invasion of T.
1
5,19,20
gondii into host cells, preventing proliferation of the parasite.
Recently, we have
demonstrated that the TgCDPK1 inhibitor 1 (1294) is an orally available inhibitor that
possesses antiꢀtoxoplasmosis activity in vitro and in vivo, with no apparent signs of
15
toxicity. Furthermore, 1 was able to block vertical transmission of Neospora caninum in
ACS Paragon Plus Environment

Page 7 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a pregnant mouse model, and since N. caninum is closely related to T. gondii, it speaks to
2
1,22
a potential therapeutic application for toxoplasmosis in pregnancy.
Unfortunately, we
have found that 1 has potent human EtherꢀàꢀgoꢀgoꢀRelated Gene (hERG) inhibitory
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
23
activity –associated with long QꢀT syndrome–and, thus, presents a significant
2
4
cardiotoxicity risk in humans. Here, we describe the identification of optimized PPꢀ
based TgCDPK1 inhibitor 32, which does not have an hERG liability but possesses the
favorable antiꢀparasitic activity of lead compound 1.
Results and Discussion
Molecular Design and Synthesis
TgCDPK1 contains a Cꢀterminal calciumꢀbinding domain of 4 EFꢀhands
(
CDPK activation domain) linked to an Nꢀterminal kinase domain. The ATPꢀbinding cleft
of TgCDPK1, which is the site of interaction with PPꢀbased inhibitors, is similar in
structure to mammalian protein kinases with the exception that it contains a rare glycine
residue at the gatekeeper position. The starting point for this study was compound 1,
which is based on a PP scaffold that forms two hydrogenꢀbonding interactions with the
hinge region of TgCDPK1 (Figure 1). The Cꢀ3 (R ) position of 1 is derivatized with a
1
bulky 6ꢀethoxyꢀ2ꢀnaphthyl group that occupies a hydrophobic pocket adjacent to
TgCDPK1’s glycine gatekeeper residue. As glycine is a very uncommon gatekeeper
residue, the space occupied by the 6ꢀethoxyꢀ2ꢀnaphthyl group provides a means to obtain
high selectivity over mammalian kinases, which often have larger gatekeepers that
constrain access to the hydrophobic back pocket. The 6ꢀalkoxyꢀ2ꢀnaphthyl substitution
pattern appears to be optimal for complementing the hydrophobic gatekeeper pocket of
TgCDPK1 and enhancing overall inhibitor selectivity. At the Nꢀ1 position, 1 contains an
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Nꢀmethylꢀ4ꢀpiperidinemethylene group, directed toward the sugar pocket. This basic
group increases potency for TgCDPK1 and, in combination with 6ꢀalkoxyꢀ2ꢀnaphthyl
substituents at the Cꢀ3(R ) position, the 4ꢀpiperidinemethyl group greatly enhances
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
25
kinase selectivity.
Several synthetic strategies were used to generate derivatives of 1. The synthetic
route shown in Scheme 1 was used to generate inhibitors that contain a 6ꢀethoxyꢀ2ꢀ
naphthyl group at the R position and various 4ꢀpiperidinemethylene or 3ꢀ
1
azetidinemethylene derivatives at the R position (Table 1). Inhibitors that contain an Nꢀ
2
methylꢀ4ꢀpiperidinemethylene R group and various 6ꢀalkoxyꢀ2ꢀnaphthyl or 6ꢀalkoxyꢀ2ꢀ
2
quinolinyl groups at the R position were generated by either alkylation of tꢀbutylꢀ4ꢀ((4ꢀ
1
aminoꢀ3ꢀ(6ꢀhydroxynaphthalenꢀ2ꢀyl)ꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ1ꢀ
yl)methyl)piperidineꢀ1ꢀcarboxylate, followed by Boc deprotection and reductive
alkylation, or by palladiumꢀcatalyzed SuzukiꢀMiyaura couplings between appropriate
aryl boronic acids and boronate esters and tꢀbutylꢀ4ꢀ((4ꢀaminoꢀ3ꢀiodoꢀ1Hꢀpyrazolo[3,4ꢀ
d]pyrimidinꢀ1ꢀyl)methyl)piperidineꢀ1ꢀcarboxylate, followed by Boc deprotection and
reductive alkylation (Scheme 2). Boronic esters and boronic acids that are not
2
6
commercially available were prepared from aryl bromides using standard methods.
Compounds containing variable groups at the R position were generated by alkylating 3ꢀ
2
iodoꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4ꢀamine with alkyl halides, alkyl mesylates, or
expoxides followed by SuzukiꢀMiyaura coupling to R ꢀboronic acids or boronate pinacol
1
esters (Scheme 3).
Inhibitor Testing Cascade
ACS Paragon Plus Environment

Page 9 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A multiꢀassay testing cascade was used to determine if newly generated analogs
demonstrated superior antiꢀparasitic properties compared to compound 1. Compounds
were first tested for the ability to block the activity of recombinant TgCDPK1 using a
luminescenceꢀbased assay that measures the consumption of ATP in the presence of the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
9
peptide substrate Syntide 2. All assays were performed in the presence of 2 mM CaCl
and at an ATP concentration–10 ꢀM–equal to TgCDPK1’s Michaelis constant (Km).
