Discovery of novel, potent, and selective inhibitors of 3-phosphoinositide-dependent kinase (PDK1)

Article abstract of DOI:10.1021/jm201019k

Analogues substituted with various amines at the 6-position of the pyrazine ring on (4-amino-7-isopropyl-7H-pyrrolo[2,3-d]pyrimidin-5-yl)pyrazin-2- ylmethanone were discovered as potent and selective inhibitors of PDK1 with potential as anticancer agents. An early lead with 2-pyridine-3-ylethylamine as the pyrazine substituent showed moderate potency and selectivity. Structure-based drug design led to improved potency and selectivity against PI3Kα through a combination of cyclizing the ethylene spacer into a saturated, five-membered ring and substituting on the 4-position of the aryl ring with a fluorine. ADME properties were improved by lowering the lipophilicity with heteroatom replacements in the saturated, five-membered ring. The optimized analogues have a PDK1 K_i of 1 nM and >100-fold selectivity against PI3K/AKT-pathway kinases. The cellular potency of these analogues was assessed by the inhibition of AKT phosphorylation (T308) and by their antiproliferation activity against a number of tumor cell lines. (Figure presented)

Full text of DOI:10.1021/jm201019k

Article
pubs.acs.org/jmc
Discovery of Novel, Potent, and Selective Inhibitors of
3-Phosphoinositide-Dependent Kinase (PDK1)
Sean T. Murphy,*,† Gordon Alton,† Simon Bailey,† Sangita M. Baxi,† Benjamin J. Burke,†
Thomas A. Chappie,‡ Jacques Ermolieff,† RoseAnn Ferre,† Samantha Greasley,† Michael Hickey,†
John Humphrey,‡ Natasha Kablaoui,§ John Kath,† Steven Kazmirski,§ Michelle Kraus,† Stan Kupchinsky,†
John Li,† Laura Lingardo,† Matthew A. Marx,† Dan Richter,† Steven P. Tanis,† Khanh Tran,†
William Vernier,† Zhi Xie,† Min-Jean Yin,† and Xiao-Hong Yu†
†Pfizer Global Research and Development, 10770 Science Center Drive, San Diego, California 92121, United States
‡Pfizer Global Research and Development, 558 Eastern Point Road, Groton, Connecticut 06340, United States
§Pfizer Global Research and Development, 620 Memorial Drive, Cambridge, Massachusetts 02139, United States
S
* Supporting Information
ABSTRACT: Analogues substituted with various amines at the
6-position of the pyrazine ring on (4-amino-7-isopropyl-7H-
pyrrolo[2,3-d]pyrimidin-5-yl)pyrazin-2-ylmethanone were dis-
covered as potent and selective inhibitors of PDK1 with potential
as anticancer agents. An early lead with 2-pyridine-3-ylethylamine
as the pyrazine substituent showed moderate potency and selec-
tivity. Structure-based drug design led to improved potency
and selectivity against PI3Kα through a combination of cyclizing
the ethylene spacer into a saturated, five-membered ring and
substituting on the 4-position of the aryl ring with a fluorine. ADME properties were improved by lowering the lipophilicity with
heteroatom replacements in the saturated, five-membered ring. The optimized analogues have a PDK1 Ki of 1 nM and >100-fold
selectivity against PI3K/AKT-pathway kinases. The cellular potency of these analogues was assessed by the inhibition of AKT
phosphorylation (T308) and by their antiproliferation activity against a number of tumor cell lines.
in murine mammary cells led to transformation, and isografts
into syngenic mice led to tumor formation.5 A recent review
paper discusses evidence that PDK1 is implicated in human
cancer and in resistance to chemotherapeutics.6 As such, there
is considerable activity in the pharmaceutical industry to
develop potent PDK1 inhibitors.7−10
INTRODUCTION
■
3′-Phosphoinositide dependent kinase-1 (PDK1) belongs to the
AGC family of kinases and was first identified by Kobayashi and
Cohen.1 PDK1 is part of the phosphoinositide-3 kinase
(PI3K)/AKT pathway, which is one of the major cellular
pathways involved in cell survival, differentiation, growth, and
protein expression. This signal transduction pathway is
downstream of several growth factor receptor tyrosine kinases
(RTK’s), such as the epidermal growth factor receptor, the
insulin-like growth factor receptor-1, and the insulin receptor.
Upon binding, these RTK’s activate PI3Kα, which catalyzes the
phosphorylation of phosphatidylinositol-(3,4)-biphosphate to
generate biologically active phosphaditylinositol-(3,4,5)-tri-
phosphate (PIP3). The formation of PIP3 triggers the
colocalization at the membrane of two Ser/Thr kinases,
PDK1 and AKT, both of which bind to PIP3 through their
pleckstrin homology (PH) domains. PDK1 phosphorylates the
activation loop of AKT at threonine-308 (T308), thereby
activating AKT to drive downstream signaling.2
A number of PDK1 inhibitors have been reported in the
literature, some of which are shown in Figure 1. The earliest
examples (Figure 1A−C)11−13 had been published at the time
our work began. The work of Stauffer et al.13 on the SAR of
inhibitor C demonstrated that the cellular response when
measuring AKT-T308 could be complicated by a lack of PI3K
selectivity, and their cellular activity correlated better with PI3K
inhibition than PDK1 inhibition. PI3K selectivity was not
reported for the other early inhibitors (A and B), and we were
concerned that they were not selective enough to test the
antiproliferative effects of PDK1 inhibition. Furthermore, the
report for inhibitor B showed a lack of correlation between
PDK1 inhibition and cellular potency,12 while inhibitor A is
structurally similar to sunitinib, which is a potent inhibitor of
many kinases.14
PDK1 has been shown to be a critical regulator of the PI3K/
AKT pathway, which is frequently activated in cancer cells due
to either the loss of phosphatase and tensin homologue
(PTEN) activity or the increase of PI3K/AKT activity.3
A hypomorphic mutation of PDK1 protects PTEN± mice from
developing a wide range of tumors.4 Overexpression of PDK1
Received: July 30, 2011
Published: October 31, 2011
© 2011 American Chemical Society
8490
dx.doi.org/10.1021/jm201019k|J. Med. Chem. 2011, 54, 8490−8500

Journal of Medicinal Chemistry
Article
Figure 1. Selected PDK1 inhibitors disclosed in the literature.
More recently (while this manuscript was in preparation),
three more papers have appeared describing potent and
selective PDK1 inhibitors (Figure 1D−F).8−10 Of particular
note is compound F, described by co-workers at Sunesis and
Biogen-Idec15 and profiled by researchers at Merck.10 It is
equipotent with the optimized analogues described in this work
and has excellent selectivity, and their biological results are
consistent with our findings.
To enhance our confidence in the preclinical pharmaco-
logical assessment of PDK1 as an anticancer target, we sought
potent PDK1 inhibitors with excellent selectivity over the other
pathway kinases: PI3K, AKT, S6K, and mTOR. In the present
study, we describe the discovery of 3-carbonyl-4-amino-
pyrrolopyrimidine analogues as potent PDK1 inhibitors with
more than 100-fold selectivity against 38 of 39 kinases (97%),
including the PI3K pathway kinases PI3K, AKT, S6K, and
mTOR.
solvent with cesium fluoride as a catalyst led to efficient
displacement of the chloride. This reaction was broadly
effective and provided access to several hundred analogues
from a wide variety of primary and secondary amines.
The cis-2-arylcyclopentylamines (6a and 6d, X = CH2) were
synthesized by the method of Shepherd et al.16 as shown in
Scheme 2. The N-methyl analogue (6e) was synthesized by
standard chemistry (see Scheme 2). The same general sequence
was utilized to make the 3,4-disubstituted tetrahydrofuran (6b)
and 3,4-disubstituted pyrrolidine (6c) analogues. After 6c was
coupled to 1, the Boc group was removed with HCl in
dichloromethane to give the final compound.
The synthesis of the 3,4-disubstituted tetrahydrofuran 7
(Scheme 3) followed the procedure of Andreotti and Ward.17
The 3,4-disubstituted pyrrolidine analogues 13a−c were made
from aldehyde 8 using a phosphine-catalyzed [3 + 2]
cycloaddition first disclosed by Lu and Xu.18 The elimination
of the sulfinate under basic conditions gave the pyrrole, which
was then protected with the Boc group to give 10. Reduction
with Adams’ catalyst gave the desired cis-stereochemistry on the
pyrrolidine ring 11. After saponification of the ester and Boc
removal by treatment with acid, the amine was variously
substituted with a Cbz, formyl, or acetyl group to give
intermediates 12a−c. The carboxylic acid was then converted
into an amine via a Curtius rearrangement to give the desired
amine products 13a−c. Because a significant amount of des-
Boc product was observed after the Curtius rearrangement, Boc
anhydride was added to the reaction mixture to form the Boc-
protected amine, which could then be easily purified by column
chromatography.
SYNTHESIS
■
The general synthesis for the analogues in this series is shown
in Scheme 1. Reaction of chloropyrazine 1 (see Supporting
Information for synthesis) with excess amine in a polar, aprotic
a
Scheme 1.
After coupling intermediate 13a to 1, the Cbz group was
removed by hydrogenolysis, which was followed by treat-
ment with acetic anhydride to give the desired final product
a
Conditions: (a) diisopropylethylamine, cesium fluoride, anhydrous
DMSO, 110−140 °C.
8491
dx.doi.org/10.1021/jm201019k|J. Med. Chem. 2011, 54, 8490−8500

Journal of Medicinal Chemistry
Article
a
Scheme 2.
a
Conditions: (a) 4-Z-PhMgBr, CuI, THF (Z = H or F); (b) phthalimide, diisopropyl azodicarboxylate, resin-bound PPh3, THF; (c) anhydrous
hydrazine, toluene, 90 °C (6a, 6b, and 6d) or hydrazine monohydrate, ethanol, 20 °C (6c); (d) Boc2O, CH2Cl2; (e) NaH, MeI, DMF; (f) HCl.
a
through an intermediary water. The displacement of the water in
the ATP/PDK1 structure could contribute to the high binding
efficiency of 14.20 Various replacements for the isopropyl group
were tried, but the more lipophilic groups (e.g., cyclobutyl,
cyclopentyl, sec-butyl) were not significantly more potent, and the
more polar groups (e.g., ethyl, hydroxyethyl, aminoethyl) lost 2-
fold or more in potency (data not shown).
