Synthesis and structure-activity relationships of dual PI3K/mTOR inhibitors based on a 4-amino-6-methyl-1,3,5-triazine sulfonamide scaffold

Article abstract of DOI:10.1016/j.bmcl.2012.06.078

Phosphoinositide 3-kinase (PI3K) is an important target in oncology due to the deregulation of the PI3K/Akt signaling pathway in a wide variety of tumors. A series of 4-amino-6-methyl-1,3,5-triazine sulfonamides were synthesized and evaluated as inhibitors of PI3K. The synthesis, in vitro biological activities, pharmacokinetic and in vivo pharmacodynamic profiling of these compounds are described. The most promising compound from this investigation (compound 3j) was found to be a pan class I PI3K inhibitor with a moderate (>10-fold) selectivity over the mammalian target of rapamycin (mTOR) in the enzyme assay. In a U87 MG cellular assay measuring phosphorylation of Akt, compound 3j displayed low double digit nanomolar IC₅₀ and exhibited good oral bioavailability in rats (F_oral = 63%). Compound 3j also showed a dose dependent reduction in the phosphorylation of Akt in a U87 tumor pharmacodynamic model with a plasma EC₅₀ = 193 nM (91 ng/mL).

Full text of DOI:10.1016/j.bmcl.2012.06.078

Bioorganic & Medicinal Chemistry Letters 22 (2012) 5714–5720
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and structure–activity relationships of dual PI3K/mTOR inhibitors
based on a 4-amino-6-methyl-1,3,5-triazine sulfonamide scaffold
a
a
a
a
a
Ryan P. Wurz a, , Longbin Liu , Kevin Yang , Nobuko Nishimura , Yunxin Bo , Liping H. Pettus ,
Sean Caenepeel b, Daniel J. Freeman b, John D. McCarter b, Erin L. Mullady b,e, Tisha San Miguel b, Ling Wang b,
Nancy Zhang b, Kristin L. Andrews c, Douglas A. Whittington c,e, Jian Jiang d, Raju Subramanian d,
Paul E. Hughes b, Mark H. Norman a
⇑
a Department of Chemistry Research & Discovery, Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320, USA
b Department of Oncology, Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320, USA
c Department of Molecular Structure, Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320, USA
d Department of Pharmacokinetics & Drug Metabolism, Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320, USA
e Amgen Inc., 360 Binney St., Cambridge, MA 02142, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Phosphoinositide 3-kinase (PI3K) is an important target in oncology due to the deregulation of the PI3K/
Akt signaling pathway in a wide variety of tumors. A series of 4-amino-6-methyl-1,3,5-triazine sulfona-
mides were synthesized and evaluated as inhibitors of PI3K. The synthesis, in vitro biological activities,
pharmacokinetic and in vivo pharmacodynamic profiling of these compounds are described. The most
promising compound from this investigation (compound 3j) was found to be a pan class I PI3K inhibitor
with a moderate (>10-fold) selectivity over the mammalian target of rapamycin (mTOR) in the enzyme
assay. In a U87 MG cellular assay measuring phosphorylation of Akt, compound 3j displayed low double
digit nanomolar IC50 and exhibited good oral bioavailability in rats (Foral = 63%). Compound 3j also
showed a dose dependent reduction in the phosphorylation of Akt in a U87 tumor pharmacodynamic
model with a plasma EC50 = 193 nM (91 ng/mL).
Received 7 May 2012
Revised 21 June 2012
Accepted 25 June 2012
Available online 3 July 2012
Keywords:
Phosphoinositide 3-kinase
Mammalian target of rapamycin
PI3K
mTOR
Ó 2012 Elsevier Ltd. All rights reserved.
Sulfonamide
Triazine
Oncology
The family of lipid kinases termed phosphatidylinositol 3-ki-
nases (PI3Ks) catalyze the phosphorylation of the 3-hydroxyl
position of phosphatidylinositides and play key regulatory roles
in many cellular processes including cell survival, nutrient uptake,
proliferation and differentiation.1 The PI3K family consists of at
least eight proteins that share sequence homology within their
kinase domains and yet have distinct substrate specificities and
modes of regulation.5 The best characterized members of this fam-
colorectal, glioblastoma, and gastric cancers.4 In addition, PI3K
signaling is negatively regulated by the lipid phosphatase PTEN,
which is one of the most commonly mutated proteins in human
malignancy, providing further evidence for the role of the PI3K
pathway in cancer. Hence, inhibition of PI3K, and in particular its
p110
a
subunit, is a promising target for cancer therapeutics.6,7
Many industrial and academic groups have pursued both selective
PI3K8 and dual PI3K/mTOR inhibitors,9 and several of these inhib-
itors have advanced to Phase II clinical trials.10
ily are the four class I PI3Ks (a, b, c and d isoforms) that link PI3K
activity to a large variety of cell-surface receptors, comprised of
growth factor receptors as well as G protein-coupled receptors
(GPCRs).2 There is significant evidence that the PI3K/Akt pathway
is deregulated in many human cancers.3–5 For example, the gene
In a previous communication, we reported the results of our
structure–activity relationship (SAR) efforts that focused on the
optimization of a series of sulfonamides appended to a benzothia-
zole scaffold derived from an initial high-throughput screening
lead from Amgen’s proprietary sample bank.11 These efforts re-
sulted in the discovery of compound 1 which was a potent dual
inhibitor of class I PI3Ks and mTOR (Fig. 1). However, subsequent
pharmacokinetic studies revealed that compound 1 underwent
deacetylation in vivo resulting in the formation of an active metab-
olite, which complicated further development efforts.12a In
addition, the low solubility of compound 1 (0.001, 0.001 and
encoding the p110a subunit, PIK3CA, is amplified and over-
expressed in several cervical, gastric, and ovarian cancer cell lines
and is mutated in a wide variety of other cancers including breast,
⇑
Corresponding author. Tel.: +1 805 313 5400; fax: +1 805 480 1337.