Compounds of sufficient in vitro enzyme inhibitory activity were then tested for their
2
ability to block the invasion and growth of parasites into human foreskin fibroblast cells
19
using a previously described T. gondii cellular assay. Inhibitors with comparable in
vitro enzymatic and cellular potencies as 1, were next screened for selectivity in two
assays. To determine selectivity for TgCDPK1 over mammalian kinases, compounds
were tested against the tyrosine kinase Src, which is representative of the most likely off
targets of pyrazolopyrimidineꢀbased inhibitors that contain a large aryl group at the Cꢀ3
27
(
R ) position due to its small threonine gatekeeper residue. General off target toxicity
1
was determined by testing for growth inhibition (GI ) of two mammalian cell lines,
5
0
HepG2 (liver) and CRLꢀ8155 (lymphoblast). As elimination of hERG activity was a
major goal of this study, all compounds with favorable profiles were tested for hERG
2
3,28,29
activity using a previously described assay.
Table 1. Enzymatic TgCDPK1 and Src inhibition (IC ), hERG inhibition (IC ), HepG2 and
CRLꢀ8155 growth inhibition (GI ) and T. gondii cellular assay (EC ) results for compounds
with variable R substructures (1ꢀ9). IC and GI values are the averages of at least three assays.
T. gondii cellular assay EC values are the average of at least two runs that were performed in
5
0
50
50
50
2
50
50
50
triplicate.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
NH2
N
N
N
N
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
R2
IC50 (ꢀM)
TgCDPK1 Src
human cells, GI50 (ꢀM) EC50 (ꢀM)
HepG2 CRLꢀ8155 T. gondii
R2
hERG
N
1
5
0.0029
0.0032
>10
>10
0.40
>10
>10
>10
>10
0.14
NH
1.7
0.283
N
N
6
7
0.0060
0.010
>10
>10
1.4
>10
>10
>10
>10
0.41
0.51
0.30
N
8
9
O
0.011
0.004
>10
1.1
>10
>10
>10
>10
>10
>10
0.22
0.48
O
N
In generating analogs of 1, we felt that it would be important to retain a 6ꢀ
alkoxynaphthalenꢀ2ꢀyl group at the Cꢀ3 (R ) position. In addition, due to the unexpected
1
25
contribution of the 4ꢀpiperidinemethyl group to inhibitor potency and selectivity,
compounds that contain a similar substituent at the R position were focused on first.
2
Results for compounds containing a fixed 6ꢀethoxynaphthyl at the R position and
1
various 4ꢀpiperidinemethyl or 3ꢀazetidinemethyl groups at the R position are shown in
2
Table 1. Increasing the size of the Nꢀalkyl group on the 4ꢀpiperidinemethyl moiety (7)
resulted in slightly reduced target potency and did not diminish hERG inhibition.
Replacing the 1ꢀmethylpiperidinꢀ4ꢀylmethyl group of 1 with either a 3ꢀazetidinemethyl
(5) or a 1ꢀmethylꢀ3ꢀazetidinemethyl (6) group resulted in potent inhibition of TgCDPK1,
albeit with attenuated activity against parasite proliferation, but only a slight decrease in
ACS Paragon Plus Environment

Page 11 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
hERG inhibition. However, an analog containing an acetylated 4ꢀpiperidinemethyl group
(inhibitor 8) displayed acceptable levels of inhibition of the TgCDPK1 enzyme and
parasite proliferation with no hERG activity. Inhibitor 9, which contains a 1ꢀacetylꢀ3ꢀ
azetidinemethylene group instead of a 1ꢀacetylꢀ4ꢀpiperidinemethyl moiety, also lacks
hERG activity but possesses unacceptably low activity in the T. gondii cellular assay.
These results are consistent with the observed correlation between inhibitor basicity and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
0ꢀ32
hERG inhibition.
We next explored whether hERG inhibition could be reduced in analogs of 1 that
maintain the 1ꢀmethylꢀ4ꢀpiperidinemethyl group by modifying the Nꢀ1 position (R ) of
2
the pyrazolopyrimidine scaffold (Table 2). For the 6ꢀalkoxynaphthalenꢀ2ꢀyl series,
inhibitors containing cycloalkyloxy groups larger than ethyl (10ꢀ12) were potent
TgCDPK1 inhibitors and were of reasonable potency in the T. gondii cellular assay.