Scheme 3.
Since PI3K inhibition could itself reduce the amount of
phospho-AKT and confound the analysis of inhibition of T308
phosphorylation due to PDK1, the program placed an emphasis
on obtaining analogues selective against PI3Kα. Initial analogues
afforded a few promising leads for building selective com-
pounds such as the pyridyl-containing analogue 15 (Figure 3) with
7-fold selectivity against PI3K (in contrast, 14 only had 2-fold
selectivity against PI3K). An analysis of the cocrystal structure led
to hypotheses to improve both the potency in PDK1 and the
selectivity against PI3K, as discussed below.
a
Conditions: (a) TsNH2, but-2-ynoic acid ethyl ester, PBu3, BF3-
etherate, toluene, reflux; (b) KOtBu, THF; (c) Boc2O, 4-DMAP; (d)
PtO2, H2, 45 psi, ethanol; (e) HCl(aq), dioxane, 70 °C; (f) R = OBn,
CbzCl, NaHCO3; R = Me, Ac2O; R = H; ethyl formate, 80 °C; (g)
diphenylphosphoryl azide, triethylamine, toluene, 70 °C; then KOtBu,
20 °C; then Boc2O; (h) HCl, dioxane, methylene chloride.
Figure 4 shows the cocrystal structure of 15 in PDK1. The
substituents on the ethyl linker are in a gauche conformation (see
lower right of Figure 4), rather than the usual anti conformation.
The gauche conformation is necessary to efficiently fill the pocket
under the G-loop of PDK1. To improve potency, we envisioned
locking the flexible ethyl linker into a five-membered ring which
would position the aryl ring under the G-loop and in the
bioactive conformation. The pyridine ring was replaced with a
phenyl ring because the nitrogen was not close enough to Ser-94
to make a significant hydrogen bond (3.3 Å) and because
unblocked pyridines are associated with increased risk of CYP
inhibition.21 No loss in potency was anticipated because the
phenyl analogue of 15 was equipotent (PDK1 Ki 29 nM).
Figure 5 shows 15 modeled into PI3K-γ.22 Analysis of
analogue 15 in PI3K-γ led to two hypotheses to improve the
selectivity. First, substitution of the 4-position of the terminal
aryl ring (pyridyl in 15) with a fluorine was hypothesized to
bump into Pro-810 in PI3K-γ. Second, methylation of the NH
attached to the pyrazine was expected to bump into Ile-963.
Both the fluorine and methyl groups were predicted to be at
least tolerated by PDK1 and possibly beneficial to binding.
Figure 6 shows the dramatic improvement in potency and
selectivity in the ring-locked analogue 17 compared to the
open-chain analogue 15. The ca. 15-fold increase in potency
(ΔG = 1.6 kcal/mol) is consistent with the energy gain from
freezing two rotatable bonds (est. 0.7 kcal/mol per rotatable
bond) along with additional van der Waals interactions from
one of the linker ring atoms (est. 0.8 kcal/mol per sp3
carbon).23 Furthermore, in agreement with the predictions
from the crystal structure, the selectivity against PI3Kα was
(26; see Table 1). During the hydrogenolysis of the adduct of
13a and 1, an over-reduced product was observed by LC/MS
(mass plus two, product not identified), which lowered the
overall yield. To avoid the low yield in the hydrogenolysis
reaction when making the fluoro derivatives (13b and 13c), the
formyl and acetyl groups were installed before coupling to 1, as
shown in Scheme 3. The free amine analogue (23, Table 1) was
made from the formyl derivative 24 by treating with aqueous
hydrochloric acid in methanol at ambient temperature.
Single enantiomers were desired and were accessed by the
separation of enantiomers using chiral SFC separation of
intermediates before coupling to 1 or on the final analogues (see
Experimental Section for details). The absolute stereochemistry of
the most active enantiomer was determined by crystallization in
PDK1 for several analogues (18, 20, 21, and 26; see Figure 7 for
one example), all of which showed the absolute stereochemistry as
depicted in Schemes 2 and 3 and in Table 1 (see below).
RESULTS AND DISCUSSION
■
The PDK1 program began with a screen of selected compounds
from the Pfizer corporate file. Of the resultant hits, the ligand
efficient (LE)19 inhibitor 14 (Figure 2) was selected for follow-up,
including cocrystallization with PDK1. Analysis of the X-ray
structure of 14 bound in PDK1 revealed a hydrogen bond
between the carbonyl of the ligand and Thr-222. Analysis of the
structure of ATP in PDK1 shows a hydrogen bond to Thr-222
8492
dx.doi.org/10.1021/jm201019k|J. Med. Chem. 2011, 54, 8490−8500

Journal of Medicinal Chemistry
Article
Figure 2. ATP and cmpd 14 shown with key interaction to PDK1 protein.
improved by the substitution of a fluoro at the 4-position of the
phenyl ring (18) and by methylation of the pyrazine-NH (19).
Because the N-methylated analogue 19 showed N-demethyla-
tion as a major clearance pathway and because the potency and
selectivity were not significantly different from those of the des-
methyl analogue 18, further analogues focused on the des-
methyl series. The X-ray structures of ether analogue 20 (see
Table 1 below for structure) and others (18, 21, and 26, data
not shown) in PDK1 confirmed the expected placement of the
aryl ring under the G-loop (see Figure 7).
Figure 3. Early lead 15 with moderate selectivity against PI3K.
Table 1 shows some in vitro ADME and selectivity data for
the cyclopentyl linker analogue 18. This analogue has poor
Table 1. In Vitro ADME and Selectivity for Ring-Locked
a
Analogues
Figure 4. X-ray crystal structure of 15 (orange stick representation)
bound to PDK1 (shown in green backbone trace).
a
b
All compounds are single enantiomers with ≥95% ee. % inhibition
at 3 uM; maximum value from CYP-1A2, -2C9, -2D6, and -3A4; the
most strongly inhibited enzyme was CYP-3A4 in all cases in the table.
c
d
Solubility tested at Analiza,25 pH 6.5, in μM. Passive permeability in
e
canine kidney cells, ×10−6 cm/s. Human liver microsome clearance,
Figure 5. X-ray crystal structure of 16 (blue stick representation)
bound to PI3K-γ22 and analogue 15 (orange stick representation)
modeled into the PI3K pocket by aligning the hinge binder.
f
μL/(min mg). Ki against kinase shown: green, >100 nM; yellow, 30−
100 nM; red, <30 nM.
8493
dx.doi.org/10.1021/jm201019k|J. Med. Chem. 2011, 54, 8490−8500

Journal of Medicinal Chemistry
Article
Figure 6. Ring-locked derivatives of 15.
permeability, and both analogues had excellent selectivity against all
the PI3K-pathway kinases in the panel. Compound 24 improved
the potency over 100-fold from 14 and has an LE of 0.38.26
Since the acetyl group in analogue 25 gave a boost in
selectivity over the formyl group in 24, the des-fluoro analogue
of 25 was made to see if selectivity could be maintained against
PI3K and the other pathway kinases. Indeed, the des-fluoro
analogue 26 maintained potency and selectivity. The in vitro
ADME was similar to 25 with the exception of higher CYP
inhibition (54% at 3 μM against CYP3A4).
The in vitro cellular potencies (reduction of pAKT-T308) for
analogues 24−26 are shown in Table 2. The two fluorinated
Table 2. In Vitro Cellular Potency
a
cell IC50 (nM) ± std dev
cmpd
PDK1 Ki (nM)
SKOV3
A549
Figure 7. Superposition of the X-ray crystal structures of PDK1 bound
to 15 (orange sticks/green backbone trace) and 20 (pink sticks and
backbone trace).
24
25
26
0.7
1.4
1.4
70 ± 30 (n = 8)
70 ± 40 (n = 6)
200 ± 50 (n = 5)
50 ± 20 (n = 2)
80 ± 10 (n = 2)
300 ± 130 (n = 2)
a
solubility, high clearance, and high CYP3A4 inhibition, likely
arising in part due to the compound’s high lipophilicity.24 Further
designs targeted analogues with clogP < 3 to improve the overall
probability of favorable ADME properties. Because the
4-fluorophenyl ring demonstrated good selectivity against PI3K,
that moiety was left intact and the lipophilicity was lowered by
replacing the methylenes in the cyclopentyl ring with
heteroatoms. Further SAR on replacements for the 4-
fluorophenyl ring will be reported in a future publication.
Replacement of the cyclopentyl ring with two isomeric
tetrahydrofuran rings gave analogues 20 and 21, lowering the
clogP by 1.7 units. The PDK1 potency was maintained while
the permeability and solubility improved in both cases. Isomer
21 also reduced the CYP inhibition although the selectivity
against S6K was reduced in this analogue to ca. 50-fold.
Substitution of the methylenes with nitrogen gave the basic
pyrrolidine analogues 22 and 23, which lowered the clogP by
about the same degree as that for the ethers (measured LogD
values at pH 7.4 were 1.7 and 2.4, respectively). Once again, the
PDK1 potency was maintained. While the solubility and CYP
inhibition were good, the passive permeability suffered, possibly
from a desolvation penalty of the basic amines. The selectivities
for these two isomers were dramatically different, with 22
becoming more potent against S6K and AKT.
Inhibition of pAKT-T308 using the ELISA assay in the tumor cell
lines SKOV3 or A549.
pyrrolidine amide analogues, 24 and 25, showed good potency in
two different cell lines. The des-fluoro analogue 26 was less potent
in the cell assay. The approximate 100-fold shift between Ki and
cellular IC50 was consistent with the analogues in this series.
To assess broader selectivity, analogues 24−26 were profiled
against a larger panel of kinases from Invitrogen (kinases
inhibited at 70% or more are shown in Table 3; see Supporting
Table 3. Kinases Inhibited at >70% at 1 μM
a
% inhibition at 1 μM
gene/kinase
cmpd 24
cmpd 25
cmpd 26
CHEK2/Chk2
MARK1
98
84
81
74
74
66
62
99
74
80
63
58
76
73
94
47
54
59
30
42
32
STK6/Aurora A
NTRK1/TrkA
PRKACA
GSK3β
CAMK2A
a
All values an average of n = 2 or greater.
Information for results from the full panel of 36 kinases). Both
compounds with the 4-fluoro substituent (24 and 25) inhibited
five other kinases at >70% at 1 μM, while the des-fluoro
analogue 26 only inhibited one other kinase (Chk2) above 70%.
Because activity against the kinases MARK1, Aurora A, TrkA,
and GSK3β might contribute to antiproliferative effects in
Formylation and acetylation of the basic amine in 23 gave 24
and 25, respectively. These analogues reduced the clogP by more
than 2 units from 18 while maintaining potency. As expected for
analogues in this clogP range, the solubility, CYP inhibition, and
permeability were moderate to good; however, the HLM clearance
was still high. The additional polarity did not adversely affect the
8494
dx.doi.org/10.1021/jm201019k|J. Med. Chem. 2011, 54, 8490−8500

Journal of Medicinal Chemistry
Article
nologies (Carlsbad, CA). The recombinant GST-tagged P70S6K
(a.a. 1-421), His-tagged AKT1 (full length), and GST-tagged mTOR
(a.a. 1360-2549) were purchased from Invitrogen. Recombinant
human His-tagged PDK1 catalytic domain (a.a. 51-359) was made
in-house at Pfizer La Jolla.29
All the kinetic experiments were conducted at room temperature
(∼22 °C), and the concentrations of reagents reported in the
following sections are reported as final in the media buffer. All the
experimental data were generated in duplicate and were fitted using
nonlinear regression analysis software, GraphPad Prism 5.30
culture and confound the interpretation of the results for PDK1
inhibition, the IC50 values for these kinase were measured
(Table 4). While all three analogues are greater than 100-fold
Table 4. IC50 against Off-Target Kinases
IC50 for kinase (nM)
kinase
cmpd 24
cmpd 25
cmpd 26
AurA
140
180
110
200
460
340
590
620
820
860
GSK3β
MARK1
TrkA
Selectivity Screening against Kinase Panel. Ki values were
determined for AKT1 and P70S6K (S6K), using a Caliper EZ-reader
instrument from Caliper LifeScience. For AKT1, kinetic studies were
conducted in the presence of 2.2 nM enzyme, 20 μM ATP, 5 μM
5FAM-peptide in a 100 mM Hepes pH 7.4, 10 mM MgCl2, 0.003%
Brij35, 1 mM DTT assay buffer at pH 7.4. The substrate and product
were separated on the basis of charge using upstream, downstream
voltages of −2800 V, −380 V, respectively, and a screening pressure
of 0.8 psi. For S6K, kinetic studies were conducted in the presence
of 5 nM enzyme, 50 μM ATP, 5 μM 5FAM-peptide in a 20 mM
Tris·HCl, 15 mM MgCl2, 50 mM NaCl, 0.1 mM EDTA, 1 mM DTT
assay buffer at pH 7.4. The substrate and product were separated using
upstream and downstream voltages of −2500 V and −500 V,
respectively, and a screening pressure of −1.2 psi. For PDK1, kinetic
studies were conducted in the presence of 20 nM enzyme, 50 μM
ATP, and 3 μM Sox-peptide in a 50 mM Hepes, 5 mM MgCl2, 0.01%
Brij35, 1 mM DTT assay buffer at pH 7.4. The increase of
fluorescence (λex = 360 nm and λem = 485 nm) was recorded using
a Safire TECAN plate reader. For mTOR, an in vitro LanthaScreen
kinase assay (Invitrogen - Life Technologies) was conducted to
determine the level of phosphorylation of the protein substrate
4E-BP1 at Thr46 in a TR-FRET format. The phosphorylation of
GFP-4E-BP1 at Thr46 is recognized by the Terbium-anti-p4E-BP1
antibody, which results in time-resolved fluorescence resonance energy
transfer (TR-FRET) between GFP and terbium in the antibody−
product complex. Kinetic studies were conducted in the presence
of 1 nM mTOR, 0.4 μM GFP-4E-BP1, 2 mM antibody (added at
the end of the reaction), 40 μM ATP in a 50 mM Hepes, 10 mM
MgCl2, 1 mM EGTA, 0.01% Tween-20, assay buffer at pH 7.4.
For PI3Kα, the kinase reaction mixture (i.e. 0.5 nM mouse
PI3Kα diluted in assay buffer, 30 μM PIP2, compound, 5 mM
MgCl2, and 200 μM ATP) was incubated for 30 min prior to adding
EDTA (10 mM) to stop the reaction. The mixture was added to the
Detector/Probe mix containing 480 nM GST-Grp1 PH Domain
and 12 nM TAMRA tagged fluorescent PIP3 dissolved in assay
buffer for 3 min. Polarization (mP) values decreased as TAMRA
tagged PI(3,4,5)P3 probe binding to the GST-Grp1 PH domain
was displaced by PIP3 produced by the PI3Kα enzymatic
reaction. Polarization of fluorescence studies were conducted in a
50 mM Hepes pH 7.4, 150 mM NaCl, 5 mM DTT, and 0.05%
CHAPS.
1300
910
selective against these four kinases, analogues with the
acetamide group (25 and 26) are more selective than
the formamide analogue (24). A potency of ca. 100 nM for
24 against Aurora A, MARK1, and PI3K (Table 1) could
potentially contribute to moderate activity in an antiprolifera-
tion assay, whereas there is less concern with analogues 25
and 26.
The antiproliferative activity for 24−26 was measured in two
tumor cell lines, A549 and H460 (Table 5). In A549, even
though good levels of pAKT-T308 reduction are observed, the
levels of antiproliferative activity were modest at best with 20-
to 30-fold increase between IC50 and EC50. In the H460 cell
line, lower reductions in pAKT-T308 levels were seen with
similar levels of antiproliferative activity, resulting in a lower
fold increase between IC50 and EC50 (ca. 5-fold). The modest
growth inhibition in H460 for these selective PDK1 inhibitors
was disappointing considering that both a PI3K-selective
inhibitor27 and an mTOR-selective inhibitor28 had an EC50
ca. 100 nM in the same cell line.
In summary, we have discovered potent inhibitors of PDK1
with excellent selectivity within the pathway (PI3K, AKT, S6K,
and mTOR) and against a broader range of kinases. The PI3K
selectivity was achieved by filling under the G-loop, and the
solubility, permeability, and CYP liability were improved by
lowering the lipophilicity of the selective lead 18. While the
potent and selective PDK1 inhibitors effectively reduced
pAKT-T308 in tumor cells in vitro, they had only modest
effects in preventing tumor cell growth in the in vitro
antiproliferation assay.
EXPERIMENTAL SECTION
■
Reagents and General Enzymatic Assay Conditions. EDTA,
Tris, and Hepes buffer, DMSO, ATP, DTT, magnesium chloride, Brij-
35, and Triton X-100 were purchased from Sigma-Aldrich (St Louis,
MO). Fluorescent labeled AKT substrate (5FAM-GRPRTSSFAEG-
CONH2) and S6K substrate (5FAM-KRSRTRTDSYSAGQ-COOH,
respectively) were purchased from Caliper LifeSciences (Hopkinton,
MA) and CPC Scientific (San Jose, Ca), respectively. The PDK1
Omnia peptide (Ac-Sox-PKTFCGTPEYLAPEVRREPRILSEEE-
QEMFRDFDYIAD-NH2) was purchased from Invitrogen Life Tech-
The data from the broad kinase selectivity panel were provided by
Invitrogen as fee-for-service and were generated in the presence of
1 μM of inhibitor against a panel of 35 selected kinases.
Ki Determination. Kiapp and Ki were calculated by fitting the
experimental data to the following equations:31
Table 5. In Vitro Cellular Potency in Biomarker and Functional Assays
A549 (nM)
H460 (nM)
a
b
c
a
b
c
cmpd
PDK1 Ki (nM)
pAKT IC50
proliferation EC50
fold shift
pAKT IC50
proliferation EC50
fold shift
24
25
26
0.7
1.4
1.4
50
80
1800
2200
36
28
17
120
160
320
700
1900
1700
6
12
5
300
5200
a
b
ELISA pAKT-T308 inhibition assay. See Table 2 for statistics. Cell proliferation assay using resazurin, n = 2, individual values within 30% of
c
average. Proliferation EC50/pAKT IC50.
8495
dx.doi.org/10.1021/jm201019k|J. Med. Chem. 2011, 54, 8490−8500

Journal of Medicinal Chemistry
Article
(dd, J = 11.1, 7.8 Hz, 1 H), 4.29 (q, J = 6.3 Hz, 1 H), 7.00−7.07 (m, 2
H), 7.19−7.25 (m, 2 H). LC-MS (APCI+) 182 (M + H-Boc).
cis-3-(1,3-Dioxo-1,3-dihydroisoindol-2-yl)-4-(4-
fluorophenyl)pyrrolidine-1-carboxylic Acid tert-Butyl Ester
(5c). To a solution of alcohol 4c (2.1 g, 7.5 mmol), phthalimide
(1.3 g, 9 mmol), and triphenylphosphine (2.4 g, 9 mmol) at 0 °C was
added slowly diisopropyl azodicarboxylate (1.9 g, 9 mmol). The
reaction mixture was allowed to warm to 20 °C and was stirred for
16 h. The solution was concentrated in vacuo, dissolved in ethyl acetate,
washed with 2 M aqueous sodium hydroxide and brine, dried over
sodium sulfate, and concentrated in vacuo. The crude was purified by
flash chromatography to give the title compound 5c as a white solid
(1)
(2)
where
[E]o and [I]o are the total active enzyme and inhibitor concentrations,
respectively; Ki is the inhibition binding constant; Vi and Vo are the
rates of peptide phosphorylation in the presence or in the absence of
inhibitor, respectively; [A]o is the total ATP concentration.
1
(2.6 g, 85%). H NMR (400 MHz, CDCl3) δ ppm 1.54 (s, 9 H),
3.76−3.87 (m, 1 H), 3.87−4.09 (m, 3 H), 4.99 (spt, J = 6.2 Hz, 1 H),
5.11 (td, J = 7.6, 4.0 Hz, 1 H), 6.79−6.86 (m, 2 H), 7.11 (dd, J = 8.4,
5.4 Hz, 2 H), 7.63−7.68 (m, 2 H), 7.68−7.73 (m, 2 H). LC/MS
(APCI+): 311 (M + H-Boc).
pAKT-T308 ELISA Assay. Cells were plated at 50,000 cells/well
in a 96-well microtiter plate and allowed to adhere overnight. Inhibitor
compounds were added to each well (3-fold serial dilution, range
0.0169 nM to 10 μM) for 2 h prior to cell lysis. After compound
treatment, cells were lysed and lysate transferred to the ELISA plate.
The pAKT T308 ELISA was performed per the manufacturer’s
directions (Cell Signaling Technology). All assays were run in
duplicate and have been repeated at least twice.
Cell Proliferation Assay. Cells were plated at 3,000 cells/well in
a 96-well microtiter plate and allowed to adhere overnight. Inhibitor
was added to each well (3-fold serial dilution, range 0.0169 nM to
10 μM) and incubated. At 72 h postcompound addition, 250 μg/mL
of resazurin was added to all wells. Plates were incubated at 37 °C for
an additional 6 h followed by capture of the fluorescence signal using
emission 590 nm and excitation 560 nm. All assays were run in
duplicate and have been repeated at least twice.
cis-3-Amino-4-(4-fluorophenyl)pyrrolidine-1-carboxylic Acid
tert-Butyl Ester (6c). A solution of phthalimide 5c (2.6 g, 6.4 mmol)
and hydrazine monohydrate (3.25 g, 64 mmol) in ethanol (30 mL)
was stirred at 20 °C for 3 days. The reaction was concentrated from
ethanol 3-times to remove excess hydrazine. The residue was dissolved
in ethyl acetate and filtered to remove solids. The filtrate was absorbed
onto an SCX column and rinsed with methanol. The product was
eluted with 3.5 M ammonia in methanol. The eluent was concentrated
in vacuo to give the title compound 6c as a yellow oil (1.3 g, 71%). 1H
NMR (400 MHz, CDCl3) δ ppm 1.50 (s, 9 H), 3.18−3.32 (m, 1 H),
3.38 (td, J = 7.8, 5.3 Hz, 1 H), 3.49 (s, 2 H), 3.58−3.83 (m, 4 H),
7.02−7.09 (m, 2 H), 7.20 (d, J = 5.3 Hz, 2 H). LC-MS (APCI+) 181
(M + H-Boc).
cis-2-(4-Fluorophenyl)tetrahydrofuran-3-ylamine (7). Syn-
thesized following the procedure from Andreotti and Ward17 for the
analogous des-fluoro compound. 1H NMR (300 MHz, MeOH-d4)
δ ppm 2.62−2.81 (m, 1 H), 3.03−3.22 (m, 1 H), 4.41−4.56 (m,
1 H), 4.56−4.67 (m, 1 H), 4.73−4.90 (m, 1 H), 5.52 (d, J = 3.8 Hz,
1 H), 7.60−7.79 (m, 2 H), 7.94−8.12 (m, 2 H). LC-MS (APCI+) 182
(M + H).
General Chemistry Experimental. All reagents and solvents
were used as purchased from commercial sources. Flash chromatog-
raphy was run on ISCO or Biotage compatible silica gel cartridges and
eluted with ethyl acetate in heptane unless otherwise noted. All
reactions were performed under a positive pressure of nitrogen, at
ambient temperature, and in anhydrous solvents, unless otherwise
indicated. Microwave assisted reactions were run in a Smith Creator
1
(Personal Chemistry). H NMR spectra were recorded on a Bruker
Ethyl 2-(4-Fluorophenyl)-1-[(4-methylphenyl)sulfonyl]-2,5-
dihydro-1H-pyrrole-3-carboxylate (9b). A solution of 4-fluoro-
benzaldehyde (15.2 g, 122 mmol), toluene-4-sulfonamide (20.9 g, 122
mmol), and borontrifluoride-etherate (0.45 g, 3.2 mmol) in anhydrous
toluene (200 mL) was heated at reflux with a Sean−Stark trap for 4 h.
The solution was allowed to cool to 20 °C. To the solution was added
ethyl 2-butynoate (14.4 g, 128 mmol) and tri-n-butylphosphine (1.2 g,
6.1 mmol), and the solution was heated at 100 °C for 1 h. The
solution was concentrated in vacuo to give a reddish-brown oil which
was purified by flash chromatography to give the title compound 9b
(34.3 g, 72%). 1H NMR (400 MHz, CDCl3) δ ppm 1.12 (t, J = 7.1 Hz,
3 H), 2.39 (s, 3 H), 3.98−4.11 (m, 2 H), 4.33−4.45 (m, 1 H), 4.45−
4.57 (m, 1 H), 5.67−5.77 (m, 1 H), 6.75−6.83 (m, 1 H), 6.93 (t, J =
8.6 Hz, 2 H), 7.12−7.24 (m, 4 H), 7.45 (d, J = 8.3 Hz, 2 H). LC/MS
(APCI+): 390 (M + H).
instrument operating at 300 or 400 MHz as indicated. The mass
spectra were obtained using liquid chromatography mass spectrometry
(LC-MS) on an Agilent instrument using atmospheric pressure
chemical ionization (APCI) or electrospray ionization (ESI). All test
compounds showed >95% purity as determined by combustion
analysis. Combustion analyses were performed by Atlantic Microlab,
Inc. (Norcross, GA). Chiral SFC separations were run on a 21.2 mm ×
250 mm column at 100 bar, 35 °C, and 60−65 mL/min unless
otherwise noted.
General Procedure for Final Compounds (2). A mixture of
(4-amino-7-isopropyl-7H-pyrrolo[2,3-d]pyrimidin-5-yl)-(6-chloro-
pyrazin-2-yl)methanone (1) (100 mg, 0.32 mmol), primary or
secondary amine (0.47 mmol), cesium fluoride (146 mg, 0.96
mmol), and Hunig’s base (165 mg, 1.3 mmol) in anhydrous DMSO
(1.2 mL) was heated at 120 °C for 16 h. The crude was filtered to
remove salts and purified by reverse phase HPLC to provide title
compounds as yellow solids.
1-tert-Butyl 3-Ethyl 2-(4-Fluorophenyl)-1H-pyrrole-1,3-dicar-
boxylate (10b). To a solution of tosylate 9b (17.2 g, 44 mmol) in
anhydrous dimethylformamide (200 mL) at 0 °C was added dropwise
a solution of potassium tert-butoxide in tetrahydrofuran (1.0 M,
46 mL, 46 mmol) over 1.5 h. The solution was then allowed to warm
to 20 °C over 30 min and stirred at 20 °C for 1.5 h. To the solution
of pyrrole was added Boc anhydride (19.2 g, 88 mmol) and 4-
dimethylaminopyridine (110 mg, 0.9 mmol). The solution was stirred
at 20 °C for 2 h. The tetrahydrofuran was removed in vacuo. The
residue was diluted with saturated aqueous ammonium chloride
(500 mL) and extracted with diethyl ether (400 mL). The organics
were washed with brine (200 mL), dried with magnesium sulfate, and
concentrated in vacuo. The crude was purified by flash chromatog-
(trans)-3-(4-Fluorophenyl)-4-hydroxypyrrolidine-1-carboxylic
Acid tert-Butyl Ester (4c). A solution of 6-oxa-3-aza-bicyclo[3.1.0]-
hexane-3-carboxylic acid tert-butyl ester (3c, 2.0 g, 11 mmol) in
2-Me-THF (5 mL) was added dropwise to a solution of the 4-
fluorophenylmagnesium bromide (2.0 M in THF, 11 mL, 22 mmol)
and copper iodide (100 mg, 0.54 mmol) at 0 °C. The reaction was
allowed to warm to 20 °C and stirred for 16 h. The reaction was
cooled to 0 °C and quenched with sat. aq ammonium chloride. The
product was extracted with ethyl acetate. The organics were washed
with sodium bicarbonate and brine, dried over sodium sulfate, and
concentrated in vacuo. The crude was purified by flash chromatog-
1
raphy to give the title compound 10b (12.7 g, 86%). H NMR (400
1
raphy to give the title compound 4c as a yellow oil (2.1 g, 69%). H
MHz, CDCl3) δ ppm 1.12 (t, J = 7.1 Hz, 3 H), 1.30 (s, 9 H), 4.09 (q,
J = 7.1 Hz, 2 H), 6.68 (d, J = 3.3 Hz, 1 H), 7.09 (t, J = 8.7 Hz, 2 H),
7.24−7.31 (m, 2 H), 7.33 (d, J = 3.5 Hz, 1 H). LC/MS (APCI+): 334
(M + H).
NMR (400 MHz, CDCl3) δ ppm 1.49 (s, 9 H), 2.02 (br s, 1 H),
3.24 (q, J = 7.6 Hz, 1 H), 3.31 (dd, J = 11.5, 5.7 Hz, 1 H), 3.48
(dd, J = 11.3, 7.6 Hz, 1 H), 3.76 (dd, J = 11.3, 6.3 Hz, 1 H), 3.88
8496
dx.doi.org/10.1021/jm201019k|J. Med. Chem. 2011, 54, 8490−8500

Journal of Medicinal Chemistry
Article
1-tert-Butyl 3-Ethyl (2S,3R)-2-(4-Fluorophenyl)pyrrolidine-
1,3-dicarboxylate (11b). A mixture of pyrrole 10b (12.7 g, 38
mmol) and platinum(IV) oxide (7.5 g, 55 mmol) in ethanol (200 mL)
was shaken in a Parr shaker with 45 psi hydrogen at 20 °C for 24 h.
The catalyst was removed by filtering through Celite and rinsing with
ethanol. CAUTION: Care must be taken to not let the filter cake go
dry to avoid a fire. The filtrate was concentrated in vacuo to give the
(23 mL, 92 mmol), and the solution was stirred at 20 °C for 18 h. The
solvent was removed in vacuo, and the white solid was dried under
high vacuum at 50 °C to give the title compound 13b (4.3 g, 100%).
1H NMR (400 MHz, DMSO-d6, 1.3:1 mixture of rotamers A:B) δ ppm
1.66 (s, 1.3 H, rotamer B), 1.77−1.97 (m, 1 H), 2.00 (s, 1.7 H,
rotamer A), 2.13−2.30 (m, 1 H), 3.38−3.51 (m, 0.4 H, rotamer B),
3.58−3.68 (m, 0.6 H, rotamer A), 3.73−3.88 (m, 1 H), 3.88−4.02 (m,
1 H), 5.18 (d, J = 7.3 Hz, 0.6 H, rotamer A), 5.17 (d, J = 7.3 Hz, 0.4 H,
rotamer B), 7.15−7.31 (m, 2 H), 7.31−7.44 (m, 2 H), 8.10 (br s,
1.7 H, rotamer A), 8.19 (br s, 1.3 H, rotamer B). LC/MS (APCI+) 223
(M + H).
[4-Amino-7-(propan-2-yl)-7H-pyrrolo[2,3-d]pyrimidin-5-
yl][6-(methylamino)pyrazin-2-yl]methanone (14). 1H NMR
(400 MHz, CDCl3) δ ppm 1.56 (d, J = 6.8 Hz, 6 H), 3.12 (d, J =
5.1 Hz, 3 H), 4.81 (br s, 1 H), 5.11 (quin, J = 6.8 Hz, 1 H), 8.12 (s, 1
H), 8.34 (s, 1 H), 8.53 (s, 1 H), 8.75 (s, 1 H). LC/MS (ESI+): 312
(M + H). Anal. Calcd for C15H17N7O·2.85H2O: C, 49.92; H, 5.99; N,
26.71. Found: C, 49.67; H, 6.31; N, 27.03.
[4-Amino-7-(propan-2-yl)-7H-pyrrolo[2,3-d]pyrimidin-5-
yl](6-{[2-(pyridin-3-yl)ethyl]amino}pyrazin-2-yl)methanone
(15). 1H NMR (400 MHz, CDCl3) δ ppm 1.47 (d, J = 6.6 Hz, 6 H),
3.03 (t, J = 7.1 Hz, 2 H), 3.72−3.84 (m, 2 H), 4.85 (t, J = 5.7 Hz, 1 H),
5.07 (quin, J = 6.8 Hz, 1 H), 7.22−7.27 (m, 1 H), 7.51−7.58 (m, 1 H),
8.09 (s, 1 H), 8.34 (s, 1 H), 8.48−8.57 (m, 4 H). LC/MS (APCI+):
403 (M + H). Anal. Calcd for C21H22N8O·1.4H2O: C, 59.10; H, 5.73;
N, 26.07. Found: C, 58.98; H, 5.84; N, 26.20.
1
title compound 11b (12.1 g, 94%). H NMR (400 MHz, CDCl3) δ
ppm 1.04 (t, J = 7.2 Hz, 3 H), 1.19 (s, 6 H), 1.37−1.49 (m, 3 H),
2.02−2.13 (m, 1 H), 2.34−2.49 (m, 1 H), 3.31−3.45 (m, 1 H), 3.45−
3.58 (m, 1 H), 3.76−3.82 (m, 1 H), 3.87 (m, 2 H), 5.01−5.24 (m, 1
H), 6.92−7.00 (m, 2 H), 7.06−7.17 (m, 2 H). LC/MS (APCI+): 238
(M + H-Boc).
cis-1-Acetyl-2-(4-fluorophenyl)pyrrolidine-3-carboxylic Acid
(12b). A solution of ester 11b (12.1 g, 36 mmol) in dioxane (100 mL)
and aqueous 6 N hydrochloric acid (100 mL) was heated at
70 °C for 40 h. The solution was cooled to 0 °C, and solid sodium
bicarbonate was added slowly and carefully until the final pH was 8.
To the solution was added acetic anhydride (11 g, 107 mmol), and the
solution was stirred at 20 °C for 3.5 h. The solution was cooled to
0 °C, and aqueous 6 N hydrochloric acid was added dropwise to lower
the pH to 2. The aqueous solution was extracted with methyl tert-butyl
ether. The organics were washed with brine, dried with magnesium
sulfate, and concentrated in vacuo. The oil was reconcentrated from
heptane and ethyl acetate to give a light yellow solid. The solid was
triturated with heptane (ca. 250 mL) using vigorous stirring, collected
by vacuum filtration, and rinsed with heptane to give a white solid
which was dried overnight at 40 °C under high vacuum to give the title
[4-Amino-7-(propan-2-yl)-7H-pyrrolo[2,3-d]pyrimidin-5-
yl](6-{[(1R,2R)-2-phenylcyclopentyl]amino}pyrazin-2-yl)-
methanone (17). Enantiomers were separated on a Chiracel OD-H
column with 30% methanol, 70% CO2; the more active enantiomer
1
compound 12b (8.4 g, 93%). H NMR (400 MHz, DMSO-d6, 1:1
mixture of rotamers A:B) δ ppm 1.67 (s, 1.5 H, rotamer A), 1.98 (s,
1.5 H, rotamer B), 1.91−2.07 (m, 1 H), 2.07−2.29 (m, 1 H), 3.34−
3.62 (m, 2 H), 3.70−3.82 (m, 0.5 H, rotamer A), 3.91 (t, J = 9.1 Hz,
0.5 H, rotamer B), 5.18 (d, J = 8.3 Hz, 0.5 H, rotamer A), 5.26 (d, J =
8.6 Hz, 0.5 H, rotamer B), 7.00−7.11 (m, 1 H, rotamer A), 7.11−7.27
(m, 2 H plus 1 H, rotamer B), 12.33 (br s, 1 H). LC/MS (APCI+):
252 (M + H).
1
eluted first and is the (+) isomer. H NMR (400 MHz, DMSO-d6) δ
ppm 1.51 (d, J = 6.8 Hz, 3 H), 1.55 (d, J = 6.8 Hz, 3 H), 1.66−1.89
(m, 2 H), 1.90−2.04 (m, 1 H), 2.03−2.19 (m, 3 H), 3.37−3.47 (m, 1
H), 4.55−4.67 (m, 1 H), 5.01 (quin, J = 6.8 Hz, 1 H), 7.08 (dq, J =
8.6, 4.2 Hz, 1 H), 7.12−7.20 (m, 5 H), 7.46 (br s, 1 H), 7.97 (s, 1 H),
8.10 (s, 1 H), 8.22 (s, 1 H), 8.36 (br s, 1 H), 8.80 (s, 1 H). LC/MS
(ESI+): 442 (M + H). Anal. Calcd for C25H27N7O·0.50H2O: C, 66.82;
H, 6.20; N, 21.62. Found: C, 66.65; H, 6.26; N, 21.76.
cis-2-(4-Fluorophenyl)-1-formylpyrrolidine-3-carboxylic
Acid (12c). A solution of ester 11b (9.15 g, 27 mmol) in dioxane
(60 mL) and aqueous 6 N hydrochloric acid (53 mL) was heated at
70 °C for 2 days. The solution was cooled to 0 °C, and solid sodium
bicarbonate was added slowly and carefully until the final pH was 8.
To the solution was added ethyl formate (54 mL, 675 mmol), and the
reaction mixture was heated at 80 °C for 40 h. The reaction was
cooled to about 0 °C and acidified to pH 1 using 6 N aqueous
hydrochloric acid (ca. 30 mL). The product was extracted with ethyl
ether (600, 300, and 200 mL). The combined organics were dried over
sodium sulfate and concentrated in vacuo to give the title compound
(4-Amino-7-isopropyl-7H-pyrrolo[2,3-d]pyrimidin-5-yl)-{6-
[(1R,2R)-2-(4-fluorophenyl)cyclopentylamino]pyrazin-2-yl}-
methanone (18). Single enantiomers were obtained by separating
the N-Boc derivative of 6a using a Chiralpak AD-H column eluting
with 9% methanol and 91% CO2; the desired enantiomer eluted first
22
and had optical rotation [α]D +47 (c 0.25, MeOH). The Boc group
1
was removed with HCl before coupling to 1. H NMR (400 MHz,
DMSO-d6) δ ppm 1.48 (d, J = 6.8 Hz, 3 H), 1.53 (d, J = 6.6 Hz, 3 H),
1.62−1.86 (m, 2 H), 1.86−2.18 (m, 4 H), 3.35−3.45 (m, 1 H), 4.52−
4.63 (m, 1 H), 4.90−5.06 (m, 1 H), 6.89−7.00 (m, 2 H), 7.09−7.20
(m, 3 H), 7.45 (br s, 1 H), 7.96 (s, 1 H), 8.10 (s, 1 H), 8.20 (s, 1 H),
8.32 (br s, 1 H), 8.75 (s, 1 H). LC/MS (APCI+): 460 (M + H). Anal.
Calcd for C25H26N7OF·0.75H2O: C, 63.63; H, 5.72; N, 20.48. Found:
C, 63.48; H, 5.86; N, 20.73.
1
12c (5.87 g, 92%). H NMR (400 MHz, DMSO-d6, 1.6:1 mixture of
rotamers A:B) δ ppm 1.94−2.09 (m, 1 H), 2.09−2.25 (m, 1 H), 3.36−
3.48 (m, 1 H plus 0.6 H, rotamer A), 3.65 (td, J = 9.8, 7.3 Hz, 0.4 H,
rotamer B), 3.70−3.78 (m, 0.6 H, rotamer A), 3.98 (td, J = 9.2, 2.8 Hz,
0.4 H, rotamer B), 5.16 (d, J = 8.3 Hz, 0.4 H, rotamer B), 5.34 (d, J =
8.3 Hz, 0.6 H, rotamer A), 7.04−7.26 (m, 4 H), 8.07 (s, 0.6 H, rotamer
A), 8.21 (s, 0.4 H, rotamer B), 12.39 (br s, 1 H). LCMS (APCI+): 238
(M + H).
cis-1-[3-Amino-2-(4-fluorophenyl)pyrrolidin-1-yl]ethanone
Hydrochloride Salt (13b). A solution of acid 12b (8.2 g, 33 mmol),
triethylamine (5.0 g, 49 mmol), and diphenylphosphoryl azide (12.5 g,
45 mmol) in toluene (1100 mL) was heated at 70 °C for 4 h. The
solution was cooled to 20 °C, 1.0 M potassium tert-butoxide solution
in tetrahydrofuran (68 mL, 68 mmol) was added, and the resulting
mixture was stirred at 20 °C for 20 min. Boc anhydride (12 g, 55
mmol) and 4-dimethylaminopyridine (200 mg, 1.6 mmol) were added,
and the mixture was stirred at 20 °C for 16 h. The solution was washed
with saturated aqueous ammonium chloride solution (2 × 600 mL)
and brine (500 mL), dried with magnesium sulfate, and concentrated
in vacuo. The crude was purified by flash chromatography to give cis-
tert-butyl [1-acetyl-2-(4-fluorophenyl)pyrrolidin-3-yl]carbamate (5.3 g,
50%, 16 mmol), which was dissolved in a solution of anhydrous
methylene chloride (160 mL) and 4 M hydrochloric acid in dioxane
(4-Amino-7-isopropyl-7H-pyrrolo[2,3-d]pyrimidin-5-yl)(6-
{[(1R,2R)-2-(4-fluorophenyl)cyclopentyl]methylamino}pyrazin-
2-yl)methanone (19). The single enantiomer was prepared from the
single enantiomer of N-Boc protected 6a (see 18 above) as shown in
1
Scheme 2. H NMR (400 MHz, DMSO-d6) δ ppm 1.39−1.48 (m, 6
H), 1.61−1.75 (m, 1 H), 1.98−2.17 (m, 5 H), 3.32 (s, 3 H), 3.50−3.59
(m, 1 H), 4.91−5.07 (m, 2 H), 6.98 (t, J = 8.8 Hz, 2 H), 7.10 (dd, J =
8.7, 5.7 Hz, 2 H), 7.49 (br s, 1 H), 8.20 (s, 1 H), 8.26 (s, 1 H), 8.27−
8.31 (m, 1 H), 8.31 (s, 1 H), 8.61 (s, 1 H). LC/MS (APCI+): 474
(M + H). Anal. Calcd for C26H28N7OF·0.50H2O: C, 64.69; H, 5.94;
N, 20.25. Found: C, 64.71; H, 6.06; N, 20.32.
(4-Amino-7-isopropyl-7H-pyrrolo[2,3-d]pyrimidin-5-yl){6-
[(3S,4R)-4-(4-fluorophenyl)tetrahydrofuran-3-ylamino]-
pyrazin-2-yl}methanone (20). Single enantiomers were obtained
by separating the N-Boc derivative of 6b using a Chiralpak AD-H
column eluting with 9% methanol and 91% CO2; the desired
22
enantiomer eluted first and had optical rotation [α]D +61 (c 2.28,
MeOH). The Boc group was removed with HCl before coupling to 1.
1H NMR (400 MHz, DMSO-d6) δ ppm 1.49 (d, J = 6.6 Hz, 3 H), 1.55
8497
dx.doi.org/10.1021/jm201019k|J. Med. Chem. 2011, 54, 8490−8500

Journal of Medicinal Chemistry
Article
(d, J = 6.6 Hz, 3 H), 3.73 (q, J = 6.9 Hz, 1 H), 3.83 (dd, J = 8.7, 5.2 Hz,
1 H), 4.05 (dd, J = 8.6, 7.1 Hz, 1 H), 4.14−4.22 (m, 2 H), 4.85−4.94 (m,
1 H), 4.99 (quin, J = 6.7 Hz, 1 H), 6.96−7.04 (m, 2 H), 7.19 (dd, J = 8.7,
5.7 Hz, 1 H), 7.42 (d, J = 7.8 Hz, 1 H), 7.48 (br s, 1 H), 8.01 (s, 1 H),
8.14 (s, 1 H), 8.21 (s, 1 H), 8.29 (br s, 1 H), 8.69 (s, 1 H). LC/MS
(APCI+): 462 (M + H). Anal. Calcd for C24H24N7FO2·0.60H2O: C,
60.97; H, 5.30; N, 20.61. Found: C, 61.03; H, 5.38; N, 20.76.
chiral SFC using a Chiracel OD-H column eluted with 40% methanol
1
and 60% CO2. [α]22 +595 (c 0.21, MeOH). H NMR (400 MHz,
D
DMSO-d6, 1.4:1 mixture of rotamers A:B) δ ppm 1.37−1.50 (m, 6 H),
1.59 (s, 1.3 H, rotamer B), 2.00 (s, 1.7 H, rotamer A), 1.93−2.09 (m, 1
H), 2.16−2.30 (m, 1 H), 3.37−3.45 (m, 0.4 H, rotamer B), 3.52−3.64
(m, 0.6 H, rotamer A), 3.83 (t, J = 10.5 Hz, 0.4 H, rotamer B), 3.98 (t,
J = 9.5 Hz, 0.6 H, rotamer A), 4.48−4.61 (m, 0.6 H, rotamer A), 4.70−
4.84 (m, 0.4 H, rotamer B), 4.88−5.04 (m, 1 H), 5.21 (d, J = 7.3 Hz,
0.4 H, rotamer B), 5.26 (d, J = 7.3 Hz, 0.6 H, rotamer A), 6.87−6.96
(m, 1.2 H, rotamer A), 6.96−7.09 (m, 2 H), 7.14 (t, J = 8.7 Hz, 0.8 H,
rotamer B), 7.26 (d, J = 6.8 Hz, 0.4 H, rotamer B), 7.35 (d, J = 6.6 Hz,
0.6 H, rotamer A), 7.48 (br s, 1 H), 8.02 (s, 1 H), 8.20 (s, 1 H), 8.33
(s, 1 H), 8.34 (br s, 1 H), 8.92 (s, 1 H). LC/MS (APCI+): 503 (M +
H). Anal. Calcd for C26H27N8FO2·0.70H2O: C, 60.71; H, 5.48; N,
21.61. Found: C, 60.62; H, 5.56; N, 21.75.
1-{(2R,3R)-3-[6-(4-Amino-7-isopropyl-7H-pyrrolo[2,3-d]-
pyrimidine-5-carbonyl)pyrazin-2-ylamino]-2-phenylpyrroli-
din-1-yl}ethanone (26). The single enantiomer was obtained by
separating intermediate 13a using a Chiralpak AS-H column eluted
with 20% methanol and 80% CO2; the desired enantiomer of 13a was
the (+) isomer. 1H NMR (400 MHz, DMSO-d6, 1.3:1 mixture of
rotamers A:B) δ ppm 1.36−1.50 (m, 6 H), 1.56 (s, 1.3 H, rotamer B),
2.00 (s, 1.7 H, rotamer A), 2.02−2.14 (m, 1 H), 2.16−2.31 (m, 1 H),
3.37−3.47 (m, 0.4 H, rotamer B), 3.60 (td, J = 10.7, 6.3 Hz, 0.6 H,
rotamer A), 3.77−3.88 (m, 0.4 H, rotamer B), 3.98 (t, J = 9.1 Hz, 0.6
H, rotamer A), 4.47−4.62 (m, 0.6 H, rotamer A), 4.71−4.83 (m, 0.4
H, rotamer B), 4.91−5.01 (m, 1 H), 5.20 (d, J = 7.3 Hz, 0.4 H,
rotamer B), 5.29 (d, J = 7.3 Hz, 0.6 H, rotamer A), 6.89 (d, J = 7.1 Hz,
1.1 H, rotamer A), 6.94 (d, J = 7.1 Hz, 0.9 H, rotamer B), 7.13−7.25
(m, 1.7 H, rotamer A), 7.25−7.34 (m, 1.3 H, rotamer B plus 0.4 H,
rotamer B), 7.37 (d, J = 7.1 Hz, 0.6 H, rotamer A), 7.48 (br s, 1 H),
8.00 (s, 1 H), 8.20 (s, 1 H), 8.32 (s, 0.4 H, rotamer B), 8.33 (s, 0.6 H,
rotamer A), 8.36 (br s, 1 H), 8.94 (s, 1 H). LC/MS (ESI+): 485
(M + H). Anal. Calcd for C26H28N8O2·2.0H2O: C, 60.21; H, 5.98; N,
21.29. Found: C, 59.99; H, 6.20; N, 21.52.
(4-Amino-7-isopropyl-7H-pyrrolo[2,3-d]pyrimidin-5-yl){6-
[(2R,3R)-2-(4-fluorophenyl)tetrahydrofuran-3-ylamino]-
pyrazin-2-yl}methanone (21). Single enantiomers were obtained
by separating the N-Boc derivative of 7 using a Chiralpak AD-H
column eluting with 10% methanol and 90% CO2 at 140 bar; the
25
desired enantiomer eluted first and had optical rotation [α]D +30
(c 0.83, MeOH). The Boc group was removed with HCl before coupl-
ing to 1. 1H NMR (300 MHz, CDCl3) δ ppm 1.56 (t, 6 H), 2.11−2.27
(m, 1 H), 2.48−2.61 (m, 1 H), 4.08 (td, J = 8.5, 6.8 Hz, 1 H), 4.30 (td,
J = 8.2, 5.9 Hz, 1 H), 4.48 (d, J = 7.4 Hz, 1 H), 4.73−4.87 (m, 1 H),
5.06−5.22 (m, 2 H), 7.03 (t, J = 8.7 Hz, 2 H), 7.29−7.34 (m, 2 H), 7.86
(s, 1 H), 8.35 (s, 1 H), 8.40 (s, 1 H), 8.43 (s, 1 H). LC/MS (APCI+):
462 (M + H). Anal. Calcd for C24H24N7FO2·0.75H2O·0.17DMSO: C,
60.13; H, 5.33; N, 19.81. Found: C, 59.87; H, 5.47; N, 20.08.
(4-Amino-7-isopropyl-7H-pyrrolo[2,3-d]pyrimidin-5-yl){6-
[(3S,4S)-4-(4-fluorophenyl)pyrrolidin-3-ylamino]pyrazin-2-yl}-
methanone Formate Salt (22). The enantiomers were separated by
chiral SFC using a Chiracel OJ-H column eluted with 30% methanol
and 70% CO2, with the desired enantiomer eluting first. The single
enantiomer was repurified by preparative HPLC (acetonitrile/water
1
gradient with formic acid). H NMR (400 MHz, DMSO-d6) δ ppm
1.49 (d, J = 6.8 Hz, 3 H), 1.54 (d, J = 6.6 Hz, 3 H), 3.11 (dd, J = 11.2,
3.9 Hz, 1 H), 3.23−3.35 (m, 1 H), 3.38−3.53 (m, 2 H), 3.55−3.68 (m,
1 H), 4.72−4.84 (m, 1 H), 4.98 (quin, J = 6.8 Hz, 1 H), 6.97 (t, J = 8.8
Hz, 2 H), 7.23 (dd, J = 8.3, 5.8 Hz, 2 H), 7.43 (br s, 1 H), 7.64 (d, J =
8.3 Hz, 1 H), 7.97 (s, 1 H), 8.10 (s, 1 H), 8.20 (s, 1 H), 8.29 (s, 2 H),
8.64 (s, 1 H). LC/MS (ESI+): 461 (M + H). Anal. Calcd for
C24H25FN8O·1.0HCO2H·1.8H2O: C, 55.8; H, 5.67; N, 20.58. Found:
C, 55.71; H, 5.72; N, 20.79.
(4-Amino-7-isopropyl-7H-pyrrolo[2,3-d]pyrimidin-5-yl){6-
[(2R,3R)-2-(4-fluorophenyl)pyrrolidin-3-ylamino]pyrazin-2-yl}-
methanone Formate Salt (23). The single enantiomer was
synthesized from the single enantiomer of 24 (see below) as described
in the text. 1H NMR (400 MHz, DMSO-d6) δ ppm 1.49 (d, J = 6.8 Hz,
3 H), 1.55 (d, J = 6.8 Hz, 3 H), 1.81−1.95 (m, 1 H), 2.19−2.31 (m, 1
H), 2.92−3.04 (m, 1 H), 3.18−3.29 (m, 1 H), 4.35 (d, J = 6.3 Hz, 1
H), 4.60−4.74 (m, 1 H), 4.99 (quin, J = 6.8 Hz, 1 H), 6.88−6.99 (m, 2
H), 7.20−7.31 (m, 3 H), 7.45 (br s, 1 H), 7.94 (s, 1 H), 8.08 (s, 1 H),
8.20 (s, 1 H), 8.30 (br s, 1 H), 8.72 (s, 1 H). LC/MS (ESI+): 461 -
(M + H). Anal. Calcd for C24H25N8FO·1.2HCO2H·1.4H2O: C, 55.96;
H, 5.54; N, 20.70. Found: C, 55.95; H, 5.63; N, 20.71.
(2R,3R)-3-[6-(4-Amino-7-isopropyl-7H-pyrrolo[2,3-d]-
pyrimidine-5-carbonyl)pyrazin-2-ylamino]-2-(4-fluorophenyl)-
pyrrolidine-1-carbaldehyde (24). The enantiomers were separated
by chiral SFC using a Chiracel OD-H column eluted with 40%
methanol and 60% CO2. [α]22D +454 (c 0.37, MeOH). 1H NMR (400
MHz, DMSO-d6, 1.7:1 mixture of rotamers A:B) δ ppm 1.29−1.43 (m,
6 H), 1.81−2.03 (m, 1 H), 2.09−2.23 (m, 1 H), 3.29−3.40 (m, 0.6 H,
rotamer A), 3.57 (td, J = 10.2, 7.1 Hz, 0.4 H, rotamer B), 3.72 (t, J =
10.5 Hz, 0.6 H, rotamer A), 3.97 (t, J = 9.3 Hz, 0.4 H, rotamer B),
4.46−4.56 (m, 0.4 H, rotamer B), 4.56−4.67 (m, 0.6 H, rotamer A),
4.87 (sxt, J = 6.8 Hz, 1 H), 5.13 (d, J = 7.3 Hz, 0.4 H, rotamer B), 5.26
(d, J = 7.3 Hz, 0.6 H, rotamer A), 6.83−6.92 (m, 2 H), 6.92−6.99 (m,
0.7 H, rotamer B), 6.99−7.06 (m, 1.3 H, rotamer A), 7.27 (d, J = 6.8
Hz, 0.4 H, rotamer B), 7.32 (d, J = 6.8 Hz, 0.6 H, rotamer A), 7.39 (br
s, 1 H), 7.92 (s, 0.4 H, rotamer B), 7.94 (s, 0.6 H, rotamer A), 8.02 (s,
0.6 H, rotamer A), 8.12 (s, 1 H), 8.16 (s, 0.4 H, rotamer B), 8.22 (s,
0.4 H, rotamer B), 8.23 (s, 0.6 H, rotamer A), 8.26 (br s, 1 H), 8.77 (s,
0.4 H, rotamer B), 8.82 (s, 0.6 H, rotamer A). LC/MS (APCI+): 489
(M + H). Anal. Calcd for C25H25N8FO2·1.80H2O: C, 57.58; H, 5.4; N,
21.59. Found: C, 57.64; H, 5.53; N, 21.51.
ASSOCIATED CONTENT
* Supporting Information
Further experimental details; full kinase profiling from
Invitrogen for analogues 24, 25, and 26; and crystallographic
methods and data for 15 and 20 in PDK1. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
Accession Codes
PDB nos.: cmpd 15 in PDK1, 3RWQ; cmpd 20 in PDK1,
3RWP.
AUTHOR INFORMATION
Corresponding Author
*Present address: Takeda San Diego, 10410 Science Center
Drive, San Diego, CA 92121. E-mail: sean.murphy@takedasd.
com. Phone: 858-731-3596.
■
ACKNOWLEDGMENTS
■
We thank Deepak Dalvie for biotransformation analysis;
Muhammad Alimuddin, Christine Aurigemma, Jason Ewanicki,
and Jeff Ellerass for analytical support; and Paul Richardson and
Kevin Bunker for synthetic support.
ABBREVIATIONS USED
■
ADME, absorption, distribution, metabolism, and excretion;
APCI, atmospheric pressure chemical ionization; ATP,
adenosine triphosphate; CHAPS, 3-[(3-cholamidopropyl)-
dimethylammonio]-1-propanesulfonate; CYP, cytochrome
P450; 4-DMAP, N,N-dimethyl-4-aminopyridine; DTT, dithio-
threitol; EDTA, ethylenediaminetetraacetic acid; EGTA, ethyl-
ene glycol tetraacetic acid; ESI, electrospray ionization; GST,
1-[(2R,3R)-3-[6-(4-Amino-7-isopropyl-7H-pyrrolo[2,3-d]-
pyrimidine-5-carbonyl)pyrazin-2-ylamino]-2-(4-fluorophenyl)-
pyrrolidin-1-yl]ethanone (25). The enantiomers were separated by
8498
dx.doi.org/10.1021/jm201019k|J. Med. Chem. 2011, 54, 8490−8500

Journal of Medicinal Chemistry
Article
́
́
(13) Stauffer, F.; Maira, S. M.; Furet, P.; Garcia-Echeverria, C.
Imidazo[4,5-c]quinolines as inhibitors of the PI3K/PKB-pathway.
Bioorg. Med. Chem. Lett. 2008, 18, 1027−1030.
glutathione S-transferase; Hepes, 4-(2-hydroxyethyl)-1-piper-
azineethanesulfonic acid; HLM, human liver microsome; LE,
ligand efficiency; PDK1, 3-phosphoinositide-dependent kinase;
PH, pleckstrin homology; PI3K, phosphoinositide-3 kinase;
PIP3, phosphaditylinositol-(3,4,5)-triphosphate; PTEN, phos-
phatase and tensin homologue; SFC, supercritical fluid
chromatography; RTK, receptor tyrosine kinase; TAMRA,
tetramethylrhodamine; T308, threonine-308; TR-FRET, time-
resolved fluorescence resonance energy transfer; Tris, tris-
(hydroxymethyl) aminomethane; Ts, 4-toluenesulfonyl
(14) Karaman, M. W.; Herrgard, S.; Treiber, D. K.; Gallant, P.;
Atteridge, C. E.; Campbell, B. T.; Chan, K. W.; Ciceri, P.; Davis, M. I.;
Edeen, P. T.; Faraoni, R.; Floyd, M.; Hunt, J. P.; Lockhart, D. J.;
Milanov, Z. V.; Morrison, M. J.; Pallares, G.; Patel, H. K.; Pritchard, S.;
Wodicka, L. M.; Zarrinkar, P. P. A quantitative analysis of kinase
inhibitor selectivity. Nat. Biotechnol. 2008, 26, 127−132.
(15) Lind, K. E.; Lin, E. Y.; Nguyen, T. B.; Tangonan, B. T.; Cao, K.;
Erlanson, D. A.; Guckian, K.; Simmons, R. L.; Lee, W.; Sun, L.;
Hansen, S.; Pathan, N.; Zhang, L. Pyridinonyl PDK1 inhibitors. PCT
Int. Appl. 2008, WO2008/5457 A2.
(16) Shepherd, T. A.; Aikins, J. A.; Bleakman, D.; Cantrell, B. E.;
Rearick, J. P.; Simon, R. L.; Smith, E. C. R.; Stephenson, G. A.;
Zimmerman, D. M.; Mandelzys, A.; Jarvie, K. R.; Ho, K.; Deverill, M.;
Kamboj, R. K. Design and synthesis of a novel series of 1,2-
disubstituted cyclopentanes as small, potent potentiators of 2-amino-3-
(3-hydroxy-5-methyl-isoxazol-4-yl)propanoic acid (AMPA) receptors.
J. Med. Chem. 2002, 45, 2101−2111.
(17) Andreotti, D.; Ward, S. E. Preparation of N-[2-phenyl-
tetrahydrofuran-3-yl]-2-propanesulfonamide derivatives as glutamate
receptor modulators. PCT Int. Appl. 2007, WO2007090841.
(18) Xu, Z.; Lu, X. Phosphine-catalyzed [3 + 2] cycloaddition
reactions of substituted 2-alkynoates or 2,3-allenoates with electron-
deficient olefins and imines. Tetrahedron Lett. 1999, 40 (3), 549−552.
(19) Hopkins, A. L.; Groom, C. R.; Alex, A. Ligand efficiency:
A useful metric for lead selection. Drug Discovery Today 2004, 9,
430−431.
REFERENCES
■
(1) Kobayashi, T.; Cohen, P. Activation of serum- and glucocorticoid-
regulated protein kinase by agonists that activate phosphatidylinositide
3-kinase is mediated by 3-phosphoinositide-dependent protein kinase-1
(PDK1) and PDK2. Biochem. J. 1999, 339, 319−328.
(2) Mora, A.; Komander, D.; van Aalten, D. M.; Alessi, D. R. PDK1,
the master regulator of AGC kinase signal transduction. Semin. Cell
Dev. Biol. 2004, Apr 15, 161−170.
(3) Raimondi, C.; Falasca, M. Targeting PDK1 in Cancer. Curr. Med.
Chem. 2011, 18, 2763−2769.
(4) Bayascas, J. R.; Leslie, N. R.; Parsons, R.; Fleming, S.; Alessi,
D. R. Hypomorphic mutation of PDK1 suppresses tumorigenesis in
PTEN(±) mice. Curr. Biol. 2005, 15, 1839−1846.
(5) Zeng, X.; Xu, H.; Glazer, R. I. Transformation of mammary
epithelial cells by 3-phosphoinositide-dependent protein kinase-1
(PDK1) is associated with the induction of protein kinase Calpha.
Cancer Res. 2002, 62, 3538−3543.
(6) Li, Y.; Yang, K. J.; Park, J. Multiple implications of 3-
phosphoinositide-dependent protein kinase 1 in human cancer. World
J. Biol. Chem. 2010, 1, 239−247.
(7) Peifer, C.; Alessi, D. R. Small-molecule inhibitors of PDK1.
ChemMedChem 2008, 3, 1810−1838.
(8) Angiolini, M.; Banfi, P.; Casale, E.; Casuscelli, F.; Fiorelli, C.;
Saccardo, M. B.; Silvagni, M.; Zuccotto, F. Structure-based
optimization of potent PDK1 inhibitors. Bioorg. Med. Chem. Lett.
2010, 20, 4095−4099.
(20) Michel, J.; Tirado-Rives, J.; Jorgensen, W. L. Energetics of
displacing water molecules from protein binding sites: Consequences
for ligand optimization. J. Am. Chem. Soc. 2009, 131, 15403−15411.
(21) Zlokarnik, G.; Grootenhuis, P. D. J.; Watson, J. B. High
throughput P450 inhibition screens in early drug discovery. Drug
Discovery Today 2005, 10, 1443−1450.
(22) PI3K-γ crystal structures were used as a surrogate for
developing binding hypotheses for PI3K-α, which was used in the
enzymatic assay. The crystal structure of 16 in PI3K-γ is available from
the PDB, entry 2V4L.
(9) Najafov, A.; Sommer, E. M.; Axten, J. M.; Deyoung, M. P.; Alessi,
D. R. Characterization of GSK2334470, a novel and highly specific
inhibitor of PDK1. Biochem. J. 2010, 433, 357−369.
(10) Nagashima, K.; Shumway, S. D.; Sathyanarayanan, S.; Chen, A.
H.; Dolinski, B.; Xu, Y.; Keilhack, H.; Nguyen, T.; Wiznerowicz, M.;
Li, L.; Lutterbach, B. A.; Chi, A.; Paweletz, C.; Allison, T.; Yan, Y.;
Munshi, S. K.; Klippel, A.; Kraus, M.; Bobkova, E. V.; Deshmukh, S.;
Xu, Z.; Mueller, U.; Szewczak, A. A.; Pan, B. S.; Richon, V.; Pollock,
R.; Blume-Jensen, P.; Northrup, A.; Andersen, J. N. Genetic and
pharmacological inhibition of PDK1 in cancer cells: Characterization
of a selective allosteric kinase inhibitor. J. Biol. Chem. 2011, 286,
6433−6448.
(11) (a) Islam, I.; Bryant, J.; Chou, Y. L.; Kochanny, M. J.; Lee, W.;
Phillips, G. B.; Yu, H.; Adler, M.; Whitlow, M.; Ho, E.; Lentz, D.;
Polokoff, M. A.; Subramanyam, B.; Wu, J. M.; Zhu, D.; Feldman, R. I.;
Arnaiz, D. O. Indolinone based phosphoinositide-dependent kinase-1
(PDK1) inhibitors. Part 1: Design, synthesis and biological activity.
Bioorg. Med. Chem. Lett. 2007, 17, 3814−3818. (b) Islam, I.; Brown,
G.; Bryant, J.; Hrvatin, P.; Kochanny, M. J.; Phillips, G. B.; Yuan, S.;
Adler, M.; Whitlow, M.; Lentz, D.; Polokoff, M. A.; Wu, J.; Shen, J.;
Walters, J.; Ho, E.; Subramanyam, B.; Zhu, D.; Feldman, R. I.; Arnaiz,
D. O. Indolinone based phosphoinositide-dependent kinase-1 (PDK1)
inhibitors. Part 2: Optimization of BX-517. Bioorg. Med. Chem. Lett.
2007, 17, 3819−3825.
(23) Andrews, P. R.; Craik, D. J.; Martin, J. L. Functional group
contributions to drug−receptor interactions. J. Med. Chem. 1984, 27,
1648−1657.
(24) Gleeson, M. P. Generation of a set of simple, interpretable
ADMET rules of thumb. J. Med. Chem. 2008, 51, 817−834.
(25) Analiza Inc., 3615 Superior Avenue, Suite 4407B, Cleveland,
OH 44114. Solubility assay: http://www.analiza.com/adme/aqueous-
solubility.html.
(26) The LE from 14 to 24 decreased from 0.42 to 0.38. A decrease
in LE is often observed during lead optimization. Researchers at
Johnson and Johnson ( Reynolds, C. H.; Tounge, B. A.; Bembenek,
S. D. Ligand binding efficiency: Trends, physical basis, and
implications. J. Med. Chem. 2008, 51, 2432−2438, ) have suggested
causes for this effect and introduced the Fit Quality parameter to
account for it. The Fit Quality from 14 to 24 increases from 0.73 to
0.89 and indicates an efficient use of heavy atoms in the optimization
process.
(12) Gopalsamy, A.; Shi, M.; Boschelli, D. H.; Williamson, R.;
Olland, A.; Hu, Y.; Krishnamurthy, G.; Han, X.; Arndt, K.; Guo, B.
Discovery of dibenzo[c,f][2,7]naphthyridines as potent and selective
3-phosphoinositide-dependent kinase-1 inhibitors. J. Med. Chem. 2007,
50, 5547−5549.
(27) Liu, K.; Zhu, J.; Smith, G. L.; Yin, M.; Bailey, S.; Chen, J. H.;
Hu, Q.; Huang, Q.; Li, C.; Li, Q. J.; Marx, M. A.; Paderes, G.;
Richardson, P. F.; Sach, N. W.; Walls, M.; Wells, P. A.; Zou, A. Highly
selective and potent thiophenes as PI3K inhibitors with oral antitumor
activity. Med. Chem. Lett. 2011, 2, 809−813.
8499
dx.doi.org/10.1021/jm201019k|J. Med. Chem. 2011, 54, 8490−8500

Journal of Medicinal Chemistry
Article
(28) Johnson, T. W.; Bailey, S.; Baxi, S. M.; Brooun, A.; Dinh, D. M.;
He, M.; Hoffman, J. E.; Kan, J. L. C.; Lam, H.; Li, C.; Mehta, P. P.;
Sach, N. W.; Smeal, T.; Walls, M.; Wells, P.; Yin, M.; Yu, X.; Zou, A.
The chemistry and biology of mTOR selective inhibitors. AACR
Special Conference: Targeting PI3K/mTOR Signaling in Cancer, Feb.
24−27, 2011, San Francisco, CA.
(29) Hofler, A.; Nichols, T.; Grant, S.; Lingardo, L.; Esposito, E. A.;
Gridley, S.; Murphy, S. T.; Kath, J. C.; Cronin, C. N.; Kraus, M.; Alton,
G.; Xie, Z.; Sutton, S.; Gehring, M.; Ermolieff, J. Study of the PDK1/
AKT signaling pathway using selective PDK1 inhibitors, HCS, and
enhanced biochemical assays. Anal. Biochem. 2011, 414, 179−186.
(30) GraphPad Prism, version 5.01; GraphPad Software Inc.:
San Diego.
(31) Morrison, J. F. Kinetics of the reversible inhibition of enzyme
catalysed reactions by tight-binding inhibitors. Biochim. Biophys. Acta
1969, 185, 269−286.
8500
dx.doi.org/10.1021/jm201019k|J. Med. Chem. 2011, 54, 8490−8500