Mtb PKNA/PKNB Dual Inhibition Provides Selectivity Advantages for Inhibitor Design to Minimize Host Kinase Interactions

Article abstract of DOI:10.1021/acsmedchemlett.7b00239

Drug resistant tuberculosis (TB) infections are on the rise and antibiotics that inhibit Mycobacterium tuberculosis through a novel mechanism could be an important component of evolving TB therapy. Protein kinase A (PknA) and protein kinase B (PknB) are both essential serine-threonine kinases in M. tuberculosis. Given the extensive knowledge base in kinase inhibition, these enzymes present an interesting opportunity for antimycobacterial drug discovery. This study focused on targeting both PknA and PknB while improving the selectivity window over related mammalian kinases. Compounds achieved potent inhibition (K_i ≈ 5 nM) of both PknA and PknB. A binding pocket unique to mycobacterial kinases was identified. Substitutions that filled this pocket resulted in a 100-fold differential against a broad selection of mammalian kinases. Reducing lipophilicity improved antimycobacterial activity with the most potent compounds achieving minimum inhibitory concentrations ranging from 3 to 5 μM (1-2 μg/mL) against the H37Ra isolate of M. tuberculosis.

Full text of DOI:10.1021/acsmedchemlett.7b00239

Letter
Cite This: ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX
Mtb PKNA/PKNB Dual Inhibition Provides Selectivity Advantages for
Inhibitor Design To Minimize Host Kinase Interactions
Tiansheng Wang,*,† Guy Bemis,† Brian Hanzelka,†,§ Harmon Zuccola,† Michael Wynn,†
Cameron Stuver Moody,† Jeremy Green,† Christopher Locher,†,∥ Aixiang Liu,‡ Hongwu Gao,‡
Yuzhou Xu,‡ Shaohui Wang,‡,⊥ Jie Wang,‡ Youssef L. Bennani,† John A. Thomson,†,# and Ute Muh†,∇
̈
†Vertex Pharmaceuticals Incorporated, 50 Northern Avenue, Boston, Massachusetts 02210, United States
‡Shanghai ChemPartner Co. Ltd., 998 Halei Road, Pudong New Area, Shanghai 201203, China
S
* Supporting Information
ABSTRACT: Drug resistant tuberculosis (TB) infections are on the rise and antibiotics that inhibit Mycobacterium tuberculosis
through a novel mechanism could be an important component of evolving TB therapy. Protein kinase A (PknA) and protein
kinase B (PknB) are both essential serine-threonine kinases in M. tuberculosis. Given the extensive knowledge base in kinase
inhibition, these enzymes present an interesting opportunity for antimycobacterial drug discovery. This study focused on
targeting both PknA and PknB while improving the selectivity window over related mammalian kinases. Compounds achieved
potent inhibition (Ki ≈ 5 nM) of both PknA and PknB. A binding pocket unique to mycobacterial kinases was identified.
Substitutions that filled this pocket resulted in a 100-fold differential against a broad selection of mammalian kinases. Reducing
lipophilicity improved antimycobacterial activity with the most potent compounds achieving minimum inhibitory concentrations
ranging from 3 to 5 μM (1−2 μg/mL) against the H37Ra isolate of M. tuberculosis.