However, the hERG activity of these compounds was comparable to 1. Unfortunately,
introduction of more polar groups at the 6 position of the naphthyl ring (13ꢀ16), which
reduces overall inhibitor lipophilicity and would be expected to diminish hERG activity,
led to less potent inhibition of TgCDPK1 and subsequent reduced potency in the T.
gondii cellular assay. A more favorable inhibitor profile was obtained by replacing the 6ꢀ
ethoxyꢀ2ꢀnaphthyl group of 1 with a 2ꢀethoxyquinolinꢀ6ꢀyl moiety (17). Inhibitor 17
showed comparable levels of TgCDPK1 enzyme inhibition and efficacy in the invasion
assay as 1, with no detectable Src inhibition or mammalian cell toxicity. Compound 17
also possesses 25ꢀfold lower activity against hERG than 1. However, replacing the ethyl
group of 17 with more hydrophobic substituents, like trifluoroethyl (18) or cyclopropyl
(19), led to unacceptable levels of hERG inhibition (Table 2).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 2. Enzymatic TgCDPK1 and Src inhibition (IC ), hERG inhibition (IC ), HepG2 and
5
0
50
CRLꢀ8155 growth inhibition (GI ) and T. gondii cellular assay (EC ) results for compounds
50
50
with variable R substructures (10ꢀ19). IC and GI values are the averages of at least three
1
50
50
assays. T. gondii cellular assay EC values are the average of at least two runs that were
5
0
performed in triplicate. Compounds that did not show an IC < 0.010 ꢀM for TgCDPK1 or an
50
IC > 5 ꢀM for hERG were deꢀprioritized for further testing.
5
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
NH2
R1
N
N
N
N
N
IC50 (ꢀM)
human cells, GI50 (ꢀM) EC50 (ꢀM)
R
1
TgCDPK1 Src hERG HepG2 CRLꢀ8155
T. gondii
O
1
0.0029 >10 0.40
0.0040 >10 0.90
>10
>10
>10
>10
0.14
O
1
0
0.61
O
11
12
0.0025 >10 0.20
0.0015 >10 0.37
>10
ꢀ
>10
>10
0.32
0.12
O
O
O
13
14
0.015
0.010
>10
>10
ꢀ
ꢀ
ꢀ
ꢀ
>10
>10
ꢀ
>10
>10
ꢀ
0.71
O
ꢀ
ꢀ
O
O
OH
1
5
0.046
0.024
OH
O
1
1
6
7
>10
ꢀ
>10
>10
>10
>10
ꢀ
N
O
0.0061 >10 ~10
0.27
N
O
CF
3
1
8
0.0022
ꢀ
0.50
1.7
ꢀ
ꢀ
0.33
N
O
19
0.0016 >10
>10
-
0.098
-Not tested
Finally, we investigated whether the cyclic amine substituent at the R position of
2
1
can be replaced with another group that confers similar potency, selectivity, and
solubility. We have previously shown that inhibitors possessing R groups that are
2
connected to the Nꢀ1 position of the PP scaffold through a methylene linkage demonstrate
the highest degree of selectivity for TgCDPK1 over mammalian kinases with small
ACS Paragon Plus Environment

Page 13 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
25
gatekeepers, like the tyrosine kinases Src and Abl. Therefore, a panel of analogs with
variable R groups, which extend into the riboseꢀbinding pocket of the ATPꢀbinding site
2
was generated (Table 3). In general, inhibitors of this class are potent inhibitors of the
TgCDPK1 enzyme and efficacious in the T. gondii cellular assay. Furthermore, all of the
inhibitors in Table 3 are highly selective (>150ꢀfold) for TgCDPK1 over Src. Consistent
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
with basicity at the R position playing a major role in hERG inhibition, inhibitor 20,
2
which contains a dimethylamino group at the R position, has a similar level of hERG
2
activity as 1. Compounds 21ꢀ23, which all contain an isobutyl group at the R position,
2
are potent inhibitors of the TgCDPK1 enzyme and are at least as effective as 1 against the
parasite. Importantly, these compounds are >3500ꢀfold selective for TgCDPK1 over Src,
demonstrate no detectable growth inhibition of HepG2 and CRLꢀ8155 cells, and lack
hERG activity. Compounds 24ꢀ29, which all contain a cycloalkyl ether group linked to
the Nꢀ1 (R ) position through a methylene, are also effective inhibitors of TgCDPK1 and
2
parasite proliferation. All of these compounds, except 28, have an IC >10 ꢀM for Src
5
0
activity and do not inhibit the growth of mammalian cells. Furthermore, all of the
inhibitors in this series, except 24 (IC = 1.9 ꢀM), have an IC of greater than 10 ꢀM
5
0
50
for hERG activity. Inhibitors 30ꢀ36, which either contain a 2ꢀmethylpropanꢀ2ꢀol or a 2,2ꢀ
dimethylpropanꢀ1ꢀol group at the R position also possess desirable profiles: potent
2
TgCDPK1 enzyme inhibition, effective inhibition of parasite proliferation, and no
detectable offꢀtarget inhibition or toxicity.