E-mail address: rwurz@amgen.com (R.P. Wurz).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.06.078

R. P. Wurz et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5714–5720
5715
O
may contribute to the decreased solubility of this series. Therefore,
we proposed replacing the N-acetyl benzothiazole functionality of
compound 1 with a different linker system that lacked these
features, thereby eliminating the inherent deacetylation liability
as well as providing a potential means to enhance the solubility
of the molecules. Improved solubility could help increase the rate
of dissolution of a molecule and lead to better oral bioavailability.
Based on these considerations, we set out to explore a different lin-
ker system for the sulfonamide series.12
Separately, we have identified a second generation of PI3K
inhibitors resulting from optimization of a lead exemplified by
compound 2.15 Compound 2 was a potent pan class I PI3K inhibitor
and was selective against mTOR. While compound 2 had improved
solubility (0.2, 0.006, 0.007 mg/mL in 0.01 N HCl, PBS and SIF,
respectively) over compound 1, it had moderate in vivo pharmaco-
kinetic (PK) properties (CL = 1.7 L/h/kg in rats). We believed that
one possible approach to addressing the shortcomings of both of
these inhibitors was to pursue a hybrid strategy exemplified by
generic structure 3 wherein the chloropyridyl sulfonamide affinity
pocket moiety of 1 replaced the 2-methoxypyridyl group of 2 (
Fig. 1).
O
O
N
N
N
S
HN
N
N
N
H
SO2Me
Cl
F
O
N
N
S
NH
N
NH2
N
1
2
PI3Kα Ki 1.2 nM
mTOR IC50 2.1 nM
pAKT IC50 6.3 nM
PI3Kα Ki 9.1 nM
mTOR IC50 4.7 μM
pAKT IC50 15 nM
O
O
S
HN
R1
R
R2
N
N
N
N
H
N
N
3
NH2
Figure 1. Hybridization of compounds 1 and 2 leading to the discovery of the 4-
amino-6-methyl-1,3,5-triazine sulfonamide class of inhibitors (3).
Insight into the possible mode of binding for this proposed hy-
brid came from the X-ray co-crystal structures of compounds 1 and
2 in PI3Kc
, displayed in Figure 2.16 In the case of compound 1, the
0.013 mg/mL, in 0.01 N HCl, phosphate buffer (PBS) and simulated
intestinal fluid (SIF), respectively, Table 3) may have contributed to
the compound’s low oral bioavailability (Foral = 12%).13 Although
formulation efforts on compound 1 eventually led to improved oral
exposures through the use of a pH adjustment,14 the long-term sta-
bility of the resulting suspension also presented a development
challenge. We postulated that the presence of the acetamide
hydrogen bond donor-acceptor functionality coupled with the
fused bicyclic aromatic ring system of the N-acetyl benzothiazole
N-acetyl benzothiazole group, also referred to as the ‘linker binder’
moiety, forms two key hydrogen bonds with the backbone NH and
carbonyl of Val882. In this pair of interactions, the thiazole nitro-
gen acts as the hydrogen bond acceptor and the NH of the N-acetyl
group serves as the hydrogen bond donor.11 Whereas in compound
2, the 4-amino-6-methyl-1,3,5-triazine motif interacts with the
Val882 residue.12,17 Both compounds contain a pyridine moiety
whose nitrogen engages in a hydrogen bond with a water molecule
bridging between Tyr867 and Asp841 in the ‘affinity pocket’.
Figure 2. Overlay of the X-ray co-crystal structures of compounds 1 and 2 in the ATP binding site of unphosphorylated PI3Kc determined to 2.85 Å and 2.95 Å resolution,
respectively. PDB codes: 3QK0 and 4DK5. Dashed lines indicate hydrogen bonds. For inhibitor 1, C: pink; N: blue; O: red; S: yellow; F: light blue; Cl: dark green. For inhibitor 2,
C: green; N: blue; O: red; S: yellow.

5716
R. P. Wurz et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5714–5720
Table 1
Table 2
Enzyme and cellular assay results for compounds arising from modifications to the
sulfonamide of the affinity pocket binding motif
Enzyme and cellular assay results arising from modifications to the ribose pocket
binding motif
O
O
HN
R1
O
O
S
S
HN
R1
Cl
R
R2
N
N
N
N
N
H
N
N
H
N
N
N
N
NH2
N
NH2
Compd R1
PI3K
(nM)
a
Ki
pAkt (U87 MG)
IC50 (nM)a
cLogP Cell
Compd
R
R1
R2
PI3K
Ki
pAkt
(U87
cLogP
a
shift
(nM)a MG)
IC50
3a
F
16
187
2.9
11.7
(nM)a
Me
N
3b
3c
20
11
20
36
1.0
1.6
1.0
3.3
Cl
Me
Cl
Cl
Cl
Cl
3g
3h
3i
3.5
18
95
95
1.7
1.5
2.3
Me
OMe
Cl
Me
Me
161
114
3d
3e
3f
11
43
16
25
408
85
1.4
2.2
1.8
2.3
9.5
5.3
3j
N
12
13
24
2.1
2.5
N
N
N
Cl
Cl
Cl
Cl
3k
380
O
O
a
Data represents an average of at least two separate determinations.
3l
29
53
2.3
N
Cl
Cl
3m
3n
3o
N
O
8.6
11b
9.3
248
3.9
6.3
2.5
1.4
1.8
Compound 2 exploits the ‘ribose pocket’ of the ATP binding pocket
with a methyl(methanesulfonylpiperazine) group appended to the
5-position of the pyridine central ring.
Cl
Cl
Me
N
OMe
OMe
Herein, we report the synthesis and biological evaluation of a
series of compounds based on hybrid structure 3. Initially we
focused on inhibitors designed to explore the affinity pocket in
which the ribose pocket was unelaborated (R2 = H, Table 1). In sub-
sequent analogs, substituents were introduced to take advantage
of the ribose pocket in an attempt to improve potency, solubility
and PK properties (Table 2).
Analogs were prepared in a straightforward manner according
to Scheme 1. The 5-amino-pyridinyl-3-sulfonamide building
blocks 6 were accessed from commercially available 3-amino-5-
bromopyridines 4 followed by treatment with the desired sulfonyl
chlorides. Transformation of bromides 5 into amines 6 involved the
palladium-catalyzed cross-coupling of bromides 5 with benzophe-
none imine18 followed by hydrolysis with 1 N HCl. The triazine
linker binder moiety was prepared using a Suzuki coupling19 of
4-chloro-6-methyl-1,3,5-triazin-2-amine (7) with suitably func-
tionalized 2-fluoropyridine boronic acids 8 furnishing 4-(2-fluoro-
pyridin-3-yl)-6-methyl-1,3,5-triazin-2-amines 9. Base-promoted
Cl
OMe
O
3p
3q
3r
N
O
13
25
2.0
7.6b
7.7b
1.4b
2.2
1.3
1.2
Cl
Cl
Me
Me
O
O
N
N
Cl
Cl
Me
N
3s
3t
9.7
1.9
5.3b
3.9b
0.8
1.1
O
O
a
Data represents an average of at least two separate determinations.
Data represents a single measurement.
b
lated that the compound’s relatively high cLogP = 2.9 maybe a
contributing factor.21 With this in mind, replacement of the 4-flu-
orophenylsulfonamide with a methylsulfonamide (compound 3b)
resulted in a compound whose cLogP value was substantially re-
duced (cLogP = 1.0). This compound exhibited nearly a 10-fold
improvement in cell-shift when compared to compound 3a. It is
noteworthy to mention, the solubility of this compound was also
markedly improved to 5.8, 2.9 and 46.2 mg/mL in 0.01 N HCl, PBS
and SIF, respectively.
SNAr displacement of the 2-fluoropyridine
9 with the fully
elaborated 5-amino-pyridinyl-3-sulfonamides 6 afforded inhibi-
tors 3a–t. In this manner, three points of diversity (R, R1 and R2)
necessary for analoging could be accessed in a relatively modular
fashion.20
The inhibitory activities of the PI3K
a
enzyme for all compounds
Other analogs were prepared with small groups for R1. A cyclo-
propyl sulfonamide (compound 3c) was well tolerated, albeit a
slightly greater cell-shift was observed as compared to compound
3b. In a similar fashion, the N,N-dimethylamino-sulfonamide
analog (compound 3d) was equipotent to 3b in both enzyme and
cellular assays; however, further elaboration to an iso-propylmeth-
ylamino- or a morpholinylsulfonamide led to significant erosion in
were determined using a modified in vitro AlphaScreen assay.11
The cellular activities were determined with a U87 MG cell-based
assay measuring the phosphorylation of Akt (pAkt) at serine-473 as
a readout. The first compound (3a) in which we replaced the 2-
methoxypyridine with the chloropyridinyl sulfonamide moiety in
the affinity pocket displayed encouraging enzyme potency (PI3K
a
Ki = 16 nM, Table 1); however, there was a 12-fold shift in potency
in the U87 MG cell-based assay (pAkt IC50 = 187 nM). Although
there are a number of factors, such as permeability, pKa, and melt-
ing point, that could contribute to the large cell-shift, we postu-
PI3Ka cellular potencies (compounds 3e–f).
Preliminary results from Table 1 suggested small sulfonamide
groups (R1) were preferred in order to achieve cellular potencies
<50 nM. To further probe the relationship between cellular

R. P. Wurz et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5714–5720
5717
Table 3
HCl, PBS and SIF, respectively) as compared to its unsubstituted
analog (compound 3f); however, a large cell-shift persisted.
The next modification was to introduce a methoxy-group to the
5-position of the pyridine central ring (R2 = OMe) to reduce the
cLogP. This modification reduced cLogP and decreased the cell-shift
resulting in three compounds with single digit nanomolar potencies
in the cell-based assay (compounds 3n–p). The morpholinyl-sulfon-
amide (compound 3p) displayed a much smaller cell-shift as com-
pared to previous analogs (compound 3p vs 3f and 3m). Further
elaboration of the 5-methoxypyridine (R2 = OMe) to a 5-methoxy-
ethoxy-pyridine (3q, R2 = O(CH2)2OMe) did not affect the cellular
or enzyme potencies (compound 3q vs 3n).
To determine if additional gains in potency could be made
through the judicious choice of the ribose pocket substitution
(modifications to R2), SAR generated from optimization of com-
pound 2 was revisited and the analysis suggested hydrogen bond
acceptors such as a tetrahydropyran or methylmorpholine could
serve to improve potency.16 Indeed, introduction of a tetrahydro-
pyran into the ribose pocket resulted in compound 3r that had
Pharmacokinetic properties of selected compounds
Compd
In vitro liver
In vivo rat PKb
Microsomal stabilitya
Rat
CL
L/min/
Human
CL
ivc
pod
AUC
(ng h/
mL)
CL
(L/h/
kg)
Vdss
(L/
MRT
(h)
%F
(
l
mg)
(l
L/min/
mg)
kg)
1
2
<20
9
<14
27
<14
20
<14
<14
<14
6
<14
47
<14
26
<14
<14
0.007
1.7
0.17
2.6
26.9
1.6
4.5
2.5
1.3
2.0
1.1
1.4
103 3200
77
33
63
1.8
22
1199
4238
3360
419
3b
3j
3n
3o
3q
3r
0.39
0.39
0.21
0.52
0.46
1.76
0.40
0.97
0.28
1.01
0.52
2.57
859
28e 1210
27
321
a
Single experimental value; estimated clearance from percent parent compound
lM) remaining following a 30 min incubation in liver microsomes (0.25 mg/mL)
(1
and NADPH (1 mM).
b
Pharmacokinetic parameters following administration in male Sprague Dawley
rat; mean values from 3 animals per dosing route.
excellent enzyme potency (PI3Ka Ki = 2.0 nM) and cellular potency
c
Dosed at 1 mg/kg as a solution in DMSO.
po doses were 2 mg/kg in 1% pluronic F68, 2% hydroxypropyl methylcellulose
(pAkt IC50 = 1.4 nM).22 A similar improvement in potency was also
realized through the introduction of a methylmorpholine (com-
pounds 3s–t).
d
(HPMC), 15% hydroxypropyl b-cyclodextrin (HPBCD), 82% water/methanesulfonic
acid pH 2.2.
e
po dose was 2 mg/kg in 100% DMSO.
The X-ray structure of compound 3r co-crystallized in PI3K
c
was obtained (Fig. 3). The expected interactions between the pro-
tein and the triazine linker binder are present as previously ob-
served for compound 216 forming two hydrogen bonds with the
potency and physicochemical properties, introduction of polar
groups to the 5-position of the central pyridine ring (R2 of 3) was
explored as the crystallographic overlays of compounds 1 and 2
suggested that modifications to this portion of the inhibitor should
be well tolerated (see Fig. 2). The first compound (3g) prepared in
this series contained a chloride in the 5-position of the pyridine
(R2 = Cl). This compound was found to be equipotent to the parent
compound (R2 = H, compound 3b) (Table 2). Compounds 3h–i
examined the influence of modifying the 2-substituent (R = Cl) on
the affinity pocket binding motif. When the R-group was changed
to Me or OMe, a >30-fold loss in enzyme potency resulted and
additional modifications at this position were not pursued
Four additional analogs (compounds 3j–m) containing the 5-
chloropyridine substitution pattern (R2 = Cl) as the central pyridine
ring were examined and the potencies of the resulting compounds
mirrored previous results from Table 1. The methoxyethylmethyl-
amine sulfonamide (compound 3l) was designed to improve
PI3Kc residue Val882 of the linker region. The tetrahydropyran
moiety projects into the ribose pocket and the 2-chloro-pyridyl-
sulfonamide occupies the affinity pocket. Somewhat surprisingly,
the catalytic Lys833 sits up and away from the inhibitor and makes
only a weak hydrogen bond to an oxygen atom of the sul-
fone.11,12,16,23 The sulfonamide motif sits in a pocket formed by
the side chain of Lys833 and by Pro810, Met804, and Ser806. All
of these residues are conserved across the class I PI3K’s and all
but one are conserved in mTOR (Met804 is Ile697 in mTOR).
Although the water is unresolved, there are likely two hydrogen
bonds formed via an ordered water molecule located between
the chloropyridine ring nitrogen and the Tyr867 and Asp841 resi-
dues as previously observed for compound 1 (Fig. 2).11 Comparison
of the X-ray co-crystal stuctures of compound 2 and 3r helps ratio-
nalize the preference for a 2-chloropyridine versus a 2-methyl- or
2-methoxy-pyridine (compound 3g vs 3h and 3i).24 The preference
for the chloro-substitution in the hybrid series likely arises as a re-
sult of the affinity pocket binding motif lying deeper into the pock-
et than in compound 2, thus the more sterically demanding
solubility and was found to have modest enzyme potency on PI3K
a
with a low cell-shift. The morpholinylsulfonamide (compound 3m)
exhibited improved solubility (1.0, 1.0 and 11.1 mg/mL in 0.01 N
O
O
HN
R1
O
O
NH2
S
S
HN
R1
a
b
R
R
R
N
Br
N
N
N
NH2
Br
d
4
5
6
3a-t
R2
N
N
Cl
N
R2
N
c
F
N
+
F
N
NH2
B(OH)2
N
NH2
7
8
9
Scheme 1. General scheme for the synthesis of N-(5-((3-(4-amino-6-methyl-1,3,5-triazin-2-yl)pyridin-2-yl)amino)pyridin-3-yl)sulfonamides (3). Reagents and conditions:
(a) R1SO2Cl, pyridine, DMAP; (b) (i) Pd2(dba)3 (5 mol %), Xantphos (10 mol %), benzophenone imine (1.1 equiv), NaOt-Bu (4 equiv), DMF, microwave, 140 °C, 20 min; (ii) THF,
1 N HCl, RT, 30 min; (c) (4-NMe2C6H4Pt-Bu2)2PdCl2 (Amphos, 5 mol %), KOAc (3 equiv), 100 °C, 1,4-dioxane, 16 h; (d) LiHMDS or NaHMDS (4 equiv.), DMF or THF, 0 °C to rt,
11–87%.

5718
R. P. Wurz et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5714–5720
Figure 3. X-ray co-crystal structure of compound 3r bound in the ATP binding site of unphosphorylated PI3Kc determined to 3.0 Å resolution. PDB code: 4F1S. Dashed lines
indicate hydrogen bonds (distances are in Angstroms). For the inhibitor, C: green; N: blue; O: red; S: yellow; Cl: dark green.
Table 4
O
O
In vitro (enzyme assay) profiles of selected compoundsa
S
N
NH2
Compd
PI3K Ki (nM)
mTOR IC50 (nM)
hVps34 IC50 (nM)
Cl
Cl
Cl
N
N
N
N
a
b
c
d
N
N
N
H
N
H
3b
3j
3o
3q
20
4.4
0.6
4.7
16
2.4
1.0
2.2
14
1.2
0.4
1.1
6.1
198
163
74
1288
606
—
7.7
9.3
18
N
N
N
NH2
N
NH2
32
—
a
10
11
Data represents an average of at least two separate determinations. Details of
the assays were previously reported.11
PI3Kα Ki = 311 nM
pAKT IC50 = >5.0 μM
PI3Kα Ki = 360 nM
pAKT IC50 = >5.0 μM
Figure 4. Additional modifications to the affinity pocket binding moiety.
ance (0.39 L/kg/h; 12% of liver blood flow), low volume of distribu-
tion (0.40 L/kg), and modest oral bioavailability (Foral = 33%). This
profile represented an improvement in oral exposure compared
to compound 2. Compound 3j showed good solubility (1.0, 2.0
and 11.6 mg/mL, in 0.01 N HCl, PBS and SIF, respectively) and
exhibited the most favorable PK profile of the compounds tested
with an oral bioavailability of 63% and an improved volume of dis-
tribution (0.97 L/kg) as compared to compound 3b. These favorable
properties translated into an MRT of 2.5 h and significant oral
exposure (AUC = 3340 ng h/mL). Three compounds (3n, 3o, 3q)
that had limited solubility,25 but good cellular potencies, all
showed low to moderate bioavailabilities resulting in low oral
exposures. Compound 3r which had the best cellular potency,
unfortunately, showed high clearance of 1.8 L/kg/h despite having
excellent liver microsomal stability, suggesting a non-P450 clear-
ance mechanism in rats.
2-methoxypyridine results in a decline in potency. Additionally,
the 2-chloropyridine renders the sulfonamide nitrogen NH more
acidic (pKa = 5.7) than the corresponding 2-methylpyridine analog
(pKa = 6.9) thus favoring interaction with Lys833.
Two additional compounds were synthesized to further probe
the importance of the sulfonamide group on the potency of the
inhibitor. Compound 10, wherein the nitrogen of the methylsulf-
onamide was methylated, was found to lose potency in the PI3K
a
enzyme assay (PI3K Ki = 311 nM) (Fig. 4). This decline in potency
a
could be the result of unfavorable interactions arising from projec-
tion of a hydrophobic methyl group toward the polar Lys833
residue. Additionally, the steric interaction between the 2-chloro-
pyridine and the N-methylsulfonamide would alter the orientation
of the sulfonamide group. Compound 11, in which the sulfonamide
was removed entirely, also resulted in substantial loss in potency
as the affinity pocket is no longer optimally occupied.
The SAR investigations described above enabled the identifica-
tion of several potent PI3K inhibitors with IC50 values 6 50 nM in
the U87 MG pAkt cellular assay and with favorable PK profiles. In
The pharmacokinetic (PK) properties of some of the most potent
compounds were examined (Table 3). Compound 3b had low clear-
addition to these compounds being potent inhibitors of PI3K
were also potent inhibitors of the other class I PI3K isoforms
a, they

R. P. Wurz et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5714–5720
5719
80000
70000
60000
50000
40000
30000
20000
References and Notes
1200
1000
1. Crabbe, T.; Welham, M. J.; Ward, S. G. Trends Biochem. Sci. 2007, 32, 450.
2. Wymann, M. P.; Zvelebil, M.; Laffargue, M. Trends Pharmacol. Sci. 2003, 24,
366.
800
600
400
200
0
3. Vivanco, I.; Sawyers, C. L. Nat. Rev. Cancer 2002, 2, 489.
4. Liu, P.; Cheng, H.; Roberts, T. M.; Zhao, J. J. Nat. Rev. Drug. Disc. 2009, 8, 627.
5. Yap, T. A.; Garrett, M. D.; Walton, M. I.; Raynaud, F.; de Bono, J. S.; Workman, P.
Curr. Opin. Pharm. 2008, 8, 393.
6. Ihle, N. T.; Powis, G. Mol. Cancer Ther. 2009, 8, 1.
7. Marone, R.; Cmiljanovic, V.; Giese, B.; Wymann, M. P. Biochim. Biophys. Acta
2008, 1784, 159.
10000
0
8. For selected recent examples of selective PI3K inhibitors see: (a) Heffron, T. P.;
Wei, B.; Olivero, A.; Staben, S. T.; Tsui, V.; Do, S.; Dotson, J.; Folkes, A. J.;
Goldsmith, P.; Goldsmith, R.; Gunzner, J.; Lesnick, J.; Lewis, C.; Mathieu, S.;
Nonomiya, J.; Shuttleworth, S.; Sutherlin, D. P.; Wan, N. C.; Wang, S.;
Wiesmann, C.; Zhu, D.-Y. J. Med. Chem. 2011, 54, 7815; (b) Folkes, A. J.;
Ahmadi, K.; Alderton, W. K.; Alix, S.; Baker, S. J.; Box, G.; Chuckowree, I. S.;
Clarke, P. A.; Depledge, P.; Eccles, S. A.; Friedman, L. S.; Hayes, A.; Hancox, T. C.;
Kugendradas, A.; Lensun, L.; Moore, P.; Olivero, A. G.; Pang, J.; Patel, S.; Pergl-
Wilson, G. H.; Raynaud, F. I.; Robson, A.; Saghir, N.; Salphati, L.; Sohal, S.; Ultsch,
M. H.; Valenti, M.; Wallweber, H. J. A.; Wan, N. C.; Wiesmann, C.; Workman, P.;
Zhyvoloup, A.; Zvelebil, M. J.; Shuttleworth, S. J. J. Med. Chem. 2008, 51, 5522;
(c) Knight, Z. A.; Chiang, G. G.; Alaimo, P. J.; Kenski, D. M.; Ho, C. B.; Coan, K.;
Abraham, R. T.; Shokat, K. M. Bioorg. Med. Chem. 2004, 12, 4749.
9. For selected recent examples of dual PI3K/mTOR inhibitors see: (a) Sutherlin, D.
P.; Bao, L.; Berry, M.; Castanedo, G.; Chuckowree, I.; Dotson, J.; Folks, A.;
Friedman, L.; Goldsmith, R.; Gunzner, J.; Heffron, T.; Lesnick, J.; Lewis, C.;
Mathieu, S.; Murray, J.; Nonomiya, J.; Pang, J.; Pegg, N.; Prior, W. W.; Rouge, L.;
Salphati, L.; Sampath, D.; Tian, Q.; Tsui, V.; Wan, N. C.; Wang, S.; Wei, B.;
Wiesmann, C.; Wu, P.; Zhu, B.-Y.; Olivero, A. J. Med. Chem. 2011, 54, 7579; (b)
Sutherlin, D. P.; Sampath, D.; Berry, M.; Castanedo, G.; Chang, Z.; Chuckowree,
I.; Dotson, J.; Folkes, A.; Friedman, L.; Goldsmith, R.; Heffron, T.; Lee, L.; Lesnick,
J.; Lewis, C.; Mathieu, S.; Nonomiya, J.; Olivero, A.; Pang, J.; Prior, W. W.;
Salphati, L.; Sideris, S.; Tian, Q.; Tsui, V.; Wan, N. C.; Wang, S.; Wiesmann, C.;
Wong, S.; Zhu, B.-Y. J. Med. Chem. 2010, 53, 1086; (c) Maira, S.-M.; Stauffer, F.;
Brueggen, J.; Furet, P.; Schnell, C.; Fritsch, C.; Brachmann, S.; Chène, P.; De
Pover, A.; Schoemaker, K.; Fabbro, D.; Gabriel, D.; Simonen, M.; Murphy, L.;
Finan, P.; Sellers, W.; García-Echeverría, C. Mol. Cancer Ther. 2008, 7, 1851.
10. (a) Holmes, D. Nat. Rev. Drug Disc. 2011, 10, 563; (b) Courtney, K. D.; Corcoran,
R. B.; Engelman, J. A. J. Clin. Oncol. 2010, 28, 1075.
Veh
3 mpk 10 mpk 30 mpk
Figure 5. Effect of compound 3j in
a
U87 tumor pharmacodynamic model
measuring the inhibition of Akt (Ser 473) phosphorylation. Oral dose–response
study at 6 h post-dose. Statistical significance was evaluated by Dunnett’s method.
Bars represent the average SD (n = 3).
(PI3Kb, PI3Kc and PI3Kd) and mTOR (Table 4). As a general trend,
the hybrid class of inhibitors of type 3 containing the N-(5-ami-
no-2-chloropyridin-3-yl)sulfonamides were only modestly selec-
tive for PI3K over mTOR and had modest selectivity over hVps34,
the class III PI3K.26
Based on the favorable rat PK properties of compound 3j, it was
chosen for testing in an in vivo mouse tumor pharmacodynamic
(PD) model. Female CD1 nu/nu mice were implanted with U87 MG
tumor cells (5 Â 106 cells) and dosed orally with 3j at 3, 10, and
30 mg/kg. Six hours post-dosing, tumor cells and plasma were har-
vested. Total Akt and pAkt (Ser473) levels in tumor were measured
with a quantitative electrochemiluminescence immunoassay.11
Compound 3j showed a dose-dependent inhibition of Akt (Ser473)
phosphorylation, indicating that it effectively inhibited PI3K
in vivo (Fig. 5). The total plasma concentration for 50% tumor PD
reduction (EC50) in this experiment was calculated to be 193 nM
(91 ng/mL; 95% CI: 116, 71 ng/mL).
11. D’Angelo, N. D.; Kim, T.-S.; Andrews, K.; Booker, S. K.; Caenepeel, S.; Chen, K.;
Freeman, D.; Jiang, J.; McCarter, J. D.; San Miguel, T.; Mullady, E. L.; Schrag, M.;
Subramanian, R.; Tang, J.; Wahl, R. C.; Wang, L.; Whittington, D. A.; Wu, T.; Xi,
N.; Xu, Y.; Yakowec, P.; Zalameda, L. P.; Zhang, N.; Hughes, P.; Norman, M. H. J.
Med. Chem. 2011, 54, 1789.
The selectivity of compound 3j against a panel of 50 protein and
lipid kinases was examined in the AMBIT KINOME scan platform.27
12. Additional efforts towards this goal resulted in the exploration of an
imidazolopyridazine scaffold, see: (a) Stec, M. M.; Andrews, K. L.; Booker, S.
K.; Caenepeel, S.; Freeman, D. J.; Jiang, J.; Liao, H.; McCarter, J.; Mullady, E. L.;
San Miguel, T.; Subramanian, R.; Tamayo, N.; Wang, L.; Yang, K.; Zalameda, L.
P.; Zhang, N.; Hughes, P. E.; Norman, M. H. J. Med. Chem. 2011, 54, 5174;
Quinoline/quinoxaline scaffolds were also evaluated, see: (b) Nishimura, N.;
Siegmund, A.; Liu, L.; Yang, K.; Bryan, M. C.; Andrews, K. L.; Bo, Y.; Booker, S. K.;
Caenepeel, S.; Freeman, D.; Liao, H.; McCarter, J.; Mullady, E. L.; San Miguel, T.;
Subramanian, R.; Tamayo, N.; Wang, L.; Whittington, D. A.; Zalameda, L.;
Zhang, N.; Hughes, P. E.; Norman, M. H. J. Med. Chem. 2011, 54, 4735.
13. The initial formulation was: 20% captisol, 2% HPMC, 1% Tween 80.
14. The formulation was the same as ref. 13 except dissolve compound 1 at pH 9.0
then readjust pH to 7.0.
15. Smith, A. L.; D’Angelo, N. D.; Bo, Y. Y.; Booker, S. K.; Cee, V. J.; Herberich, B.;
Hong, F.-T.; Jackson, C. L. M.; Lanman, B. A.; Liu, L.; Nishimura, N.; Pettus, L. H.;
Reed, A. B.; Tadesse, S.; Tamayo, N. A.; Wurz, R. P.; Yang, K.; Andrews, K. L.;
Whittington, D. A.; McCarter, J. D.; San Miguel, T.; Zalameda, L.; Jiang, J.;
Subramanian, R.; Mullady, E. L.; Caenepeel, S.; Freeman, D. J.; Wang, L.; Zhang,
N.; Wu, T.; Hughes, P. E.; Norman, M. H. J. Med. Chem. 2012, 55, 5188.
In this panel, the competitive binding of 3j at 1 lM was measured
as a percentage of control (POC). For all of the kinases assayed, no
competitive binding (POC <50%) by 3j was observed. Compound 3j
also showed moderate permeability and low efflux (12.2 Â 10
À6 cm/s and ER = 1.4 and 2.0, respectively in human and rodent
LLC-PK1 cell line transfected with a MDR1 gene), and no hERG lia-
bility, but exhibited high plasma protein binding (99.8%, 99.8%,
99.65%, 97.13% in human, rat, dog and mouse, respectively).
In conclusion, a structurally novel class of pan class I PI3K inhib-
itors containing a 4-amino-6-methyl-1,3,5-triazine sulfonamide
scaffold has been generated based on a molecular structure guided
hybridization of the two previously reported classes of PI3K inhib-
itors. From this series, compound 3j was highly selective against 50
other kinases and also had superior solubility properties over com-
pound 1 and an improved PK profile over compound 2. Compound
3j was efficacious in a tumor PD assay in mice with an EC50 value of
193 nM (91 ng/mL).
16. PI3K
highly homologous to PI3K
isoform. The X-ray crystal structure of PI3K
c
is used as the more amenable surrogate protein whose structure is
. Residue numbers in the text refers to the PI3K
has been reported, see: Huang, C.-
a
c
a
H.; Mandelker, D.; Schmidt-Kittler, O.; Samuels, Y.; Velculescu, V. E.; Kinzler, K.
W.; Vogelstein, B.; Gabelli, S. B.; Amzel, L. M. Science 2007, 318, 1744.
17. A structurally related linker binder moiety has been reported in PF-04691502,
see: (a) Cheng, H.; Bagrodia, S.; Bailey, S.; Edwards, M.; Hoffman, J.; Hu, Q.;
Kania, R.; Knighton, D. R.; Marx, M. A.; Ninkovic, S.; Sun, S.; Zhang, E. Med.
Chem. Commun. 2010, 1, 139; See also: (b) Liu, K. K. C.; Huang, X.; Bagrodia, S.;
Chen, J. H.; Greasley, S.; Cheng, H.; Sun, S.; Knighton, D.; Rodgers, C.; Rafidi, K.;
Zou, A.; Xiao, J.; Yan, S. Bioorg. Med. Chem. Lett. 2011, 21, 1270.
18. Bhagwanth, S.; Adjabeng, G. M.; Hornberger, K. R. Tetrahedron Lett. 2009, 50,
1582.
19. (a) Guram, A. S.; King, A. O.; Allen, J. G.; Wang, X.; Schenkel, L. B.; Chan, J.;
Bunel, E. E.; Faul, M. M.; Larsen, R. D.; Martinelli, M. J.; Reider, P. J. Org. Lett.
2006, 8, 1787; (b) Guram, A. S.; Wang, X.; Bunel, E. E.; Faul, M. M.; Larsen, R. D.;
Martinelli, M. J. J. Org. Chem. 2007, 72, 5104.
Acknowledgments
The authors thank Randy Hungate, Terry Rosen, Rick Kendall,
and Glenn Begley for their support of this research program. Thanks
also go to Paul Andrews for his assistance with the in vitro cellular
assays. In addition, we are grateful to Jin Tang and Peter Yakowec
for expression and purification of the PI3Kc enzyme.
Supplementary data
20. For detailed experimental procedures see: Andrews, K.; Bo, Y. Y.; Booker, S.;
Cee, V. J.; D’Angelo, N.; Herberich, B. J.; Hong, F.-T.; Jackson, C. L. M.; Lanman, B.
A.; Liao, H.; Liu, L.; Nishimura, N.; Norman, M. H.; Pettus, L. H.; Reed, A. B.;
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012. 06.078.

5720
R. P. Wurz et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5714–5720
Smith, A. L.; Tadesse, S.; Tamayo, N. A.; Wu, B.; Wurz, R.; Yang, K. WO
2010126895, November 4, 2010.
21. Physicochemical properties were calculated using the Daylight Toolkit, within
Amgen’s proprietary software (ADAPT). For more information see: (a) Daylight
Chemical Information Systems Inc. http://www.daylight.com.; (b) Cho, S. J.;
Sun, Y.; Harte, W. J. Comput.-Aided Mol. Des. 2006, 20, 249.
Elkins, P. A.; Gardiner, C. M.; Garver, E.; Gilbert, S. A.; Gontarek, R. R.; Jackson, J.
R.; Kershner, K. L.; Luo, L.; Raha, K.; Sherk, C. S.; Sung, C.-M.; Sutton, D.;
Tummino, P. J.; Wegrzyn, R. J.; Auger, K. R.; Dhanak, D. ACS Med. Chem. Lett.
2010, 1, 39.
24. See Supplementary data for a superposition of the X-ray co-crystal structures
of compounds 2 and 3r in PI3Kc.
22. It should be noted that the inhibition of phosphorylation of Akt (pAkt) at S473
of PTEN-null cell lines, such as U87 MG, are also sensitive to mTOR inhibition,
therefore the cell IC50 values can be a composite of the various PI3K isoforms
as well as mTOR, see: Neshat, M. S.; Mellinghoff, I. K.; Tran, C.; Stiles, B.;
Thomas, G.; Petersen, R.; Frost, P.; Gibbons, J. J.; Wu, H.; Sawyers, C. L. Proc.
Natl. Acad. Sci. U.S.A. 2001, 98, 10314.
23. Knight, S. D.; Adams, N. D.; Burgess, J. L.; Chaudhari, A. M.; Darcy, M. G.;
Donatelli, C. A.; Luengo, J. I.; Newlander, K. A.; Parrish, C. A.; Ridgers, L. H.;
Sarpong, M. A.; Schmidt, S. J.; Van Aller, G. S.; Carson, J. D.; Diamond, M. A.;
25. For solubility data of selected potent PI3K inhibitors see the Supplementary
data.
26. Backer, J. M. Biochem. J. 2008, 410, 1.
27. For more information, see: http://www.discoverx.com/technology/technology-
kinomescan.php. The Ambit kinase panel included: Abl, Akt1, Akt2, AMPK,
AurA, BTK, CAMK2, CAMK4, CDC7, CDK2, CHK1, CK1d, CHK2, DYRK1
Erk2, FGFR1, FLT3, FYN, GSK3, HGK, IGF1R, INSR, IRAK4, KDR, LCK, LYN,
MAPKAPK2, MARK1, MET, MSK1, MST2, p38 , p70S6 K, PAK2, PIM2, PKAC2,
PKC, PKCb2, PKD2, PKG , PRAK, RAF, ROCK2, RSK1, SGK1, SRC, SYK, TAK1.
a, Erk1,
a
a