KEYWORDS: Protein kinase A (PknA), protein kinase B (PknB), Mycobacterium tuberculosis, dual targeting,
structure activity relationship (SAR)
uberculosis (TB) remains a major unmet medical need due
to widespread drug resistance, HIV coinfection, and
moderate antibacterial activity was achieved, and no data on
PknA was reported. We expand on these results by targeting both
PknA and PknB. Dual targeting has the potential to significantly
lower the frequency at which resistance develops. However,
broadening the inhibitor specificity may increase the risk of
introducing adverse side effects through the inhibition of
mammalian kinases. We report here the optimization of a lead
series that achieved potent inhibition of both PknA and PknB
with Ki values in the single digit nanomolar range. Structures of
compound-enzyme cocrystals allowed the identification of a
unique feature in the binding site of PknB. Compound
substitutions that exploit this feature achieved a 100-fold
selectivity window over a range of mammalian kinases. For
T
ineffective healthcare management with approximately two
billion latent infections, ten million new cases, and two million
deaths each year.1 Antibacterial compounds that act through a
novel mechanism of inhibition may offer advantages to treat
infections with drug-resistant strains. Of the 11 serine/threonine
(Ser/Thr) protein kinases described in Mycobacterium tuber-
culosis (Mtb),2,3 protein kinase A (PknA) and protein kinase B
(PknB) are especially attractive drug targets because they are
essential for bacterial growth both in culture medium and in Mtb-
infected host macrophages.4−6 The availability of protein crystal
structures for PknB7−9 will enable the design of potent and
selective compounds. PknA and PknB are crucial for cell wall
formation, resuscitation from dormancy, and performing multi-
ple regulatory functions in metabolism and adaptation to
environmental stress.10−13 Antimycobacterial activity through
PknB inhibition has been explored by others.14,15 However, only
Received: July 5, 2017
Accepted: November 28, 2017
Published: November 28, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acsmedchemlett.7b00239
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
Scheme 1. Diversification Strategy and Initial Quinazoline SAR
many of the lead compounds, improved enzymatic potency
translated to improved antimycobacterial activity. Several
compounds achieved a minimum inhibitory concentration
(MIC) around 3−5 μM (1−2 μg/mL).
The project was initiated by screening a collection of 1078
compounds representing kinase inhibitor chemical diversity
against the PknA and PknB kinase domains. Compounds with Ki
< 10 μM were modeled into the active site of PknB in complex
with phosphomethylphosphonic acid adenylate ester (PDB ID:
1O6Y)9 and into a homology model of PknA to get an
approximation of dual targeting. Hit compounds that were able
to dock into both sites with similar poses and reasonable strain
energies were prioritized for follow-up. Quinazoline 1 was
selected as our starting point because it showed relatively strong
PknB inhibition (Ki = 150 nM) and a good MIC against Mtb
H37Ra (33 μM). Although 1 did not show inhibition of PknA,
the sequence similarity between PknA and PknB (43% identity in
the kinase domains; 56% identical in the ATP binding pocket)
made it likely that we would be able to build in inhibition for both
isoforms.
Table 1. Representative PknA and PknB Enzyme Inhibition
(Ki): SAR of 2-Aryl-Substituted Quinazolines
Scheme 1 summarizes the diversification strategy used for
quinazoline 1. Since 4-(aminopyrazolyl)-substituted quinazo-
lines inhibit several mammalian host kinases as well as Mtb
kinases by binding to the hinge region, we wanted to identify
compd
R1
R2
PknB Ki (nM) PknA Ki (nM)
23
24
25
26
27
28
29
30
22
31
32
33
CH2CN
H
5
>4000
ND
H
CH2CN
38
50
15
23
23
29
88
69
24
6
H
CN
>4000
>4000
2300
1800
>4000
>4000
>4000
22
H
CONH2
H
CH2OH
CH2CONH2
CONH2
CN
H
H
Figure 1. Compound 22 docked into the ATP binding site of PknB, and
the mammalian kinase GSK3β, CDK2, and SRC. Labels identify PknB
residue numbers corresponding to cyan structure. For clarity, amino acid
side chains are not shown. Straight arrows indicate the direction of the
CO bond between the 3rd and 4th residues in the sequences shown,
and the curved arrow indicates flipping of the peptide bond between D
and G residues. Alignment was done using C-alpha atoms. Red G is the
residue that allows PKnB to undergo conformational change.