Table 3: Enzymatic TgCDPK1 and Src inhibition (IC ), hERG inhibition (IC ), HepG2 and
5
0
50
CRLꢀ8155 growth inhibition (GI ) and T. gondii cellular assay (EC ) results for compounds
with variable R and R substructures (20ꢀ37). IC and GI values are the averages of at least
three assays. T. gondii cellular assay EC values are the average of at least two runs that were
50
50
1
2
50
50
5
0
performed in triplicate.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
NH2
R1
N
N
N
R2
N
IC50 (ꢀM)
TgCDPK1 Src hERG
human cells, GI50 (ꢀM) EC50 (ꢀM)
HepG2 CRLꢀ8155 T. gondii
R2
R
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
N
1
0.0029
0.0045
>10 0.40
>10 0.082
>10
>10
>10
>10
0.14
0.26
N
O
N
20
O
2
1
2
0.00080 >10 >10
>10
ꢀ
>10
>10
0.12
O
2
0.0011
0.0012
3.7
>10
0.071
N
O
O
23
>10 >10
>10
>10
0.090
2
4
5
0.0021
0.0018
>10 1.9
>10 >10
>10
>10
>10
>10
0.045
0.21
O
N
O
2
O
2
6
7
0.0020
0.0020
>10 >10
>10 >10
>10
>10
>10
>10
0.076
0.078
N
O
2
O
O
28
0.0024
0.37 >10
>10 >10
>10
>10
0.077
N
O
29
30
31
0.0018
0.0056
0.0033
>10
>10
>10
>10
>10
>10
0.081
0.75
0.17
O
>10
ꢀ
N
O
>10 >10
OH
O
32
33
0.001
0.002
>10 >10
>10 >10
>10
>10
>10
>10
0.060
0.16
N
O
O
3
4
5
0.00090
0.0013
3.1
7.4
>10
>10
>10
>10
>10
>10
0.056
0.018
N
N
O
O
O
3
OH
CF
3
36
0.017
0.094
>10
>10
ꢀ
ꢀ
>10
>10
>10
>10
0.70
ꢀ
CF
3
CF
3
37
OH
-Not tested
Pharmacokinetic Profiling of Potent and Selective TgCDPK1 Inhibitors
ACS Paragon Plus Environment

Page 15 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Based on the abilities of a number of inhibitors to potently block T. gondii
proliferation in mammalian cells, while demonstrating little or no offꢀtarget toxicity or
hERG inhibition, we next prioritized compounds based on their solubility in water and
pharmacokinetic (PK) properties (Table 4). The aqueous solubility of inhibitors 8, 17, 29,
31ꢀ34, and 35 was determined at pH = 6.5. The initial PK profiles of these inhibitors were
determined after a single 10 mg/kg oral dose in three Balb/c mice, with sampling
conducted at the time points indicated in the Experimental Section. As a reference, after a
10 mg/kg per os (PO) dose, lead inhibitor 1 demonstrated a maximum concentration
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
Cmax) of 0.75 ±0.15 µM and total exposure (area under the plasma concentrations versus
time curve, AUC) of 430 ±84 µM·min. Inhibitors 29, 34, and 35 were poorly soluble (<20
M) relative to 1 (solubility = 82 ꢀM) and all three compounds had lower exposures and
ꢀ
higher clearances than 1. Inhibitors 8 and 17 demonstrated better solubility but also had a
lower exposure and higher clearance than 1, which is likely due to the more efficient
metabolism of these compounds. In contrast, inhibitors 31ꢀ33, which all contain a 2ꢀ
methylpropanꢀ2ꢀol at the R position, were highly soluble and 32 and 33 reached
2
maximum serum concentrations (Cmax) >10ꢀfold higher and total exposure >30ꢀfold
higher than 1 (Table 4 and Figure 2). Compound 31 was absorbed rapidly with a Tmax of
5
0 min and a Cmax nearly 10ꢀfold higher than 1. Yet, the overall exposure of 31 was only
twice as high as 1 and much lower than 32 and 33. Based on this mouse pharmacokinetic
data, compounds 32 and 33 were selected for further pharmacokinetic and tolerability
studies.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 4. Solubility (in phosphaseꢀbuffered saline, pH= 6.5) and PK properties of the most potent
and selective TgCDPK1 inhibitors dosed at 10 mg/kg PO to mice.
Solubility
(ꢀM)
T
max
C
max
AUCo->inf
(ꢀM.min)
Half life
(min)
233 ±26
clearance
(mL/min)
0.7
(min)
120 0.75 ±0.15
50
0.89 ±0.17
200 0.26 ±0.04
(ꢀM)
1
8
82
430 ±58
48
54
94 ±51
128 ±53
67±60
68 ±64
223 ±182
50 ±73
2.5
1.8
110
0.2
0.2
0.3
6.7
21
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
7
9
1
2
3
17
30
50
0.88 ±0.23
5.2 ±1.0
13 ±1
>100
>100
>100
6.2
850 ±411
114 ±115
32
320
560
40
13700 ±1500 1190 ±510
33
7.8 ±1.4 16600 ±4300 1110 ±400
34
0.39 ±0.16
3.9 ±1.1
31 ±8
40 ±5
35
8.0
30
3.9 ±1.1
41 ±21
Figure 2. Comparison of oral PK for compounds 32 and 33 in mice. Both compounds were dosed
at 10 mg/kg, PO.
