Optimizing Pyrazolopyrimidine Inhibitors of Calcium Dependent Protein Kinase 1 for Treatment of Acute and Chronic Toxoplasmosis

Article abstract of DOI:10.1021/acs.jmedchem.0c00419

Calcium dependent protein kinase 1 (CDPK1) is an essential Ser/Thr kinase that controls invasion and egress by the protozoan parasite Toxoplasma gondii. The Gly gatekeeper of CDPK1 makes it exquisitely sensitive to inhibition by small molecule 1H-pyrazolo

Full text of DOI:10.1021/acs.jmedchem.0c00419

Subscriber access provided by Uppsala universitetsbibliotek
Article
Optimizing pyrazolopyrimidine inhibitors of calcium dependent
protein kinase 1 for treatment of acute and chronic toxoplasmosis
James W Janetka, Allen T. Hopper, Ziping Yang, Jennifer
Barks, Mary Savari Dhason, Qiuling Wang, and L. David Sibley
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00419 • Publication Date (Web): 18 May 2020
Downloaded from pubs.acs.org on May 18, 2020
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 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 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.
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 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
/15/20
Optimizing pyrazolopyrimidine inhibitors of calcium dependent
protein kinase 1 for treatment of acute and chronic toxoplasmosis
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
1*
2a
2
3
James W. Janetka , Allen T. Hopper , Ziping Yang , Jennifer Barks , Mary Savari
3
3
3*
Dhason , Qiuling Wang , L. David Sibley
1Department of Biochemistry and Molecular Biophysics, Washington University School
of Medicine, St. Louis. MO 63110
2Vyera Pharmaceuticals, 600 Third Avenue, New York, NY 10016
3Department of Molecular Microbiology, Washington University School of Medicine, St.
Louis. MO 63110
a
current address: Sage Therapeutics, 215 First Street, Cambridge, MA 02142
Running title: Optimizing CDPK1 inhibitors
Key words: serine threonine protein kinase, toxoplasmosis, calcium dependent kinases,
pharmacokinetics, structure activity relationship, small molecule inhibitors, CDPK1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 68
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
ABSTRACT: Calcium dependent protein kinase 1 (CDPK1) is an essential Ser/Thr
kinase that controls invasion and egress by the protozoan parasite Toxoplasma gondii.
The Gly gatekeeper of CDPK1 makes it exquisitely sensitive to inhibition by small
molecule 1H-Pyrazolo[3,4-d]Pyrimidine-4-amine or (PP) compounds that are bulky ATP
mimetics. Here we rationally designed, synthesized, and tested a series of novel PP
analogs that were evaluated for inhibition of CDPK1 enzyme activity in vitro and
parasite growth in cell culture. Optimal substitution on the PP scaffold included 2-pyridyl
ethers directed into the hydrophobic pocket and small carbocyclic rings accessing the
ribose-binding pocket. Further optimization of the series led to identification of the lead
compound 3a that displayed excellent potency, selectivity, safety profile, and efficacy in
vivo. The results of these studies provide a foundation for further work to optimize
CDPK1 inhibitors for the treatment of acute and chronic 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
INTRODUCTION
Toxoplasma gondii is a widespread parasite of domestic, companion and wild animals
1
that also frequently causes infections in humans . The organism’s life cycle involves an
acute phase dominated by rapidly growing tachyzoite stages that proliferate and
2
disseminate widely within the body . Following a vigorous immune response, the
parasite differentiates into a semi-dormant state characterized by tissue cysts that
3
contain bradyzoites, which replicate slowly and asynchronously . Although the turnover
of the chronic tissue stages is not precisely understood, available evidence suggests
that there are successive waves of cysts that are formed, expand slowly and rupture to
4
release bradyzoites . The majority of these released stages are destroyed by the
2
ACS Paragon Plus Environment

Page 3 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
adaptive immune response, and yet they give rise to daughter cysts, such that the
chronic infection is perpetuated. When ingested by members of the cat family,
bradyzoites can initiate development of the sexual phase within enterocytes of the gut
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
5
resulting in shedding of oocysts in the feces . Oocysts are highly persistent in the
environment and serve as a source of infection for many species of intermediate hosts
2
including rodents and domestic animals . Because T. gondii infection occurs commonly
in food animals, it poses a risk to food safety and animal well being.
Humans become infected with T. gondii by ingestion of tissue cysts in undercooked
6
meat, or oocyst contaminated food or water . Most human infections give rise to mild
illness that is effectively controlled by the immune response; however, toxoplasmosis
possess a serious risk to the immunocompromised and due to congenital infection7.
The severity of toxoplasmosis is related to the fact that tachyzoites disseminate to all
organs and readily forms tissue cysts within neurons in the brain and endothelial cells in
7
the retina . When these chronic infections reactivate due to decreased immunity, the
tachyzoite forms re-emerge and proliferate rapidly to cause encephalitis and retinitis,
8
respectively . Hence, drugs must readily cross the blood brain barrier in order to be
effective in control of toxoplasmosis. Current therapy for toxoplasmosis relies on
9
inhibition of the folate pathway with pyrimethamine and sulfadiazine . Complications
from this treatment include intolerance to sulfa drugs, intolerance of pyrimethamine due
to gastrointestinal distress, and the potential for bone marrow suppression that requires
10
co-administration of leucovorin (i.e. folic acid) . Although the current standard of care
can suppress active infection due to proliferation of tachyzoites, it has no effect on the
10a
semi-dormant bradyzoite forms . Although there are backup treatments that can be
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 68
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
used in the event of toxicity (i.e. clindamycin, atovaquone, azithromycin), they are also
9
unable to cure chronic infection . Consequently, those at risk of toxoplasmosis need to
be treated for long term, which is particularly problematic for the immunocompromised
or newborns. As such, there is a need to develop new treatments with an improved
safety profile and that would target the bradyzoites stages that are responsible for
infection.
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
The intracellular phases of T. gondii infection rely on active motility to power cell
11
invasion and egress . Among the components that are essential for this process, is the
calcium dependent protein kinase 1 (CDPK1), which controls microneme secretion and
12
thus is essential for parasite motility, cell invasion and egress . The recognition that
13
CDPK1 contains a Glycine (Gly) gatekeeper residue allowed repurposing of bulky
small molecule ATP mimetics previously used to probe kinase function based on
14
engineering of mammalian kinases to render them selectively susceptible . Based on
this discovery, we and others have explored the existing 1H-pyrazolo[3,4-d]pyrimidine-
4-amine or ‘pyrazolopryimidine’ (PP) kinase inhibitor scaffold (Table 1) identifying
compounds with good enzyme activity against CDPK1 13a, 15 16. Other scaffolds based
on the PP framework have also been shown to exploit this feature of the small
17
gatekeeper residue , and the shared nature of this enlarged ATP binding pocket has
been used to demonstrate the general utility of CDPK1 inhibitors for inhibiting a variety
18
of related parasites .
Previous studies have shown that within the core PP scaffold, the 4-position NH2
and adjacent N5 of the pyrimidine ring interact with the hinge region of the kinase; the
R1 substituent at the C3 position protrudes into the large hydrophobic pocket defined by
4
ACS Paragon Plus Environment

Page 5 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the Gly gatekeeper residue; and the pyrazolo N1 substituent at the R2 position extends
17
into the ribose sugar binding region of the ATP pocket . Analogs from the PP series
were shown to control acute infection in the murine model15c, 19 and to either decrease
the burden of chronic infection16b or cure immunocompromised mice with chronic
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
20
infection . The ability of PP-based inhibitors to reduce or prevent chronic infection may
be related to the reliance of bradyzoites on CDPK1 for invasion, as suggested by ex
20
vivo treatment of tissue cysts from chronically infected mice . The effectiveness of PP
inhibitors against bradyzoites is especially noteworthy as current therapies have little
effect on these stages.
Although CDPK1 shows promise as a target for developing new inhibitors of parasite
infection, existing analogs have been limited by serval undesirable features. Some of
the previously described compounds exhibited problems such as lack of kinase
selectivity (e.g. inhibition of Src or other human kinases)15a, 15b, 16c, 20 or undesired
21
inhibition of the hERG channel while others displayed adverse neurological symptoms
16b
when used to treat large animals . Other potent analogs in vitro showed low
metabolic stability in liver microsomes (mouse and human) and high in vivo clearance in
mice limiting their potential for advancement15c, 20. Herein we sought to rationally design
and optimize new potent diverse PP CDPK1 inhibitors with improved drug-like
properties, metabolic stability, pharmacokinetic and safety profiles for advancement to
the clinic.
RESULTS AND DISCUSSION
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 68
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
We previously reported on a unique series of PP CDPK1 inhibitors 15c, 20 having a t-butyl
at the pyrazolo R position and a phenyl ether at the R1 position (1a-c; Table 1). We
2
discovered the meta R position on the phenyl ether was preferred for potency further
3
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
identifying trifluoromethyl, chloro and methyl substituents as optimal substituents. While
these analogs showed excellent potency and moderate cell activity, their stability in
mouse liver microsomes was extremely low (half-life ~10 min). Thus, we initiated
studies directed at evaluating new CDPK1 inhibitors using the PP scaffold with
significant modification of R1 and R2 groups.
Inhibition of CDPK1
To evaluate potential inhibitors of CDPK1, we expressed recombinant parasite enzymes
in E. coli and used an ELISA assay to detect phosphorylation of a conserved peptide,
20
as described previously . CDPK1 activity was tested across an 11-point dilution series
ranging from 20 M to 0.17 nM. We also tested compounds for inhibition of parasite
growth using a previously described type 1 RH transgenic parasite that expresses -
22
galactosidase . Growth of T. gondii tachyzoites in monolayers of human foreskin
fibroblasts was tested across an 11-point dilution series ranging from 10 M to 0.13 nM
compound.
Rational design of novel PP-based CDPK1 inhibitors
Using the X-ray structure of compound 1b (Table 1) bound to CDPK1 (PDB code: 4IHP)
20
and our previous structural studies of other analogs , it was hypothesized that addition
of a nitrogen in the ortho ring position (2-pyridyl; 2b) would lock the 4-Cl-2-pyridyl ring in
6
ACS Paragon Plus Environment

Page 7 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the bound confirmation through an internal H-bond to the 4 amino NH (Figure 1A)
2
while at the same time significantly change the physical properties and metabolic profile
of the compound. Therefore, we synthesized analogs 2a and 2b and found that they
retained similar values for enzyme inhibition, but were ~ 10-fold more potent in inhibition
of parasite growth. Additionally, the stability in human liver microsomes of 2b improved
5-fold over matched pair 1b with a decent half-life of 60 min. Further exploration of R3
substituents in the 4-position produced cyano compound 2c that displayed excellent
metabolic stability in human liver microsomes (half-life >145 min) and increased stability
in mouse microsomes as well (half-life 46 min). Next, we evaluated if the tert-butyl R2
group could be replaced with a smaller, less lipophilic and often more metabolically
stable cyclopropyl group. Compounds 3a, 3b and 3c had improved metabolic stability
and similar CDPK1 potency as their t-butyl matched pairs (2a-c), but cellular potency
dropped-off by 2 to 5-fold (Table 1).
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
Structure activity relationships (SAR) of R Based on this preliminary work
1
identifying the new O-linked pyridyl R , we pursued a library of PP analogs with varied
1
R1 substituents where the R group was held constant as cyclopropyl (Table 2).
2
Interestingly, the analog where R is a hydroxy group and R is cyclopropyl showed
3
2
good enzyme and growth inhibition (4l; Table 2). This outcome was unexpected as
polar functional groups in this position are typically less active (e.g. 4x and 4m). We
further explored these findings by modeling this compound bound to CDPK1 (Figure
1B). We hypothesize that the ring is twisted 180 degrees relative to 2b where the
hydroxy group is making two H-bonds to backbone amide carbonyls of CDPK1. The
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 68
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
model also places the pyridyl nitrogen within proximity for beneficial electronic
interactions with a salt bridge Lys-Asp (Figure 1B).
It became clear that a pyridin-2-yl ring was preferred over other pyridine and di-nitrogen
heteroaryl ring systems. Further analysis of SAR indicated that the position of the
substituent off the 2-pyridyl ring was also important to achieve high enzyme inhibition,
with the 4 position being optimal as in previous compounds 2a-c. For example,
compare 6-chloropyridin-2-yl 4b with 4-chloropyridin-2-yl 3b with CDPK-1 IC s of 988
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
50
and 8.4 nM, respectively (Table 2). This result was not surprising and is consistent with
the location on the best non-pyridyl analogs reported previously. A compound worth
mentioning is 2-pyrazine 4r that had good enzyme inhibition (IC = 66 nM) comparable
50
to that of the unsubstituted 2-pyridyl 4a (IC = 38 nM). Subsequently, we constructed
50
derivatives with varied R3 groups at the 4-position of the pyridine-2-yl ring where we
found as before with 2a-c, the CF (3a), Cl (3b) and Me (4u) groups had the best
3
potency in both the enzyme and growth inhibition assays.
Evaluation of 4-(trifluoromethyl)pyridin-2-yl PP R substituents Having identified
2
the CF group as preferred in the 4-position based on the combination of potency and
3
stability, we next explored analogs with a diverse range of R substitutions on the
2
pyrazolo nitrogen (Table 3). We focused on carbocycles and non-aromatic
heterocycles (5 and 6) and found that cyclopropyl (3a), cyclobutyl (7a, Table 4),
cyclopentyl (5a, Table 3) and cyclohexyl (5b, Table 3) rings were mostly similar in
enzyme and cell activity, with CDPK1 IC s of 16, 9.8, 23 and 15 nM, respectively.
50
However, cyclopropyl (3a) was clearly differentiated from the others with superior
8
ACS Paragon Plus Environment

Page 9 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
metabolic stability in both mouse and human liver microsomes. The comparable
oxygen containing heterocyclic analogs 3-oxetain, 3-tetrahydrofuran, 4-tetrahydropyran
and 3-tetrahydropyran (5c, 5d, 5e, 5f, Table 3) had reduced and variable CDPK-1
potencies of 440, 48, 29 and 116 nM, respectively. The nitrogen containing heterocyclic
analogs 4-piperidine and 3-piperidine (5g, 5i, Table 3) showed similar decreased
activity against the CDPK1 enzyme with IC s of 68 and 57 nM. One important piece of
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
50
SAR was determined by comparing 4-piperidine 5g to 3,3 gem-difluoro-4-piperidine
analog 5h. The difluoro analog was predicted to reduce the pKa of the piperidine ring
nitrogen from a calculated value of 9.2 to 7.1 and thus hypothesized to improve cell
penetration and potency. As predicted, 5h showed a 10-fold increase in growth
inhibition, although some of this improvement is likely due to its 3-fold increase in
enzyme inhibition potency. Unfortunately, this strategy was not successful for 3-
piperidine 5i when compared to 3,3-gem-di-fluoro-5-piperidine 5j, which was >3-fold
less potent in growth inhibition. In contrast to the heterocycles, those rings appended
externally with an oxygen or nitrogen largely retained enzyme inhibition with varied
effects on growth inhibition and metabolic stability. The most promising results were
obtained with either the 4-membered or 6-membered rings. Some representative
examples include the 4-position cis and trans cyclohexyl alcohols, 6c and 6b (Table 3),
respectively. Both of these analogs had good enzyme and growth inhibition plus
excellent half-life in human microsomes but only moderate stability in the mouse
microsomes. Appending a methylene spacer between the ring and alcohol (6d, 6e,
Table 3) had a negative effect on the stability, while methylation of the alcohols (6f, 6g
Table 3) decreased potency 2 to 3-fold without improvement in mouse microsome
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 68
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
stability. However, 2-hydroxycyclohexanes (6j, Table 3) fared better, maintaining
potency with increased mouse microsome half-life; which was largely dependent upon
the stereochemistry with optimal substitution being trans and with the R, R
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
stereochemistry as with 6j. 2-Hydroxycyclohexane stereoisomers 6k and 6i were of
substantially weaker activity. The absolute stereochemistry was determined by small
molecule X-ray crystallography (Figure 2). Compound 6j had an EC of 77 nM in the
50
growth inhibition assay in addition to >145 min half-life in human and 66 min half-life in
mouse microsomes. Likewise, amino substituted cycloalkyls largely showed good
inhibition of enzyme and growth but interestingly was somewhat dependent on relative
and absolute stereochemistry. For example, comparison of cis-3-amino-cyclopentane
(6m) to the trans isomer (6l) shows the cis form was more potent in enzyme inhibition
and also had excellent stability in human microsomes and a moderate 84 min half-life in
mouse liver microsomes. A similar trend is noted for the 4-methylamino-cyclohexyl
versions as seen with 6n and 6o where the cis isomer 6n is also 3x more potent in
growth inhibition (Table 3). The dimethyl analog 6q and the desmethyl analog 6p both
had similar enzyme and growth inhibition plus >145 min half-life in human microsomes
but only 17 and 55 min half-life in the mouse liver microscome assay, respectively.
Other issues were identified with these promising amino carbocycles including cellular
efflux as indicated from the MDCK permeability assay and thus potentially poor CNS
permeability and low oral bioavailability. Thus, from this study, 4-CF -2-pyridyl
3
cyclopropyl 3a and 2-hydroxycyclohexyl 6j pyrazolopyrimidines both emerged as new
lead compounds for inhibition of CDPK1.
1
0
ACS Paragon Plus Environment

Page 11 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Multivariate analysis of 2-pyridyl-4-CF , Cl, CN, and OH analogs with optimal R
3
2
groups Shown in Table 4 are several analogs with other optimized R2 groups that were
synthesized in all three 4-CF , Cl, and CN series of 2-pyridyl PP CDPK1 inhibitors (7-
3
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
13). Many of these optimal R2 groups are substituted cyclobutyl derivatives bearing
either a hydroxy or fluoro substitution. As with previous SAR, we found that the trend
for inhibition of CDPK1 activity and parasite growth inhibition was Cl>CF >CN.
3
Consistent with this pattern, the analogs of 7b, 7a and 7c (Table 4) showed IC values
50
for enzyme inhibition of 6.2, 9.8 and 10 nM and EC values for growth inhibition of 27,
50
95 and 272 nM. The trend was reversed for stability in microsomes where CN analog
7c had a half-life in human microsomes of 141 min while 7b and 7a had 20 and 61 min
half-lives, respectively. In an attempt to improve upon 3a, the cyclopropyl was
exchanged with a cyclopropylmethyl in compounds 12a-c (Table 4) but this led to
decreased metabolic stability. A similar result was found from adding a methylene
spacer as in 13a-c in comparison to 5j. On the other hand, addition of a gem-difluoro
group to cyclohexyl 5b with 11a and 11b (Table 4) resulted in a >145 min half-life in
human microsomes but only ~25 min half-life in mouse microsomes. Better results
were obtained with both the hydroxy and fluoro-cyclobutyl R groups. In this case, cis
2
versus trans stereochemistry was not as important as seen with earlier analogs of
different R ring sizes. For example, cis hydroxy analogs 8a, 8b and 8c has similar
2
enzyme inhibition to trans hydroxy analogs 9a, 9b, and 9c (Table 4) but the latter
analogs did have a slight increase in growth inhibition relative to the cis variants. The
most promising result was discovered with trans-3-fluorocyclobutyl analog 10c.
Although the 4-Cl 2-pyridyl analog 10b clearly had the best potency in parasite growth
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 68
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
inhibition, the 4-CF analog 10a was not far behind, and the 4-CN version 10c had only
3
a slightly diminished activity (Table 4). Excitingly, we elucidated that 10c also had good
stability in both human and mouse microsomes with a half-life >145 min in both assays.
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
In another study we looked at different R groups where R was a 4-hydroxy-2-pyridine
2
(
14; Table 4) due to the promising results from initial cyclopropyl R analog 4l (Figure
2
1B and Table 2). We surmised that the hydroxy group could potentially decrease
metabolism; however, the best compound with a 4,4-difluorocyclohexyl R 14d had a
2
half-life of only 68 min in mouse microsomes.
Based on compound 10c, we decided to evaluate one more set of compounds
shown in Table 5 that had either the preferred trans-3-hydroxy (16a, 16b, 16c) or 3-
fluoro (15a, 15b, 15c) cyclobutyl R group. In this case, we switched the R group back
2
1
to a phenyl ether as opposed to pyridyl ether. Our combined results clearly
demonstrate that the 2-pyridyl group in the new series of PP CDPK1 inhibitors is not
only responsible for gains in cellular activity but also the enhancements attained in
metabolic stability.
Synthesis
Analogs of the PP scaffold were synthesized utilizing both existing and new synthetic
transformations/methodologies which are outlined in this section. Scheme 1 outlines the
synthesis of compounds shown in Table 2 specifically bearing a cyclopropyl R group.
2
Starting from cyclopropyl boronic acid (17), the hydrazine (18) was obtained from
23
reaction with di-tert-butyl (E)-diazene-1,2-dicarboxylate and copper acetate followed
by Boc deprotection. A novel condensation of 18 with ketene ethylene acetal 1924
1
2
ACS Paragon Plus Environment

Page 13 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
provided cyclopropyl pyrazole intermediate 20. Cyclization to the ring fused 5-6
pyrazolopyrimidine 21 was performed by reaction with formamide. Subsequent
hydrolysis of the ethylene glycol ether with potassium hydroxide in diphenyl ether at
high temperature yielded key intermediate alcohol 22. The final target compounds such
as 3a were all obtained by substitution reactions of the appropriate chloro-pyridine (2-
chloro-4-(trifluoromethyl)pyridine for 3a) or bromo-heterocyclo derivative with 22 using
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
o
potassium carbonate and DMSO at 120 C.
The 2-pyridyl-4-CF pyridine PP analogs shown in Tables 1 and 3 were constructed
3
through a similar approach as in Scheme 1 but with a slight variation to install the
variable R groups using a common intermediate shown in Scheme 2A. Starting with t-
2
25
butyl pyrazolo-pyrimidine alcohol 23 (made in an analogous fashion to 22) , reaction
with the 2-chloro-4-(trifluoromethyl)pyridine as before gave compound 2a. Removal of
o
the t-butyl group was effected with conc. sulfuric acid at 0 C which gave advanced
intermediate 24. As an example of a final target synthesis, lead compound 6j is
highlighted. Cyclohexyl mesylate intermediate 27 was synthesized via a two-step
procedure first by reaction of cyclohexane-cis-1,2-diol 25 with 0.5 eq. benzyl bromide
and NaH to provide monoprotected intermediate 26 and then further reaction with mesyl
chloride. Displacement of the mesylate by 24 resulted in protected compound 27 which
was debenzylated using boron trichloride. The final compound 6j was isolated as the
pure (R,R) enantiomer following SFC chiral chromatography. A modified route was
developed shown in Scheme 2B where Intermediate 24 was di-Boc protected as 28,
alkylated on the pyrazole nitrogen with 2-chlorocyclohexan-1-one to give 29. After a
systematic study of different chiral reducing agents, the final compound 6j was obtained
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 68
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
by utilization of the chiral reducing agent (S)-CBS (Corey-Bakshi-Shibata)
oxazaborolidine catalyst which gave excellent enantioselectivity for the desired (R,R)
isomer.
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
Outlined in Scheme 3 is the general synthetic route applied in the synthesis of
the 2-pyridyl-4-CN PP analogs shown in Tables 1 and 4. Once again starting from
intermediate 23, reaction with 2-bromoisonicotinonitrile yields t-butyl target compound
2c. Unfortunately, removal of the t-butyl group with sulfuric acid also hydrolyzed the
nitrile to the carboxamide 30. Conversion back to the nitrile using TFAA followed by
reaction with pyridine in methanol gave 31. Transformation of 31 to target compound
10c was completed via alkylation with cis-3-fluorocyclobutyl mesylate 33 (from alcohol
32) using cesium carbonate and potassium iodide.
The R1 phenyl PP matched pairs of the pyridyl PP compounds shown in Table 5
were produced following the route in Scheme 4 and exemplified with compound 15a. In
this case we started with bromo PP intermediate 34, reported in our previous publication
20
of CDPKI inhibitors . S 2 displacement of trans-3-(benzyloxy)cyclobutyl mesylate 36
N
(
from alcohol 35 in two steps) with 34 followed by debenzylation produced the cis 3-
benzyloxy)cyclobutyl N-pyrazolopyrimidine bromide intermediate 37 which was
(
subjected to addition of 3-(trifluoromethyl)phenol using cesium carbonate and copper
chloride to yield PP alcohol 38. Boc-protection of the 4-amino group gave 39 followed
by fluorination with DAST with inversion of stereochemistry at the alcohol center then
Boc-deprotection gave trans-3-(fluoro)cyclobutyl PP target compound 15a as the direct
matched pair of the pyridyl derivative 10a.
1
4
ACS Paragon Plus Environment

Page 15 of 68
Journal of Medicinal Chemistry
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
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 68
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
1
6
ACS Paragon Plus Environment

Page 17 of 68
Journal of Medicinal Chemistry
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
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 68
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
Evaluation of top leads
Based on the activity profiles of the compounds described above, we chose the top
three analogs 3a, 10c, and 6j for further in vitro and in vivo testing. These compounds
demonstrated very similar IC values for inhibition of enzyme activity and EC values
50
50
for inhibition of parasite growth (Figure 3). To confirm the specificity of leads 3a, 6j and
0c for the Gly gatekeeper feature, we compared the activity against wildtype CDPK1 to
1
15c
a mutant enzyme that altered at Gly128 to Met (G128M), as described previously .
This mutation resulted in loss of inhibition with all top three PP analogs even when
tested at 20 M (Table 6). These results confirm that the inhibitors are selective for Gly
1
8
ACS Paragon Plus Environment

Page 19 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
gatekeeper kinases and indicate that the inhibition of parasite growth in cell culture is
quite likely due to targeting CDPK1. At the doses used in vitro here, it is likely that the
main effect of PP inhibitors is to block parasite invasion and egress, as described
previously 12, 15c. In contrast, studies using much higher doses (i.e. several M) of some
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
26
PP compounds can disrupt parasite cell division , an effect that may be attributable to
inhibition of MAPK, which occurs at high concentrations using early PP inhibitors that
lacked specificity 27.
Additionally, we tested the specificity of these PP inhibitors against human c-Src, a
kinase that contains Thr at the gatekeeper and which shows intermediate sensitivity to
28
PP inhibitors . The IC for inhibition of c-Src was > 20 M (Table 6), demonstrating
50
that these PP inhibitors are extremely selective and show > 1,000-fold increased
potency against the parasite enzyme. To further explore the selectivity of 3a and 10c,
we screened an extensive panel of 485 human kinases (Thermo-Fisher) with
compounds at a single fixed concentration of 1 M. Compound 3a inhibited five targets
(i.e. CKƐ, BRK (PTK6), HGK (MAP4K4), TGF (ALK5), and PKAC) while 10c only
inhibited two host kinases (i.e. CKƐ, BRK (PTK6)) by ~ 50% (Figure 4, Table S1).
These findings indicate that these PP analogs are potent inhibitors of CDPK1 with
relatively low off target activity against human kinases.
We also tested the 3a, 6j, and 10c against type strains from the three major lineages
29
of T. gondii that found in North America and Europe . All three compounds exhibited
similar EC values when tested against this panel of strains (Table 7). No adverse
50
effect on the host cell monolayer was observed across this dilution series. We also
tested the potential for these inhibitors to affect the growth or viability of a variety of
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 68
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
human cell types including lines isolated from kidney, liver, brain, and gut. There was
no observed inhibition of cell growth by these PP compounds, although they enhanced
growth slightly for human brain endothelial cells (hCMEC) (Table 8). As expected, the
positive control mitomycin C showed significant inhibition of all the cell types.
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
In vitro ADME properties
To examine the in vitro properties of the lead compounds, we profiled them using a
standard set of assays designed to estimate solubility, permeability and metabolic
stability (Table 9). First, all three leads showed good kinetic solubility at pH 7.4. They
were also highly stable in the presence of human liver microsomes and human
hepatocytes, showing low clearance in both assays. All three lead compounds showed
minimal susceptibility to aldehyde oxidase, an enzyme involved in hepatic and extra-
30
hepatic phase I metabolism of xenobiotics . Finally, all three lead compounds showed
good uptake by polarized epithelial MDCK cells and little capacity for efflux in a cell line
over expressing the MDR1 transporter. These properties predict the lead compounds
should not be subject to efflux or rapid catabolism in vivo and therefore might have good
pharmacokinetic properties.
Off target activity and potential for genotoxicity
Although the lead compounds exhibited minimal activity against host kinases, we also
wanted to profile them for other potential off-target activity against other host enzymes
and receptors. We tested the three leads at a 10 µM concentration against the Safety
panel 47 (Eurofins, DiscoverX) that includes human kinases, channels, GPCRs, cyclic
2
0
ACS Paragon Plus Environment

Page 21 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
nucleotide and calcium signaling pathways, and neurotransmitter activities. All three
compounds partially inhibited the kinase Lck (Table 10), although this activity was not
seen in the kinome-wide screen. This difference likely reflects the higher concentration
of compounds used (10 M for Discover X vs. 1 M for the kinome wide screen), and
differences in ATP (Discover X assays lack ATP while the kinome wide screen was
conducted at Km for ATP). Additionally, 3a showed modest activity against an
antagonist of alpha 2A- adrenergic receptor (ADRA2A) (Table 10), an adrenergic
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
31
receptor that is broadly distributed in the body . Antagonists of ADRA2A have been
studied for their ability to enhance motor function in rodent and non-human primate
31
models of Parkinson disease . Inhibition by 3a would only be expected to be an issue
in patients also receiving antagonists of this receptor. None of the three compounds
32
showed appreciable inhibition of hERG, a K+ channel important in cardiac function .
The lack of activity against hERG channels is notable as other PP analogs that inhibit
16b
CDPK1 have previously been found to exhibit this liability . All three compounds were
negative in a mini-Ames test and for micronucleus formation (Table 10), indicating that
they have a low capacity for genotoxicity.
Finally, we tested the potential for the lead compounds to inhibit several cytochrome
P450 enzymes or CYPs, which are involved in detoxification of xenobiotics and
33
metabolism of drugs . Compound 3a showed inhibition of CYP1A2 at < 1 M and
CYP2C19 at < 10 M, but otherwise there was minimal effect on these enzymes
including on CYP3A4 and CYP2D6 (Table 11), which are responsible for metabolizing
33
most drugs . Collectively, these findings indicate that all three lead compounds
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 68
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
encompass minimal potential for genotoxicity and a low likelihood for complications
related to drug-drug interactions.
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
In vivo pharmacokinetics
In order to estimate the potential metabolic stability of the top leads in different species,
we compared the metabolism in liver microsomes from mouse, rat, rabbit, minipig and
dog. Compound 3a showed low metabolism in mouse, rat and minipig, with higher
levels in dog and rabbit, relative to the normal single pass rate of clearance due to
hepatic circulation (Table S2). Compound 10c showed low metabolic clearance in
microsomes for all species (Table S2). Compound 6j showed high metabolism in
mouse but was stable in the other species (Table S2). Because these in vitro assays
only estimate the potential for clearance in vivo, we also performed pharmacokinetic
(PK) analysis over an 8 h time course following a single 10 mg/kg oral dose vs 3 mg/kg
i.v. dose. We chose mouse for these studies since this species is the most versatile
model for showing efficacy against toxoplasmosis. Compounds 3a and 10c showed
similar maximum concentration in plasma (Cmax) and area under the curve (AUC)
values, although 3a had a slightly longer half-life (Table 12). By comparison, 6j showed
high clearance and corresponding lower Cmax and plasma AUC. All three compounds
showed moderate to high bioavailability (%F), high unbound fractions in plasma and in
brain, and importantly exhibited favorable CNS/plasma ratios indicating they are brain
penetrant (Table 12). Finally, compounds 3a and 10c showed no adverse effects when
dosed at 600 mg/kg for 7 consecutive days in mouse (data not shown). Although we
have not performed any studies on the safety of PP compounds during pregnancy,
2
2
ACS Paragon Plus Environment

Page 23 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
prevention of congenital infection is an important unmet medical need. Hence, future
studies warrant evaluation of these compounds for safety and efficacy in congenital
transmission models of toxoplasmosis.
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
Efficacy against acute infection
Due to its better performance in PK and relatively low potential for toxicity, we chose
compound 3a for further PK and efficacy studies. We compared the clearance of 3a in
groups of mice that were dosed at 3, 10, and 30 mg/kg given once by oral gavage. The
subsequent free plasma levels were plotted relative to the free EC and multiples of
50
this value (Figure 5B). From these studies, it was evident that only a dose of 30 mg/kg
was sufficient to maintain the plasma levels above the EC for ~ 8 h (Figure 5B) and in
50
all dosing schemes 3a dropped to undetectable levels by 24 h (data not shown). Based
on these PK results, we reasoned that it would likely be necessary to provide either a
higher total dose, or repeated dosing at a level of at least 30 mg/kg to protect against
toxoplasmosis.
Based on the anticipated PK profile, we designed a trial to evaluate 3a for its ability
to control acute toxoplasmosis in the laboratory mouse. Mice are a natural host for T.
gondii and they recapitulate the acute and chronic phases of infection that are seen in
1
humans and other species . Laboratory mice are somewhat more susceptible than
other animals and in particular, they are highly susceptible to type I strains, which are
34
lethal at a very low inoculum in immunocompetent outbred and inbred mice . Hence,
challenge with type I strains offers a stringent test of the ability of compounds to prevent
infection. We inoculated mice i.p. with 3,000 tachyzoites of the RH strain, which is
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 68
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
34
equivalent of > 1,000 times the lethal dose (Figure 5A). Infection was tracked using
35
bioluminescence from a firefly luciferase expressing strain of the parasite . Following
infection, the parasites amplified rapidly and spread to all of the major organs, leading to
a strong luciferase signal in the peritoneum (Figure 5A). Treatment of mice with a total
daily dose of 60 mg/kg, po of 3a using different dosing paradigms of 60 mg/kg once a
day (QD), 30 mg/kg twice a day (BID), or 15 mg/kg four times per day (QID) for a total
of 7 days were all effective at reducing parasite expansion and dissemination, as shown
by bioluminescence measurements (Figure 5C) and prevention of death in the majority
of animals (Figure 5D). To determine if animals were cured by treatment, the brain and
lungs of surviving animals were homogenized and inoculated into recipient naïve
animals that were subsequently followed without additional drug treatment. Because
this strain is highly virulent in mice, inoculation of even a single parasite of the type I RH
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
34
strain is expected to lead to death in the recipient . Bioluminescence imaging and
survival assays revealed that all of the animals inoculated with tissues from mice treated
6
0 mg/kg QD succumbed to infection whereas 20% of those inoculated with tissues
from mice treated with 30 mg/kg BID and 15 mg/kg QID survived (Figure 5D). For
these results, we conclude that more frequent dosing at lower concentration provides
equal protection from survival and greater frequency of cure.
Efficacy in the immunocompromised mouse model
One of the key target populations for treatment of toxoplasmosis is
immunocompromised patients such as those with HIV infection that have developed
36
AIDS . To simulate this defect in the murine model, we used a previously described
2
4
ACS Paragon Plus Environment

Page 25 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
immunocompromised mouse that lacks IFN- receptors and hence is unable to control
infection with a type II strain of the parasite, which is less virulent and prone to
20
developing chronic infection in wild type mice . Based on the PK studies and efficacy
in immune competent mice described above, we reasoned that higher doses would
likely be required for control in the absence of a functional immune system. As such,
we tested the PK in mice following an oral dose of 60, 90, or 120 mg/kg twice daily over
a 24 h period. Analysis of plasma levels indicate that the compound reached a peak
that exceeded the EC (free concentration) but that the concentration dropped below
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
90
the EC by 8 or 12 h (Figure 6B). To compare the efficacy of these doses in
50
preventing reactivation of toxoplasmosis, we treated immunocompromised mice using
-/-
the model described in Figure 6A. Ifngr1 mice were orally infected with tissue cysts of
the type 2 strain ME49 expressing firefly luciferase and maintained by treatment with
sulfadiazine, which suppresses tachyzoite growth but does not affect the development
37
of tissue cysts . All inoculated animals became infected as shown by visible signs of
illness and verified by positive ELISA titer at three weeks post-infection (Figure S1A).
At 22 days post-infection, treatment with sulfadiazine was discontinued and 48 h later
the animals were treated with compound 3a at 60 mg/kg by oral gavage BID for 8 or 21
days. Control animals lost weight and succumbed to infection by day 12 (Figure 6
C,D). Animals treated with 60 mg/kg for 8 days eventually succumbed to infection,
although they lasted significantly longer than controls (Figure 6D). Treatment with 60
mg/kg for 21 days lead to significantly prolonger survival and two animals survived
beyond 30 days (Figure 6D), after which they gained weight (Figure 6C), suggesting
they were cured of infection. The brain of these surviving animals was homogenized
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 68
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
-/-
and injected into recipient Ifngr1 mice that were not treated with compound. These
recipient animals, which are normally highly susceptible and unable to control infection,
remained seronegative (Figure S1 B,C), confirming that the donor mice had been cured
during treatment. In a follow up study, we repeated this trial using animals that were
treated with 60, 90 or 120 mg/kg BID for 8 days (Figure S2). The survival of mice
treated for 8 days with either 60, 90, or 120 mg/kg was similar, while greater survival
was achieved with treatment at 60 mg/kg for 21 days (Figure 6E). Collectively, these
findings indicate that the efficacy of 3a is driven primary by total time of exposure over
an extended time period. This result may reflect the fact that CDPK1 inhibitors are
expected to act to block egress and invasion, but may not have an effect on bradyzoites
while they are intracellular.
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
CONCLUSIONS
We generated several series of novel PP CDPK1 inhibitors to explore SAR and
metabolism profiles of an expanded range of analogs. Modification of the R position to
1
include pyridyl rings with N in the 2-position allowed appending various functional
groups on the 5-position of the ring that imparted greater metabolic stability including Cl,
CF and CN groups. Interestingly, an OH group in the same position was tolerated,
3
likely due to a different binding mode of the pyridyl ring. Addition of small cycloalkyl
groups or other substituted small heterocyclic groups at R also improved potency and
2
stability. By several rounds of optimizing potency, selectivity, and metabolic profiles, we
selected the set of top leads 3a, 6j, and 10c for further characterization of their safety
profile and metabolism in different species. Of the three top leads, compound 3a was
2
6
ACS Paragon Plus Environment

Page 27 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
chosen for advanced evaluation in murine models of toxoplasmosis. Lead PP 3a
prevented lethal acute infection, prolonged the survival of immunocompromised mice
and resulted in cure of chronic infection in ~25% of animals. Collectively, our studies
and their findings identify not only promising new PP CDPK1 inhibitors and SAR but
also important models and parametric analysis for further evaluation of inhibitors for
advancement to clinical trials for treatment of both acute and chronic toxoplasmosis.
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
EXPERIMENTAL SECTION
Synthetic chemistry. All synthetic chemistry and DMPK was performed at WuXi App
Tec at their China facilities in Tianjin and Shanghai. Proton NMR spectra were
recorded on a Varian 400 MHz NMR with chemical shifts δ recorded in ppm relative to
TMS. LCMS were taken on a quadrupole Mass Spectrometer on Shimadzu LCMS
2010 (Column: sepax ODS 50×2.0 mm, 5 µm) or Agilent 1200 HPLC, 1956 MSD
(Column: Shim-pack XR-ODS 30×3.0 mm，2.2 µm) operating in ES (+) ionization
mode. All final compounds were greater than 95% purity based on HPLC UV% AUC as
1
determined at 214, 254 and 280 nM wavelengths. Structures were confirmed by H
NMR and LC/MS. Final compounds in this manuscript are not known to interfere with
the assays herein (i.e., they are not PAINS). Detailed synthetic chemistry and
analytical analyses are found in the Supplemental materials.
Parasite strains and culture. In vitro growth inhibition experiments were conducted
with the 2F clone of the type I RH strain T. gondii, a transgenic line that expresses -
22
galactosidase . We also used additional T. gondii strains representing type 2
(ME49∆hx::FLUC strain) and type 3 lineages (CTG-FLUX), that were derived
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 68
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
38
independently, as described previously . The type 2 ME49∆hx::FLUC strain was used
for the reactivation studies in immunocompromised mice, as described below.
Tachyzoites of each strain were grown in monolayers of human foreskin fibroblasts
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
(
HFF, obtained from the Boothroyd laboratory at Stanford University) maintained in
complete medium (DMEM containing 10% FBS, 10 mM glutamine, and 10 g/ml
o
gentamycin) incubated at 37 C in 5% CO . In vitro cultures were shown to be
2
mycoplasma negative using the e-Myco plus kit (Intron Biotechnology).
Enzyme expression and purification. Full-length T. gondii wild type or Gly 128 Met
(G128M) mutant CDPK1 enzymes were expressed with a C-terminal His tag in
12
pET22b(+), as described previously . For protein purification, CDPK1 was expressed
BL21 (DE3)V2RpAcYc-LIC+LamP E coli, which contains the LamP phosphatase. As a
control, we also expressed cDNA fragment encoding human Src (hSrc) kinase domain
20
o
in BL21, as described previously . Following overnight growth in Terrific Broth at 37 C,
o
bacterial cultures were diluted 1:100 and grown for 3 hr at 37 C (O.D. 0.6-0.8), then
o
o
induced with 0.3 mM IPTG during overnight growth at 15 C or 30 C. Cells were lysed
using CellLytic B cell Lysis reagent (Sigma-Aldrich) and soluble proteins purified using
HIS-select Nickel Affinity Gel. Purified proteins were dialyzed against storage buffer (i.e.
5
0 mM Tris-HCL, pH7.5, 150 mM NaCl), and stored in 25% glycerol containing 0.5 mM
o
DTT at -80 C. Protein purity and concentrations were determined by SDS-PAGE and
staining with Coomassie Blue (Invitrogen).
Enzyme assays. CDPK1 activity was monitored based on phosphorylation of syntide-2
peptide (Callbiochem) that was detected using mAb MS-6E6 (MBL Intl, Corp) using an
20
o
ELISA protocol described previously . Reactions were conducted at 30 C for 20-40
2
8
ACS Paragon Plus Environment

Page 29 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
min in 20 mM HEPES, pH 7.5, 10 mM MgCl , 1 mM DTT, 2.5 mM CaCl , 0.1 mM EGTA,
2
2
0.005% Tween 20. Kinase reactions included 25 M ATP (the Km for the enzyme) using
25 nM of kinase per reaction. For comparison of the inhibition of wild type and G128M
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
mutant CDPKA, 100 nM of enzyme was used for both enzymes as the mutant enzyme
is slightly less active. To determine the potency of inhibitors, duplicate wells were
treated with a range of compound concentrations from 20 M to 0.13 nM containing 1%
20
DMSO final, as described previously . Individual IC values were determined from
50
three or more independent biological replicates each containing two technical replicates.
Human Src activity was monitored by phosphorylation of the Abltide peptide
(EAIYAAPFAKKK, Enzo Life Sciences) that was detected using a monoclonal antibody
to phosphotyrosine (Sigma) using an ELISA protocol similar to that used for CDPK1
above. Kinase reactions were conducted in 50 mM Tris-HCl, pH 7.5, 10 mM MgCl , 2.5
2
nM MnCl , 0.2 mM DTT, 0.5 mM EGTA. Reactions were conducted using 5 M ATP
2
(the Km for the enzyme) or 1 mM ATP (physiological) using 75 nM kinase per reaction.
Compounds were tested at 10 M in duplicate, in two biological replicates. For
compounds that showed > 50% inhibition at 10 M, serial dilutions from 10 M to 0.5
nM were tested in duplicate to derive IC values.
50
In vitro growth assays. -galactosidase assays Parasite growth inhibition assays were
conducted using the type I RH strain, 2F clone that expresses bacterial -galactosidase
22
(
-gal), as described previously . Compounds dissolved in DMSO as 10 mM stocks
o
that were stored at -80 C until use and then diluted in medium to reach final
concentrations. HFF monolayers grown in 96 well plates were inoculated with 5 x 102
parasites containing dilutions of compounds ranging from 10 M to 0.17 nM plus 0.1%
2
9
ACS Paragon Plus Environment