H
H
H
SO2NH2
H
H
SO2NH2
<8
<1
<8
B
DOI: 10.1021/acsmedchemlett.7b00239
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
In addition to our sp3-retaining C2-substitution strategy, we
also explored aryl C2 substitutions. Starting from the baseline 2-
phenyl analog (22: PknB Ki = 60 nM), we prepared
approximately 140 other substituted 2-phenyl analogs. Sub-
stituents at the m- and p-position of the 2-phenyl ring offered
excellent PknB potency, but showed no PknA activity. Table 1
shows SARs for a representative set of quinazoline 2-substituted
phenyl compounds. The 4-cyanomethylphenyl analog 23 had a
PknB Ki of 5 nM, while the 3-cyanomethylphenyl analog 24 had a
PknB Ki of 38 nM. Additional polar functionalities such as 4-
cyano (25), 4-CONH2 (26), and 4-hydroxymethyl (27) were
examined, and additional 3-substituted analogs were made such
as 3-acetamide 28 (Ki = 23 nM)), 3-CONH2 29 (Ki = 29 nM),
and cyano 30 (Ki = 88 nM). Overall, the substituted phenyl
analogs 24 to 30 maintained similar levels of PknB inhibitory
activity compared to their parent phenyl analog 22. It was not
until we prepared the 4-hydroxymethyl (27) and 3-acetamide
(28) compounds that we began to observe inhibition of PknA
(2.3 and 1.8 μM, respectively.) Further exploration of the H-
bonding substituents at both 3- and 4-positions provided
excellent inhibition of PknA and PknB with the primary
sulfonamide in either position (compounds 31 and 32). The 4-
SO2NH2 analog 31 reached Ki values of 24 and 22 nM for PknB
and PknA, respectively. The 3-SO2NH2 analog 32 also
demonstrated single-digit Ki values (6 and <8 nM) for PknB
and PknA. The value of this substituent was further
demonstrated when we changed the quinazoline 2-substituent
from phenyl to a thiophene ring bearing a 5-sulfonamide group
(33) and found even more potent dual PknB and PknA
inhibition. These findings are summarized in Table 1.
We next investigated opportunities for achieving selectivity
over mammalian host kinases. We chose GSK3β, CDK2, and
SRC for selectivity assessment since each was known to be
inhibited by compounds of this class. Overlay of the PknB hinge
binding region with a selection of mammalian host kinases
including GSK3β, CDK2, and SRC showed that the area
immediately surrounding the 6-chloro substituent of 1 provided
a distinct binding pocket that could be exploited for
mycobacterial kinase selectivity. As shown in Figure 1, it can
be seen that Gly-97 of PknB is “flipped” relative to the
corresponding residues of the mammalian kinases, such that
the backbone carbonyl of the adjacent residue, Asp-96, is pointed
away from the inhibitor, creating additional space that
accommodates the 6-Cl of 22.
Table 2. 6-Substituted Quinazoline with Host Kinase
Selectivity
compd
kinase Ki (nM)
34
30
35
25
50
PknB
40
9
88
46
GSK3β
CDK2
SRC
190
170
710
70
>4000
750
11
22
140
18
200
alternative hinge-binding elements in order to obtain selectivity.
We synthesized an extensive set of amino heterocyclic
replacements for the amino-pyrazole, including oxazole (2),
triazole (3), thiadiazole (4), thiazole (5, 6), pyridine (7), and
pyrimidine (8). Unfortunately, all these compounds had reduced
PknB inhibition. Our hypothesis is that these compounds
destabilized the preferred binding tautomer (in the case of the
triazole), removed a key Ar−H interaction (in the case of the
thiadiazole), or disturbed the overall geometry of interaction
with the kinase hinge area (in the cases of pyridine and
pyrimidine).
A small structure−activity relationship (SAR) study of the 6-
position, replacing Cl with F (9), methyl (10), and methoxy
(11), showed that it was possible to maintain PknB activity at this
position with a variety of substituents, an opportunity that was
later exploited for selectivity and cell-based activity (see below).
Next, an investigation of the quinazoline 2-position was
conducted. One of our goals was to preserve the sp3 character of
the azepane substituent, if possible, to retain more drug-like
physicochemical properties.16 The neutral oxazepane analog 12
had an unchanged PknB Ki of 150 nM, but when a positive charge
was introduced with the diazepane analog 13, we observed a six-
fold loss of activity. Acylation of the amine to return it to the
neutral state as either the amide or carbamate analog (14 and 15)
recapitulated the original PknB potency of 150 and 230 nM,
respectively. In accord with this developing SAR, the
diazepanone analogs 16 and 17 also gave PknB Ki potencies of
190 and 310 nM, respectively. Contraction of the diazepane ring
to the six-membered piperazine rings (18 and 19) gave weaker
activity, but surprisingly so did the acylated version found in 20.
We surmise that the much larger volume of the N-benzoyl group
was responsible for its diminished activity.