Dose Finding and Tolerability of 32 and 33
Based on the PK parameters of 32 and 33 following single dose administration to
mice, these compounds were evaluated for tolerability following escalated dosing. Both
compounds were initially dosed at 10 mg/kg on day one. As no observable adverse
ACS Paragon Plus Environment

Page 17 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
effects were detected in mice dosed with either 32 or 33, the next dose was increased to
5
0 mg/kg on day 4, followed by a final dose of 100 mg/kg on day 8. Mice showed no
overt signs of toxicity or weight loss over the 10ꢀday observation period. Next, the
plasma protein binding of both compounds was evaluated. Both compounds were highly
protein bound in mouse plasma with plasma protein binding of 96% for 32 and 88% for
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
3. Based on the protein binding values the plasma concentrations required to maintain
unbound plasma exposure above the TgCDPK1 EC are 1.33 ꢀM for 33 and 1.5 ꢀM for
50
3
2. Using these target concentrations and the single dose PK data, a multiple dosing
regimen with a loading dose of 20 mg/kg of 32 and 33 followed by a 5 mg/kg daily dose
for five days was evaluated. Blood was collected at multiple time points to determine
plasma concentrations over the course of treatment (Figure 3). Both compounds
demonstrated excellent exposure through the 24ꢀhour dosing interval with the trough
concentrations remaining >1.5 ꢀM throughout the study. The measured concentrations
following multiple dosing were similar to those predicted from the single dose PK studies
in mice (Figure 3). Neither compound showed any evident signs of toxicity compared to
mice dosed with vehicle only. Cardiac puncture blood collection was performed at the
end of the study for complete blood count and serum biochemical profiles. All results
were reported by Phoenix Central Laboratory to be within a normal range for species and
age.
In a second multiple dose study compound 32 was administered at 50 mg/kg PO
every other day for 5 doses and compound 33 was administered with a 20 mg/kg loading
dose followed by a 10 mg/kg maintenance dose every other day for 5 doses. The
exposure to 33 with this dosing regimen was similar to the one observed with daily
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
dosing of the lower dose and the plasma concentration at various time points were well
predicted based on the single dose PK results. With compound 32 plasma concentrations
remained >10 ꢀM for the entire duration of the study (Figure 3) with maximum
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
concentrations reaching 43 ꢀM. The exposure to compound 32 following the last 50
mg/kg dose was higher than predicted from single dose studies with longer halfꢀlife
suggesting possible saturation of metabolism at this dose.
30
40
30
20
20
10
10
0
0
0
0
100
200
300
400
0
50
100
150
200
250
time (h)
time (h)
8
40
30
20
10
0
60
4
0
0
0
2
0
100
200
300
400
0
100
200
300
400
time (h)
time (h)
Figure 3. Multiple dose PK studies with compound 32 and 33. The dosing regimens were
simulated based on the disposition data following single oral doses to mice (lines) and the
observed plasma concentrations (dots) compared to the simulation. (top left panel) 32 was dosed
at 20 mg/kg on day 1 and then 5 mg/kg every 24 hours for 5 doses. (bottom left panel) 32 was
dosed at 50 mg/kg every 48 hours for 5 doses. (top right panel) 33 was dosed at 20 mg/kg on day
1
and then 5 mg/kg every 24 hours for 5 doses. (bottom right panel) 33 was dosed at 20 mg/kg on
day 1 and then 10 mg/kg every 48 hours for 5 doses. Each dot is a mean of 3 mice with standard
deviation.
Based on the favorable single and multiple dose PK and the lack of observable
toxicity, compounds 32 and 33 were evaluated for tolerability and PK characteristics in
ACS Paragon Plus Environment

Page 19 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
rats following intravenous (IV) and PO dosing. Following IV administration to rats both
3
2 and 33 displayed biphasic kinetics with compound 33 reaching distribution
equilibrium more rapidly than 32 (Figure 4). Compound 32 distributed approximately to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
total body water with a volume of distribution at steady state (V ) of 0.9 L/kg. The
ss
distribution of 33 was more extensive with a V of 5.8 L/kg. Both compounds had very
ss
low systemic clearances in rats but, in contrast to mice, the clearance of 33 was 4ꢀfold
higher than that of 32. The elimination halfꢀlives of both compounds were acceptable for
multiple dosing regimens, 9.6 and 13 hours for 32 and 33, respectively (Figure 4).
Figure 4. Plasma concentration time curves for compound 32 (left) and 33 (right) following IV
and PO dosing to rats.
Table 5: In vivo pharmacokinetic parameters of 32 and 33 in rats following IV administration of
5
mg/kg. Data is shown as mean and range between animals.
T1/2 (hr)
AUC (h*
mol/L)
ꢀ
CL
mL/hr/kg)
Vss (L/kg)
.91
(
0
9.6 (9.5ꢀ
9.7)
1
67 (160ꢀ
75)
3
3
2
3
77 (73ꢀ80)
(0.89ꢀ
.93)
5.8 (4.1ꢀ 13 (12.6ꢀ
7.5) 13.4)
1
0
3
10 (216ꢀ
12)
45 (31ꢀ59)
4
Following oral administration to rats both compounds showed slow absorption
with absorption phase continuing for 12 hours after oral dosing. The plasma
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
concentrations over the 24ꢀhour period after PO dosing exceeded those observed after IV
dosing (Figure 4) suggesting that both 32 and 33 had essentially complete bioavailability
in rats. The apparent bioavailability of these compounds was greatly improved compared
to that of compound 1 (46%).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Because it is important that antiꢀtoxoplasmosis therapies are able to prevent
reactivation of parasites within tissue cysts, which largely reside in the central nervous
system (CNS), the distribution of compounds 32 and 33 to the brain was next determined.
To do this, the distribution of 32 and 33 into mouse brain (n=3) at one hour after
intraperitoneal dosing of 5 mg/kg was measured. The mean concentration of compound
3
2 at one hour in brain was 1.2 ±0.5 ꢀM, a concentration well above the TgCDPK1 EC50.
The corresponding plasma concentration of 32 was 4.1 ±1.1 ꢀM resulting in a brain to
plasma concentration ratio of 32 of 0.33 ±0.22. This brain penetration was comparable to
1
(0.31). In accordance with the larger distribution volume of compound 33, it
demonstrated a greater brain to plasma concentration ratio (1.65 ±0.84). Both the brain
and plasma concentration of 33, 1.90 ±0.24 ꢀM and 1.23 ±0.76 ꢀM, respectively, were
above the TgCDPK1 EC . As both 32 and 33 provide adequate exposure in CNS, they
50
are both excellent candidates for animal efficacy models of toxoplasmosis.
Selectivity Profiling of 32 and 33
Due to the impressive PK properties of 32 and 33, these inhibitors were profiled
further. To demonstrate that TgCDPK1 is the kinase target of these compounds in T.
gondii, compounds 32 and 33 were tested against parasitic cell lines overexpressing the
Gly128Met TgCDPK1 gatekeeper mutant or wildtype (wt) TgCDPK1. Expression of the
Gly128Met gatekeeper mutant of TgCDPK1, but not wtTgCDPK1, makes T. gondii
ACS Paragon Plus Environment

Page 21 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
highly resistant to PPꢀbased inhibitors that contain 6ꢀalkoxynaphthalenꢀ2ꢀyl groups at the
1
9,33
R position.
1
Both 32 and 33 show a dramatic loss in potency against parasites
overexpressing the Gly128Met gatekeeper mutant relative to the parent RH strain and to
T. gondii overexpressing wild type TgCDPK1 (Figure 5), which is consistent with
TgCDPK1 being the primary target through which these inhibitors exert their antiꢀ
parasitic effects.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. EC curves of inhibitors 32 (left) and 33 (right) for T. gondii overꢀexpressing either
50
wild type (wt) TgCDPK1 (black circles) or a drug resistant G128M TgCDPK1 mutant (blue
squares). All experiments were performed in triplicate.
32 and 33 were further profiled for any mammalian kinase off targets using a
panel of 80 human kinases representing different subfamilies of the kinome tree with
17
a fluorescenceꢀbased competition assay (Supporting Information Table 1). 78 of the 80
mammalian kinases tested have an IC >1.5 ꢀM (>1500ꢀfold selective for TgCDPK1)
5
0
for 32. For the two kinases–PKCν (PKD3) and MEK1–that have subꢀmicromolar IC50
values, compound 32 is a >120ꢀfold and >900ꢀfold less potent inhibitor than for
TgCDPK1, respectively. Compound 33 appears to be slightly more selective than
compound 32. Indeed, it demonstrated an IC value of greater than 5 ꢀM (>2500ꢀfold
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
selective) for 79 of the kinases tested, with only PKCν (PKD3) demonstrating a subꢀ
micromolar (IC = 0.280 ꢀM; 140ꢀfold selective) IC value.
5
0
50
Structure of the TgCDPK1ꢁ33 Complex
Previously, we demonstrated that certain substituents at the R position of the
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
25
pyrazolopyrimidine scaffold enhance inhibitor potency and selectivity for TgCDPK1.
While 1 has an IC of 2.9 nM for TgCDPK1 and is >3000ꢀfold selective for TgCDPK1
5
0
over the mammalian tyrosine kinase Src, the R isopropyl analog of 1 has an IC of 5.0
2
50
1
9,25
nM for TgCDPK1 and is only 76ꢀfold selective for TgCDPK1 over Src.
The 2ꢀ
methylpropanꢀ2ꢀol R group of inhibitors 32 and 33 seem to have a similar effect on
2
selectivity as the Nꢀmethylꢀ4ꢀpiperidinemethylene group of 1. For example, the R2
isopropyl analog of inhibitor 33 is only 40–fold selective for TgCDPK1 over Src, while
3
3, which contains a 2ꢀmethylpropanꢀ2ꢀol R group, is >5000ꢀfold selective. To gain a
2
better understanding of the contribution of the 2ꢀmethylpropanꢀ2ꢀol R group to inhibitor
2
binding, a coꢀcrystal structure of 33 bound to TgCDPK1 in the calciumꢀfree inactive form
was determined (Figure 6). As expected, 33 occupies the ATPꢀbinding cleft of
TgCDPK1, with the pyrazolopyrimidine scaffold making the same hydrophobic and
hydrogenꢀbonding contacts as the adenine ring of ATP. In addition, the 2ꢀ
cyclopropyloxyquinolinꢀ6ꢀyl R group of 33 is directed towards the Gly128 gatekeeper
1
residue of TgCDPK1, with the quinoline moiety occupying the enlarged hydrophobic
pocket that is available due to the lack of a side chain at this position. Side chains of
residues in helixꢀαC and the fourth βꢀstrand of the Nꢀterminal lobe form hydrophobic
interactions with the 2ꢀcyclopropyloxy group, most likely contributing to the favorable
potencies of inhibitors containing this substituent. The R substituent of 33 occupies the
2
riboseꢀbinding pocket of TgCDPK1, with the tertiary alcohol of the R substituent
2
ACS Paragon Plus Environment

Page 23 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
directed towards the phosphateꢀbinding loop (Pꢀloop), which caps the ATPꢀbinding site.
Superposition of the TgCDPK1ꢁ33 complex with the coꢀcrystal structure of TgCDPK1
bound to a close analog of 1 shows that the pyrazolopyrimidine scaffolds and R1
substituents of these inhibitors are well aligned within the ATPꢀbinding site. However,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
the R groups of both inhibitors adopt very different orientations. While the 2ꢀ
2
methylpropanꢀ2ꢀol R group of 33 is oriented towards the Pꢀloop of TgCDPK1, the larger
2
4
ꢀpiperidinemethylene group of the 1 analog is directed out of the ATPꢀbinding site
(Figure 6C) and forms an electrostatic interaction with the side chain of Glu155, located
on the surface of the Cꢀterminal lobe. Therefore, it appears that different interactions in
the ribose pocket can be exploited by R substituents to confer high selectivity for
2
TgCDPK1.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Structure of the TgCDPK1ꢁ33 complex (PDB accession code: 4TZR). (A) Entire view
of the 33·TgCDPK1 coꢀcrystal structure. Inhibitor 33 is shown as orange sticks. The CDPK
activation domain (CAD) of TgCDPK1 is in mint and the kinase domain is shown in grey. (B)
Expanded view of the ATPꢀbinding site from the TgCDPK1ꢁ33 complex. The backbone amides
of residues Glu129 and Tyr131 in the hinge region (shown as sticks) form two hydrogen bonds
(dashed red lines) with the pyrazolopyrimidine scaffold of 33. The 2ꢀcyclopropyloxyquinolinꢀ6ꢀyl
R group of 33 is directed towards the Gly128 gatekeeper residue. (C) An overlay of an analog of
2
1
(yellow) bound to TgCDPK1 (PDB accession code: 3SX9) and the TgCDPK1ꢁ33 (orange)
complex (PDB accession code: 3SX9)
ACS Paragon Plus Environment

Page 25 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. Calf plasma levels. Comparison of oral PK for compounds 32 and 33. Each compound
was tested in two calves. Compounds 32 was dosed at 9.3 mg/kg and compound 33 was dosed at
1
0 mg/kg, PO. Each point shows the mean and range of measurements in two calves.
Large Animal Pharmacokinetics
The lack of toxicity in the initial mouse toxicity screens allowed us to move
forward into large animal pharmacokinetic and tolerability profiling. Male calves (n=2
for each compound) were dosed orally at 10 mg/kg for compound 32 and 9.3 mg/kg for
compound 33 and blood sampled up to 12 days after dosing (Figure 7). Similar to rats,
the absorption of both compounds was slow with maximum concentrations reached at 24
hours for compound 33 and 12 hours for compound 32. The maximum plasma
concentrations were similar for the two compounds, 7.9 ꢀM and 9.8 ꢀM, for compound
32 and 33, respectively. However, the overall exposure to 33 was greater than that of 32
due to its lower oral clearance and longer halfꢀlife (Table 6). Yet, both compounds had
very low oral clearances and long systemic halfꢀlives. Similar to rats, compound 33 had a
higher apparent volume of distribution than compound 32 and the overall distribution
characteristics were similar to those observed in rats.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 6: In vivo pharmacokinetic parameters of 32 and 33 in calves following PO administration
of 10 mg/kg of 32 and 9.3 mg/kg of 33. Data are shown as the mean and range between animals.
t1/2 (hr)
AUC (h*
mol/L)
ꢀ
CL/F
mL/hr/kg)
V/F
(L/kg)
(
23 (17ꢀ
28)
5
88 (353ꢀ
25)
1.5 (1.3ꢀ
1.8)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
32
52 (31ꢀ72)
30 (22ꢀ39)
8
849 (607ꢀ
1091)
2.3 (1.4ꢀ
3.2)
51 (46ꢀ
56)
3
3
There were some notable calf health concerns following the single oral dose of
compound 33. For compound 33, both calves developed neurologic signs of toxicity on
day 4ꢀ5 postꢀdose. Both calves were observed to have a plantigrade stance and ataxia.
Despite abnormalities, both calves were ambulatory, alert, and responsive with normal
appetites. The signs became progressively worse, with notable ataxia, followed by
marked improvement by day 10 but were never completely resolved by the end of the
experiment on day 14. These signs were not observed for compound 32. One possibility
for the increased CNS toxicity was the greater brain exposure for compound 33 compared
with compound 32. The observed neurological signs of toxicity in compound 33 gave us
reasons to no longer pursue further studies of this compound. Moving forward we chose
to further characterize compound 32 before performing small animal efficacy studies and
large animal PK studies.
Further Toxicity Profiling of 32
Potential further toxicity of compound 32 was examined in mice by testing two
doses (30 mg/kg and 100 mg/kg PO) daily for five days, while observing mice for signs
of toxicity and collecting blood samples. Both groups of mice remained active, wellꢀ
groomed, and appeared normal throughout the study. Upon necropsy, there were no gross
abnormalities. Histology revealed mild focal inflammation in the spleen in two of three
ACS Paragon Plus Environment

Page 27 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mice in the 30 mg/kg group. The only abnormality seen in the 100 mg/kg group was
inflammatory infiltrate in the hepatic lobules in one of three mice. The concentrations
following 30 mg/kg doses were slightly higher than those predicted from single dose PK
data while the exposures following the 100 mg/kg doses were similar to those predicted
from single 10 mg/kg dose data (Figure 8). Similar to the early multiple dosing
experiments in mice, after the last 100 mg/kg dose the elimination of compound 32 was
slower than predicted with considerable plasma concentrations persisting at 72 hours
after the final dose.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
50
200
50
100
1
50
0
0
50
100
150
200
250
Time (h)
Figure 8. Exposure to compound 32 in multiple dose toxicity studies. The concentration profiles
of 32 following 30 mg/kg and 100 mg/kg daily doses were simulated based on the single dose 10
mg/kg data. The mean observed concentrations (n=3) with standard deviation are plotted as
triangles for 30 mg/kg dosing and circles for the 100 mg/kg dosing) with the simulation.
Next, we completed a maximally tolerated dose study to determine the lowest
observable adverse effect level (LOAEL) and the no observable adverse effect level
(NOAEL) of compound 32. Mice were dosed with single PO doses of 32 ranging from
2
00 mg/kg to 1000 mg/kg. The LOAEL was observed at 500 mg/kg and the NOAEL was
observed at 400 mg/kg. Mice had slightly ruffled fur and were less active than the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
control mice at 3 hours following the 500 mg/kg dose. The lowered activity persisted at
2
4 hours but was resolved by 30 hours. The lack of toxicity up to 500 mg/kg confirmed
the safety of the compound for small animal efficacy studies.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Large Animal Pharmacokinetics of 32
The pharmacokinetics of 32 was further explored in dogs and monkeys following
IV and PO administration. Similar to rats, 32 had a very low clearance in both species
and a relatively long halfꢀlife (Table 7). The plasma concentrations timeꢀprofile following
IV dosing was biphasic in both species with a similar extent of distribution in dogs and
monkeys as observed in rats and calves (Figure 9). The bioavailability of compound 32
was 66 ±17% in monkeys and 88 ±12% in dogs demonstrating good bioavailability in
both species as predicted from rats.
Table 7: In vivo pharmacokinetic parameters of 32 following IV and PO administration of 1
mg/kg to dogs and monkeys (n=3 for each). Data is show as mean and standard deviation.
t1/2 (hr)
AUC (h*
ꢀ
mol/L) CL (mL/hr/kg) Vss (L/kg)
Dogs
IV
11200 ±1100
9620 ±1350
90 ±10
1.7 ±0.1
1.8 ±0.3
13.2
9.6
PO
Monkeys
IV
8720 ±3400
5730 ±1500
130 ±50
PO
ACS Paragon Plus Environment

Page 29 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 9. Plasma concentration time curves for compound 32 following IV and PO dosing to
dogs (left) and monkeys (right).
In Vivo Efficacy of 32 Against Acute and Latent T. gondii Infection in Mice.
We tested 32 against a high inoculum of type I RH strain T. gondii to determine
efficacy against fulminate toxoplasmosis in two experiments (Figure 10). Unlike type II
T. gondii strains, type I strains do not typically form tissue cysts but rather cause death
within 10 days. Compound 32 was administered via oral gavage two days after infection
and the burden of infection was measured 5 days after the initiation of treatment.
Treatment with 32 at 20 mg/kg for 5 days reduced the number of T. gondii tachyzoites in
the peritoneal fluid below the limits of detection (less than 100 tachyzoites/mL).
Treatment with 32 was highly effective at 20 mg/kg daily for 5 days in reducing infection
in the spleen (more than 99%), and infection in the brain (95%). The efficacy of 32
against an acute T. gondii infection that is not adequately controlled by the mouse
immune system suggests that 32 would be active against toxoplasmosis in an
immunocompromised human.
ACS Paragon Plus Environment