To further substantiate the Mtb PknB selectivity of inhibitory
compounds over mammalian kinases, 6-Cl quinazoline analogs
were compared with 6-H compounds as summarized in Table 2.
As predicted, we found similar PknB potency for both 6-Cl and 6-
H quinazolines (34 vs 30 and 35 vs 25, respectively), while
inhibition of the three mammalian kinases was reduced 5−50-
fold with the 6-Cl compounds (30 and 25).
Thus far, we had been able to improve enzyme potency and
selectivity for the quinazoline series. However, we were unable to
improve antimycobacterial activity since, despite lowering PknB
inhibition roughly 100-fold, the MIC against Mtb H37Ra
remained at 33 μM or higher. It is possible that our inhibitors
failed to penetrate the Mtb cell wall, or they were subject to rapid
efflux.
To follow up on this result, a library of 48 amine compounds
that were substituted at the quinazoline C2 position was
generated. We selected a set of primary and secondary amines
containing one or fewer aromatic rings, with the following
constraints: MW < 200, PSA < 55 Å, and calculated logP < 2.5.
Library members were found to have PknB Ki values ranging
from >4 μM to <100 nM. The cyclobutylamino analog (21) was
the most potent PknB inhibitor (Ki = 72 nM), while most of the
other C2-amino analogs plateaued at approximately 100 nM
PknB Ki without further improvement. These findings are
summarized in Scheme 1.
Some antibacterial compounds can cross the bacterial cell wall
through channel-forming porins that enable the diffusion of
small, hydrophilic solutes.17 Efflux pumps, while not well studied
in Mtb, tend to be particularly active on lipophilic substrates.
Hence, reducing lipophilicity can be a successful strategy both for
C
DOI: 10.1021/acsmedchemlett.7b00239
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
Scheme 2. Compounds 36−38 and 46−49 and the Representative Syntheses of 37, 38, and 49
a
b
c
N,N-Dimethylaniline (2 equiv), POBr3 (3 equiv), toluene, 90 °C, 2 h, 83%. 5-Cyclopropyl-1H-pyrazol-3-amine, EtOH, RT, 4 h, 78%. 3- or 4-
(Pinacolatoboranyl)-benzenesulfonamide for 36 and 37, or 44 for 38 and 56, Pd(dppf)Cl2 (0.1 equiv), Na2CO3 (4 equiv), dioxane/H2O (4:1), 100
d
e
°C, 15 h. t-Butylamine, dioxane, 20 °C, 4 h, 94%. Bis-pinacolatodiboron (1.2 equiv), Pd(dppf)Cl2 (0.1 equiv), CH3COOK (4.0 equiv), dioxane,
100 °C, 3 h, 70%. BCl3, DCM, 20 °C, 6 h. NIS, AcOH, RT, 7 h, 95%. TMS-acetylene (1.2 equiv), (Ph3P)2PdCl2 (0.25 equiv), CuI (0.25 equiv),
EtOAc, RT, 4 h, 46%. N,N-Dimethylaniline, POBr3, 100 °C, 50 min, toluene, 51%.
f
g
h
i
Table 3. Enzyme Inhibition and MIC (μM) of Pyrimidines
against Mycobacterium tuberculosis
truncated the quinazoline ring into a simple monocyclic
pyrimidine ring. In addition, we incorporated the sulfonamide
substituent to provide potent PknA inhibition. Toward this
objective, we synthesized the pyrimidine analogs 36, 37, and 38
(Scheme 2). The preparation of 37 and 38 is described in
Scheme 2. Treatment of 5-chloropyrimidine-2,4-diol (39) with
POBr3 gave the 2,4-dibromo-5-chloro-pyrimidine (40), which
reacted with 5-cyclopropyl-1H-pyrazol-3-amine to afford 41.
Suzuki coupling of 41 with 3-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)benzene-sulfonamide then produced target
compound 37. For 38, the synthesis started with the reaction of
5-bromothiophene-2-sulfonyl chloride (42) with 2-methylpro-
pan-2-amine to form 5-bromo-N-(tert-butyl)thiophene-2-sulfo-
namide (43), which then was converted to the pinacol-boron
compd
PknB Ki (nM)
PknA Ki (nM)
Mtb MIC (μM)
36
37
38
46
47
48
13
17
<1.3
4
2900
<8
9
>100
8.3
6.25
4.2
18
12
18
<1
4
3.1
4.7
improving permeability through the bacterial cell wall and for
avoiding efflux. We set about reducing the lipophilicity of our
most potent kinase inhibitors while also minimizing their size.
Similar to the approach described by Chapman et al.,14 we
D
DOI: 10.1021/acsmedchemlett.7b00239
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
crystallography to solve the structure of compound 38
complexed with PknA and also complexed with PknB. Figure 2
shows that the aminopyrazole provides the key hinge-binding
interactions, with the cyclopropyl group interacting in the small
hydrophobic pocket provided by Met95 and Val27 in PknA and
by the analogous residues Met92 and Val25 in PknB. These
structures also provide some clues to understanding why the
SO2NH2 group provides enhanced inhibition of PknA. The
−SO2NH2 group bound to PknA shows three relatively close
polar contacts, to the side chains of Lys42, Thr158, and Asn146
(Figure 3).
We then investigated the C5-position on the pyrimidine ring
to determine if we could again exploit the “flipped glycine”
pocket for selectivity over mammalian kinases. Toward this
effort, the 5-acetylenyl compound 49 was made as shown in
Scheme 2. The 5-H analog 46 only showed two-fold selectivity
versus CDK2, while the 5-Cl compound 38 gave at least a 15-fold
improvement in Mtb PknB selectivity vs CDK2. The sterically
more demanding 5-acetylenyl compound 49 demonstrated
greater than 100-fold selectivity vs CDK2, with increased PknB
potency (Table 4) while maintaining Mtb H37Ra MIC at 4.7
μM.
Figure 2. Compound 38 bound to PknA (left) and PknB (right), and
with key amino acid residues that interact with 38.
Finally, we wanted to demonstrate a clear link between kinase
inhibition and the observed antimycobacterial activity. For this
purpose, we designed the N-methyl version of compound 48,
where the kinase hinge-binding nitrogen is blocked. Compound
57 was synthesized as shown in Scheme 3. The starting material
5-cyclopropyl-1-methyl-1H-pyrazol-3-amine 58 was prepared
from simple methylation of 3-cyclopropyl-1H-pyrazol-5-amine.
Reaction with 2,4-dibromo-5-methylpyrimidine followed by
Suzuki coupling with intermediate 44 and N-t-butyl removal
gave compound 57. Consistent with our expectations,
compound 57 lost both PknA and PknB enzyme activities (Ki
both >4 μM) as well as antimycobacterial activity (Mtb H37Ra
MIC > 100 μM).
In conclusion, starting from a 1078-compound screen of a
focused collection of diverse kinase-inhibitory compounds, we
identified a weak and nonselective quinazoline-based inhibitor of
PknB with a quinazoline core that lacked activity against PknA
and showed weak antimycobacterial activity. We successfully
optimized this series to achieve potent, dual-targeting inhibitors
with activity against both PknB and PknA. We identified a
structural pocket unique to the mycobacterial kinases that we
exploited for two related but distinct series. Compound
modifications that filled this pocket gave a significant selectivity
window over mammalian kinases. Reducing compound lip-
ophilicity (for example, quinazoline 33 has a calculated logP of
4.1, while for the analogous pyrimidine 38, 2.7) greatly improved
Figure 3. X-ray crystal structures for 38 complexed to PknA (orange)
and PknB (blue.) Structures were overlaid using PyMol for alignment.
Dotted lines represent close polar contacts discussed in the text.
Table 4. Selectivity of PknB Inhibitors with Reduced
Inhibition of the Mammalian Kinase CDK2
compd
PknB Ki (nM)
PknA Ki (nM)
CDK2 Ki (nM)
selectivity
46
38
49
4
18
9
7
2
<1.3
<2
17
147
>13
>74
11
ester 44. Suzuki coupling, followed by removal of the t-butyl
group with BCl3 provided 38.
Each of the three sulfonamide analogs 36, 37, and 38 were
potent PknB and PknA inhibitors, while 37 and 38 demonstrated
single-digit micromolar MIC against Mtb H37Ra. Additional 5-
substituted pyrimidine analogs 46, 47, and 48 were prepared and
all showed excellent PknB and PknA potency that translated to
improved Mtb H37Ra MIC values (Table 3).
To confirm our proposed binding modes and to understand
the SAR observed for PknA and PknB, we used X-ray
Scheme 3. Synthesis of 57
a
b
c
t-BuOK, 18-crown-6, CH3I, Et2O, 0 °C, 3 h, 43%. 2,4-Dibromo-5-methylpyrimidine, EtOH, RT, 8 h, 70%. 44, Pd(dppf)Cl2 (0.1 equiv), Na2CO3
d
(4 equiv), dioxane/H2O (4:1), 100 °C, 15 h, 28%. BCl3, DCM, RT, 30 min, 71%.
E
DOI: 10.1021/acsmedchemlett.7b00239
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
(2) Av-Gay, Y.; Jamil, S.; Drews, S. J. Expression and characterization of
the Mycobacterium tuberculosis serine/threonine protein kinase PknB.
Infect. Immun. 1999, 67, 5676−5682.
the antibacterial activity, with the best compounds in the series
active against Mtb H37Ra with MIC in the range of 3−5 μM (1−
2 μg/mL) and potency commensurate with clinically used
antibiotics. Thus, pending further lead optimization efforts,
kinase inhibition could be a valuable approach to the discovery of
much needed new TB therapeutics.
(3) Av-Gay, Y.; Everett, M. The eukaryotic-like Ser/Thr protein
kinases of Mycobacterium tuberculosis. Trends Microbiol. 2000, 8, 238−
244.
(4) Sassetti, C. M.; Boyd, D. H.; Rubin, E. J. Genes required for
mycobacterial growth defined by high density mutagenesis. Mol.
Microbiol. 2003, 48, 77−84.
(5) Sassetti, C. M.; Rubin, E. J. Genetic requirements for mycobacterial
survival during infection. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 12989−
12994.
(6) Fernandez, P.; Saint-Joanis, B.; Barilone, N.; Jackson, M.; Gicquel,
B.; Cole, S. T.; Alzari, P. M. The Ser/Thr protein kinase PknB is essential
for sustaining mycobacterial growth. J. Bacteriol. 2006, 188, 7778−7784.
(7) Szekely, R.; Waczek, F.; Szabadkai, I.; Nemeth, G.; Hegymegi-
Barakonyi, B.; Eros, D.; Szokol, B.; Pato, J.; Hafenbradl, D.; Satchell, J.;
Saint-Joanis, B.; Cole, S. T.; Orfi, L.; Klebl, B. M.; Keri, G. A novel drug
discovery concept for tuberculosis: inhibition of bacterial and host cell
signalling. Immunol. Lett. 2008, 116, 225−231.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acsmedchem-
lett.7b00239.
Preparation of all final compounds, kinase and MIC data,
PknA and PknB expression, biological assay protocols, and
crystallization conditions (PDF)
Spectra of compounds (PDF)
Accession Codes
(8) Young, T. A.; Delagoutte, B.; Endrizzi, J. A.; Falick, A. M.; Alber, T.
Structure of Mycobacterium tuberculosis PknB supports a universal
activation mechanism for Ser/Thr protein kinases. Nat. Struct. Biol.
2003, 10, 168−174.
PDB codes for the X-ray crystal structures described in this study
have been deposited in the PDB under the accession codes 6B2P
(PknA) and 6B2Q (PknB).
(9) Ortiz-Lombardia, M.; Pompeo, F.; Boitel, B.; Alzari, P. M. Crystal
structure of the catalytic domain of the PknB serine/threonine kinase
from Mycobacterium tuberculosis. J. Biol. Chem. 2003, 278, 13094−
13100.
(10) Prisic, S.; Dankwa, S.; Schwartz, D.; Chou, M. F.; Locasale, J. W.;
Kang, C. M.; Bemis, G.; Church, G. M.; Steen, H.; Husson, R. N.
Extensive phosphorylation with overlapping specificity by Mycobacte-
rium tuberculosis serine/threonine protein kinases. Proc. Natl. Acad. Sci.
U. S. A. 2010, 107, 7521−7526.
(11) Chou, M. F.; Prisic, S.; Lubner, J. M.; Church, G. M.; Husson, R.
N.; Schwartz, D. Using bacteria to determine protein kinase specificity
and predict target substrates. PLoS One 2012, 7, e52747.
(12) Kang, C. M.; Abbott, D. W.; Park, S. T.; Dascher, C. C.; Cantley,
L. C.; Husson, R. N. The Mycobacterium tuberculosis serine/threonine
kinases PknA and PknB: substrate identification and regulation of cell
shape. Genes Dev. 2005, 19, 1692−1704.
(13) Ortega, C.; Liao, R.; Anderson, L. N.; Rustad, T.; Ollodart, A. R.;
Wright, A. T.; Sherman, D. R.; Grundner, C. Mycobacterium
tuberculosis Ser/Thr protein kinase B mediates an oxygen-dependent
replication switch. PLoS Biol. 2014, 12, e1001746.
(14) Chapman, T. M.; Bouloc, N.; Buxton, R. S.; Chugh, J.; Lougheed,
K. E.; Osborne, S. A.; Saxty, B.; Smerdon, S. J.; Taylor, D. L.; Whalley, D.
Substituted aminopyrimidine protein kinase B (PknB) inhibitors show
activity against Mycobacterium tuberculosis. Bioorg. Med. Chem. Lett.
2012, 22, 3349−3353.
(15) Lougheed, K. E.; Osborne, S. A.; Saxty, B.; Whalley, D.; Chapman,
T.; Bouloc, N.; Chugh, J.; Nott, T. J.; Patel, D.; Spivey, V. L.;
Kettleborough, C. A.; Bryans, J. S.; Taylor, D. L.; Smerdon, D. J.; Buxton,
R. S. Tuberculosis (Oxford, U. K.) 2011, 91, 277−286.
(16) Lovering, F.; Bikker, J.; Humblet, C. Escape from flatland:
increasing saturation as an approach to improvimng clinical success. J.
Med. Chem. 2009, 52, 6752−6756.
AUTHOR INFORMATION
Corresponding Author
*E-mail: tiansheng_wang@vrtx.com. Phone: 617-341-6484.
ORCID
Tiansheng Wang: 0000-0001-5073-2670
Jeremy Green: 0000-0003-4544-6412
■
Present Addresses
§Hennessy Research Associates, 12700 Johnson Drive, Shawnee,
Kansas 66216, United States. E-mail: bhanzelka@
hennessyresearch.com.
∥Versatope Therapeutics, MassChallenge Design and Innovation
Center, 21 Drydock Lane, Boston, Massachusetts 02210, United
States. E-mail: Christopher.locher@versatope.com.
⊥GlaxoSmithKline R&D China, No. 3 Building, 898 Halei Road,
Shanghai 201203, China. E-mail: shaohui.wang@gsk.com.
#P.O. Box 2241, Acton, Massachusetts 01720, United States. E-
mail: jtexvertex@yahoo.vom.
∇Department of Microbiology, University of Iowa, 51 Newton
Road, Iowa City, Iowa 52242, United States. E-mail: ute-muh@
uniowa.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We would like to express our gratitude to ChemPartner’s
scientists Jianguo Chen, Xiaoming Li, Zhongtai Piao, and
Jingwen Zang for their synthetic (J.C., X.L., Z.P.) and biological
(J.Z.) assistance. We thank Barry Davis and Brian Ledford, both
of Vertex, for HRMS and NOE studies.
(17) Vargas, J. R.; Stanzl, E. G.; Teng, N. N. H.; Wender, P. A. Cell-
penetrating, guanidinium-rich molecular transporters for overcoming
efflux-mediated multidrug resistance. Mol. Pharmaceutics 2014, 11,
2553−2565.
ABBREVIATIONS
■
Mtb, Mycobacterium tuberculosis; PknA, Protein kinase A; PknB,
Protein kinase B; SAR, structure−activity relationship; GSK 3-
beta, glycogen synthase kinase 3 beta; CDK2, cyclin-dependent
kinase 2; MIC, minimum inhibitory concentration; PSA, polar
surface area
REFERENCES
(1) WHO. Global tuberculosis report, 2013.
■
F
DOI: 10.1021/acsmedchemlett.7b00239
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX