Potent, selective, and orally bioavailable inhibitors of mammalian target of rapamycin (mTOR) kinase based on a quaternary substituted dihydrofuropyrimidine

Article abstract of DOI:10.1021/jm200215y

Figure Presented. A series of inhibitors of mTOR kinase based on a quaternary-substituted dihydrofuropyrimidine was designed and synthesized. The most potent compounds in this series inhibited mTOR kinase with K_i < 1.0 nM and were highly (>100 - ) selective for mTOR over the closely related PI3 kinases. Compounds in this series showed inhibition of the pathway and antiproliferative activity in cell-based assays. Furthermore, these compounds had excellent mouse PK, and showed a robust PK-PD relationship in a mouse model of cancer.

Full text of DOI:10.1021/jm200215y

ARTICLE
pubs.acs.org/jmc
Potent, Selective, and Orally Bioavailable Inhibitors of
Mammalian Target of Rapamycin (mTOR) Kinase Based
on a Quaternary Substituted Dihydrofuropyrimidine
Frederick Cohen,*,† Philippe Bergeron,† Elizabeth Blackwood,‡ Krista K. Bowman,§ Huifen Chen,†
Antonio G. DiPasquale,O Jennifer A. Epler,‡ Michael F. T. Koehler,† Kevin Lau,† Cristina Lewis,||
Lichuan Liu,# Cuong Q. Ly,† Shiva Malek,|| Jim Nonomiya,|| Daniel F. Ortwine,† Zhonghua Pei,†
Kirk D. Robarge,† Steve Sideris,|| Lan Trinh,|| Tom Truong,‡ Jiansheng Wu,r Xianrui Zhao,† and
Joseph P. Lyssikatos†
†Department of Discovery Chemistry, ‡Department of Translational Oncology, §Department of Structural Biology, Department of
Biochemical Pharmacology, #Department of Drug Metabolism and Pharmacokinetics, and rDepartment of Protein Chemistry,
Genentech, Inc., 1 DNA Way, South San Francisco, California 94080, United States
OX-ray Crystallography Facility, University of California Berkeley, 32 Lewis Hall Berkeley, California 94720, United States
S Supporting Information
b
ABSTRACT: A series of inhibitors of mTOR kinase based on a
quaternary-substituted dihydrofuropyrimidine was designed and
synthesized. The most potent compounds in this series inhibited
mTOR kinase with Ki < 1.0 nM and were highly (>100ꢁ)
selective for mTOR over the closely related PI3 kinases. Com-
pounds in this series showed inhibition of the pathway and
antiproliferative activity in cell-based assays. Furthermore, these
compounds had excellent mouse PK, and showed a robust
PKꢀPD relationship in a mouse model of cancer.
’ INTRODUCTION
competitive mTOR kinase inhibitors may circumvent many of these
limitations, as this strategy enables potent inhibition of the
mTORC1 and mTORC2 complexes and will allow for pharmaco-
logical validation of complete mTOR kinase inhibition in the clinic.
Our mTOR program began with compound 1, which was found
to be a potent inhibitor of recombinant mTOR kinase (Ki = 3 nM)
with 20ꢁ selectivity versus PI3KR (Figure 1).6 The selectivity of this
molecule was unique among this series and was attributed in large part
to the ethyl urea substituent. Our efforts to improve the properties of
this lead involved a number of heterocyclic A-rings. The knowledge
gained from this effort led us to hypothesize that cyclic ethers would
provide an optimal balance of properties. We synthesized a number of
cyclic ethers such as 3ꢀ5. Molecular modeling suggested that a small
hydrophobic group above the plane of the inhibitor at the 7-position
would provide additional selectivity over the PI3 kinases, while
blocking a benzylic position through quaternization was expected
to reduce metabolism. These considerations brought us to conceptual
compound 6. A search of the literature revealed known β-keto ester
7 as a likely precursor to compounds such as 6.7
The mammalian target of rapamycin (mTOR) signaling path-
way serves as a central regulator of cell metabolism, growth,
proliferation, and survival.1,2 The identification of mTOR as an
essential downstream component of the PI3K/Akt signaling
pathway provides a strong rationale for targeting mTOR in
tumors harboring mutations or amplification of PI3K, Akt, or
mutations or deletion of the PTEN phosphatase gene, a tumor
suppressor that negatively regulates PI3K activity.3 mTOR is a
serineꢀthreonine protein kinase of the phosphoinositide 3-kinase
(PI3K) superfamily, referred to as class IV PI3Ks, which resides in
two distinct multiprotein complexes, mTOR complex-1 (mTORC1)
and mTOR complex-2 (mTORC2).4 The mTORC1 complex is
known for its critical role in protein synthesis and can phosphor-
ylate 4EBP1 and S6K1, two key regulators of translation initiation,
whereas mTORC2 functions as a key regulator of cell growth,
metabolism, and survival and can phosphorylate Akt1. The natural
product rapamycin and semisynthetic analogues selectively inhibit
the activity of mTORC1 in an allosteric manner.5 Despite potent
and selective inhibition of the mTORC1 complex, the clinical
success of rapamycin has been limited to a few rare cancers. It has
been suggested that this limited efficacy may be due to (1) lack of
mTORC2 complex inhibition, (2) hyperactivation of PI3K signal-
ing due to suppression of a negative feedback loop, and/or (3)
partial inhibition of mTORC1 activity.3 Development of ATP-
’ RESULTS AND DISCUSSION
Although β-keto ester 7 was readily prepared, conversion to
the corresponding uracil by direct condensation with urea8 or
Received: February 24, 2011
Published: April 15, 2011
r
2011 American Chemical Society
3426
dx.doi.org/10.1021/jm200215y J. Med. Chem. 2011, 54, 3426–3435
|

Journal of Medicinal Chemistry
ARTICLE
from these assays are summarized in Table 1. In general, these
compounds were very potent mTOR inhibitors, generally with Ki
< 5 nM, and were highly selective for mTOR over the PI3 kinases.
There were a few notable exceptions to this trend. We observed
(as have others)12 that removal of the methyl group from the
morpholine leads to a significant loss in selectivity (Table 1,
compare 13b with 13c). We also observed that replacement of
the ethyl group on the urea with a small heterocycle (13a vs 13g)
led to a significant loss in selectivity for PI3KR but not for PI3Kδ,
a trend that we observed in other scaffolds (data not shown).
Since the urea substituent was predicted by MetaSite13 to be a
hot spot for CYP-mediated metabolism, significant effort went
into its replacement (Supporting Information Table S1). The
only substituent we found that maintained potency and selec-
tivity was a pyridone (13i). This group was predicted by model-
ing to retain significant hydrogen bonding interactions with the
mTOR protein. Polar compounds such as 13h (TPSA = 129;
cLogP = 1.4 compare to 13a, TPSA = 88; cLogP = 2.8), made in
an attempt to increase metabolic stability, maintained potency
in the enzymatic assays, but potency in cell-based assays was
significantly reduced.
Compounds from this series were predicted to have moderate
to high clearance (Cl) in humans, rats, and mice, with rates of
>50% of liver blood flow, based on in vitro liver microsome
data. (Table 2). When the pharmacokinetic (PK) properties were
measured in mice, we observed a good correlation (within 2ꢁ)
between predicted and observed Cl. Our hopes for pyridone 13i
as a urea replacement were dashed by its low oral bio-
availability (7%). When dosed in rats, we observed Cl rates much
higher than predicted. For example, 13a and 13d had Cl rates
of 44 and 55 (mL/min)/kg, respectively. Since the urea substi-
tuent seemed to have little effect on either in vitro or in vivo
metabolic stability, we turned ourattention to the other side of the
molecule. The benzylic position was predicted to be a primary
spot of metabolism, both by first principles14 and by MetaSite.13
Thus, we removed it by oxidation to lactones 13b and 13c.
However, this had no effect on in vitro or in vivo stability (Table 2).
As we had exhausted our options for reasonable modifications
of the gem-dimethyl scaffold, we sought to redesign it in a
manner that would allow introduction of additional structural
diversity. Our criterion was rapid access to a versatile intermedi-
ate without a large increase in molecular weight. Switching one of
the methyl groups to allyl seemed to satisfy the criterion. The
synthesis (Scheme 2) began by condensation of allyl substituted
R-hydroxy ester 1415 with ethyl acrylate to provide β-ketoester
15 along with a number of byproducts that were not readily
separated. Impure 15 was condensed with ammonia to provide
16, which was readily purified in 50% yield for the two steps.
The dichloropyrimidine ring was completed as described in
Scheme 1. The allyl group was then elaborated as described in
Schemes 2 and 3, with reference to Table 3. Finally, all racemates
or diastereomeric pairs were resolved by chiral stationary phase
HPLC. The stereochemistry of the quaternary center in 22d was
unambiguously assigned as R by X-ray crystallographic analysis
(see Supporting Information). In most cases one diastereomer or
enantiomer was significantly more potent (in either the bio-
chemical or cell-based assay) and/or selective than the other.
Since the structural basis for these trends was consistent with
predictions from molecular modeling studies (vide infra), we
assigned the chirality of the remaining analogues by analogy. The
exceptions were the pairs 21a/b and 23a/b, where the stereo-
chemistry of the quaternary center remained ambiguous. As in
Figure 1. Design strategy for geminal-disubstituted dihydrofuropyri-
midine scaffold.
thiourea9 failed to afford any identifiable products. Condensation
with a benzamidine also failed.10 Since such reactions worked
well in our hands on other substrates, we attributed this lack of
reactivity to steric hindrance of the ketone by the neighboring
quaternary center. We were able to condense β-keto ester 7 with
ammonia to provide enamine 8 in good yield (Scheme 1).11
Reaction of this enamine with COCl2, pyridine, and then
ammonia provided a urea intermediate, which cyclized to uracil 9
under gentle heating. The crude uracil was heated in a mixture of
POCl3 and 1,2-dichloroethane to provide dichloropyrimidine 10.
The inhibitors were completed using standard transformations.
In vitro potencies of these compounds were measured against
recombinant mTOR kinase domain using a LanthaScreen FRET
assay, and against PI3KR and PI3Kδ using a fluorescence
polarization assay as previously described.6 Activity in a cellular
context was assessed by measuring the inhibition of phosphor-
ylation of two known mTOR substrates, Akt (serine 473) and
p70S6K in a pTEN-null prostate carcinoma NCI-PC3 cell line.
Antiproliferative activity was determined in a 3-day viability
assay using NCI-PC3 and MCF7 neo/HER2 cell lines. Results
3427
dx.doi.org/10.1021/jm200215y |J. Med. Chem. 2011, 54, 3426–3435

Journal of Medicinal Chemistry
ARTICLE
Table 1. Structure and Potency of gem-Dimethyl Scaffolda
a Assays described in ref 6. b Assayed in NCI-PC3 cell line.
modifications in the ether ring, or on the urea, substitution at the
quaternary center led to no improvement in metabolic stability,
either in vitro or in vivo. (Table 4).
Table 2. PK Profiles of gem-Dimethyl Scaffold
mouse PKa
Molecular Modeling Studies. To help rationalize the ob-
served SAR and assist in the design of potent, selective analogues,
a homology model of the mTOR kinase domain was built from a
PI3Kγ crystal structure and selected compounds were examined
by docking.16 Consistent with reported modeling analyses, we
observed that analogues with substituents on the morpholine
ring conferred higher selectivity for mTOR over the PI3 kinases.
The mTOR-selective bridged bicyclomorpholine analogues
22c and 22f fill the larger cavity predicted for this region in
mTOR because of a Leu2354 to Phe961 (PI3Kγ numbering)
swap (Figure 2).12 In addition, there is predicted to be a small
hydrophobic pocket adjacent to the inhibitor morpholine
formed primarily by the Trp2239 residue (Ile881 in PI3Kγ).
In addition to its larger size compared with that of Ile, this Trp
is predicted to be held rigidly in place in the mTOR protein
because of a Pro residue two amino acids upstream (Lys883 in
PI3Kγ) that imparts increased rigidity to the mTOR protein
in this region relative to PI3Kγ. The 3-methyl group on the
morpholine of compounds such as 13a fits well in this pocket,
packing against the Trp, without significantly disturbing the
predicted Trp conformation. This is consistent with the high
potency and selectivity observed for these 3-methylmorpho-
line analogues.
iv (1 mg/kg)
po (5 mg/kg)
AUC PPB (%)
Cl ((mL/ Vss t1/2
F
microsome
Clhep
((mL/min)/kg)
h/r/mc
min)/kg) (L/kg) (h) (μM h) (%) h/r/mc
b
3
compd
13a
13d
13i
13c
13b
13e
13f
18/27/53
14/25/49
15/29/50
13/28/56
18/21/57
16/26/43
15/23/44
16/17/38
11/18/47
46
53
44
30
1.4
1.5
1.9
0.40
0.42
2.2
6.1
140 86/92/95
22 72/70/81
0.81
0.31
1.5
7
98/96/97
0.57 0.21
23 98/95/98
13g
13h
a Compound dosed iv as a solution in 80%PEG/10%EtOH and dosed
po as a suspension in MCT. b Clhep predicted from incubation with
microsomes. c h/r/m = human/rat/mouse.
the dimethyl analogues, most of these elaborated compounds
were potent, with mTOR Ki < 5 nM, and highly selective for
mTOR over the PI3 kinases. Potency in the enzymatic assay
translated well to potency in substrate phosphorylation and
proliferation cellular assays. Similar to findings with the structural
3428
dx.doi.org/10.1021/jm200215y |J. Med. Chem. 2011, 54, 3426–3435

Journal of Medicinal Chemistry
ARTICLE
Scheme 1. Synthesis of gem-Dimethyl
Dihydrofuropyrimidinesa
Scheme 2. Synthesis of Quaternary Substituted Analoguesa
a Reagents and conditions: (a) NH4OAc, MeOH, reflux 70%;
(b) COCl2 pyridine, CH2Cl2, then NH4OH, then reflux; (c) POCl3,
1,2-dichloroethane 65%, two steps; (d) morpholine or (S)-3-methyl-
morpholine, DIPEA, DMF; (e) PhIO, KBr, montmorillonite K10,
water, MeCN 80%; (f) (RO)2BC6H4N(H)R2, (Ph3P)4Pd, base,
MeCN, water, Δ.
a Reagents and conditions: (a) NaH, then ethyl acrylate, DMSO; (b)
NH4OAc, EtOH, Δ, 50%, two steps; (c) COCl2 pyridine, 1,2-dichloro-
ethane, then NH4OH, then Δ 22%; (d) POCl3, 1,2-dichloroethane, Δ
75%; (e) morpholine, DIPEA, DMF 92%; (f) (RO)2BC6H4N(H)Et,
(Ph3P)4Pd, base, MeCN, water, Δ; (g) CH2I2, Et2Zn, PhMe, Δ; (h) cat.
OsO4, N-methylmorpholine N-oxide, THF, water; (i) NaIO4, THF,
water; (j) NaBH4, THF, MeOH, 60%; (k) MsCl, DIPEA, CH2Cl2; (l)
XH, solvent, heat.
Another region of variability between mTOR and PI3Kγ
proteins involves Leu2185 (Ile831 in PI3Kγ) and Ile2160
(Met804 in PI3Kγ). Taken together, these branched alkyl
residues in mTOR, in addition to Pro2169 (also Pro810 in
PI3Kγ), combine to create a small hydrophobic patch above the
plane of the inhibitor bicyclic ring system as shown in Figure 2B.
This region is well-suited to interact with small alkyl groups such
as the methyl group of analogues such as 22a. The hydroxyethyl
group of 22a is predicted to extend away from this site toward
the Ser2342 residue in mTOR (Asp950 in PI3Kγ). These poten-
tial interactions may explain the increased potency and selec-
tivity observed for the S-configured 22a and allowed us to assign
the stereochemistry of most of the analogues in this series by
analogy.
Finally, in the back pocket of the binding site, the urea portion
of inhibitors such as 13a and 22a is predicted to form three hydrogen
bonds, two to Asp841 and one to Lys833, effectively locking this
portion of the inhibitor in place. In the region adjacent to the urea,
mTor presents a Glu2190 from the C-helix (Asp836 in PI3Kγ). This
Glu residue in mTOR is predicted to orient in a way that exposes the
hydrophobic ethylene portion of its side chain directly toward the
ethyl substituent on the urea. The corresponding Asp in PI3Kγ
cannot adopt a similar conformation, resulting in a more open, polar
site in this area that is less tolerant of a hydrophobic substituent on the
urea. Thus, the ethylurea provides the optimal combination of
mTOR potency and selectivity versus the PI3 kinases.
the design of urea replacements in this and other related series of
mTOR inhibitors (data not shown).
To determine the dose- and time-dependent effects of these
potent and selective inhibitors on mTOR pathway signaling
in vivo, three dose levels of compounds 13a and 22a were
administered to female NCR.nude mice bearing subcutaneous
NCI-PC3 prostate tumors. Plasma and tumor samples were
collected at 1, 6, and 10 h postdose and analyzed for drug
exposure and PD target modulation. The NCI-PC3 cell line
is PTEN-null, and as a consequence, the PI3K/Akt/mTOR
signaling axis is activated in this tumor model and has been
shown to be sensitive to rapamycin and mTOR kinase inhibi-
tors alike.17 As illustrated in Figure 3, pS6RP (mTORC1
readout) and pAkt (mTORC2 readout) levels in this model
were significantly reduced following treatment, with the max-
imum inhibition scoring at ∼15ꢀ25% the baseline levels of Akt
or pS6RP of baseline (time-matched vehicle controls). The
combination of potency and high exposure of these molecules
appears to result in maximal inhibition of the pathway at early
time points, even at the lowest dose examined. However, partial
recovery of the pAkt phospho marker is evident at later times, in a
dose-dependent manner, despite drug exposures well in excess of
the cellular IC50 at all time points. Similar observations have been
made in in vitro and in vivo studies of mTOR inhibitors.18,19
The C-helix is flexible in the kinase family, and such flexibility
certainly clouds the detailed explanation of observed potency
differences between mTOR and PI3K families. However, these
hypotheses have proven consistent in explaining observed po-
tency differences among known inhibitors and have been used in
3429
dx.doi.org/10.1021/jm200215y |J. Med. Chem. 2011, 54, 3426–3435

Journal of Medicinal Chemistry
ARTICLE
Scheme 3. Further Elaboration of Quaternary Substituted
Analoguesa
Methyl 4-Amino-5,5-dimethyl-2,5-dihydrofuran-3-carbo-
xylate (8). A solution of methyl 5,5-dimethyl-4-oxodihydrofuran-3-
carboxylate (7) (19.1 g, 111 mmol), ammonium acetate (89 g, 1.15
mol), and methanol (225 mL) was heated at reflux for 20 h. The solvent
was removed under reduced pressure and the residue partitioned
between saturated NaHCO3 (500 mL) and ethyl acetate (150 mL).
The phases were separated, and the aqueous phase was extracted with
ethyl acetate (2 ꢁ 150 mL). The combined organic phases were washed
with brine (1 ꢁ 50 mL), dried (Na2SO4), filtered, and concentrated
onto Celite to afford a free-flowing powder. The residue was chromato-
graphed: 330 g column, 5ꢀ30% ethyl acetateꢀheptane to afford 12.34 g
(65%) 7 as a colorless solid: 1H NMR (400 MHz, CDCl3) δ 5.35 (s, 2H),
4.66 (s, 2H), 3.71 (s, 3H), 1.37 (s, 6H); 13C NMR (101 MHz, CDCl3) δ
166.23, 162.47, 90.77, 84.85, 69.73, 50.47, 25.72; HRMS (ESþ) m/z
172.0936 (172.0974 calcd for C8H14NO3 M þ Hþ).
7,7-Dimethyl-5,7-dihydrofuro[3,4-d]pyrimidine-2,4(1H,3H)-
dione (9). To a cool (0 °C) solution of 8 (12.34 g, 72.1 mmol), pyridine
(23.3 mL, 288 mmol), and 1,2-dichloroethane (250 mL) was added phos-
gene (20% in toluene, 50 mL, 86.5 mmol) in one portion. The mixture was
maintained at 0 °C for 3 h. Then 28% NH4OH (80 mL) was added cautiously
down the side of the flask and the mixture was stirred gently for 3 h, then
heated at 50 °C for 16 h. Water (200 mL) was added, and the phases were
separated. The organic phase was extracted with 1% NH4OH (2 ꢁ 100 mL).
The combined aqueous phases were washed with dichloromethane (3 ꢁ
20 mL) and concentrated to approximately 150 mL, which caused the product
to precipitate. The precipitate was collected on paper, rinsed with a small
amount water, and dried under high vacuum to afford 8.27 g of 9 as colorless
crystals. Further concentration of the mother liquor provided a second crop of
product, 1.07 g (71% combined yield): 1H NMR (400 MHz, DMSO-d6) δ
11.43 (s, 1H), 11.00 (s, 1H), 4.64 (s, 2H), 1.38 (s, 6H); 13C NMR (101 MHz,
DMSO-d6) δ 160.07, 157.69, 152.64, 104.56, 83.86, 67.76, 25.42; HRMS
(ESþ) m/z 183.0770 (183.0770 calcd for C8H11N2O3 M þ Hþ).
a Reagents and conditions: (a) cat. OsO4, N-methylmorpholine N-oxide,
THF, water; (b) NaIO4, THF, water; (c) NaBH4, THF, MeOH; (d)
NaH, MeI, THF; (e) MsCl, DIPEA, CH2Cl2; (f) KO-t-Bu, THF, 77%;
(g) MeMgBr, THF; (h), Swern oxidation; (i) (RO)2BC6H4N(H)Et,
(Ph3P)4Pd, base, MeCN, water, Δ.
’ CONCLUSIONS
We designed and synthesized a series of potent and selective
mTOR inhibitors. The most potent compounds from this series
have mTOR Ki < 1 nM and >100ꢁ selectivity over the PI3KR
and PI3Kδ. These compounds were also very potent in anti-
proliferative assays with IC50 < 100 nM. Furthermore, these
compounds had moderate Cl and high oral bioavailability in
mice. The combination of good potency and high plasma
exposure leads to significant knock-down of the pathway in a
murine model of cancer at moderate doses. Further efficacy
studies in murine models and selection of a clinical candidate will
be reported in due course.
2,4-Dichloro-7,7-dimethyl-5,7-dihydrofuro[3,4-d]pyrimidine
(10). A mixture of 9 (2.55 g, 14.0 mmol), phosphoryl chloride (15 mL, 160
mmol), and 1,2-dichloroethane (80 mL) was heated at 80 °C for 20 h.
The mixture was concentrated under reduced pressure at 40 °C to a solid
which was dissolved in dichloromethane (250 mL) and saturated NaHCO3
(500 mL). The phases were separated, and the aqueous phase was extracted
with dichloromethane (3 ꢁ 50 mL). The combined organic phases were
dried (Na2SO4), filtered, and concentrated to afford 2.53 g (82%) of 10
1
as a colorless solid: H NMR (400 MHz, CDCl3) δ 5.06 (s, 2H), 1.56
(s, 6H); 13C NMR (126 MHz, CDCl3) δ 179.63, 160.63, 155.43, 127.64,
85.39, 66.89, 26.12; HRMS (ESþ) m/z 219.0092 (219.0092 calcd for
C8H9Cl2N2O M þ Hþ).
(S)-2-Chloro-7,7-dimethyl-4-(3-methylmorpholino)-5,7-dihy-
drofuro[3,4-d]pyrimidine (11a). To a cool (0 °C) solution of 10
(2.53 g, 11.5 mmol), DIPEA (4.8 mL, 28 mmol), and DMF (15 mL) was
added (S)-3-methylmorpholine (1.42 g, 14 mmol). The solution was
allowed to warm slowly over 15 h. The solution was poured into
saturated NH4Cl (100 mL) and extracted with ether (3 ꢁ 50 mL).
The combined organic phases were washed with brine (1 ꢁ 25 mL),
dried (MgSO4), filtered, and concentrated to afford 3.18 g (95%) of the
title compound as a colorless solid: 1H NMR (400 MHz, CDCl3) δ 5.10
(d, J = 11.3 Hz, 1H), 5.05 (d, J = 11.3 Hz, 1H), 4.11 (brs, 1H), 3.99 (m,
2H), 3.72 (m, 2H), 3.55 (ddd, J = 11.9, 11.9, 2.8 Hz, 1H), 3.39 (ddd, J =
13.0, 13.0, 3.2 Hz, 1H), 1.47 (s, 3H), 1.46 (s, 3H), 1.36 (d, J = 6.8 Hz,
3H); 13C NMR (101 MHz, CDCl3) δ 176.19, 160.18, 158.57, 108.03,
83.96, 70.73, 68.89, 66.81, 48.36, 40.10, 26.19, 26.10, 15.03; HRMS
(ESþ) m/z 284.1165 (284.1166 calcd for C13H19ClN3O M þ Hþ).
(S)-1-(4-(7,7-Dimethyl-4-(3-methylmorpholino)-5,7-dihy-
drofuro[3,4-d]pyrimidin-2-yl)phenyl)-3-ethylurea (13a). A mix-
ture of 11a (1.65 g, 5.66 mmol), [4-ethylureido)phenyl]boronic acid
pinacol ester (2.87 g, 9.90 mmol), tetrakis(triphenylphosphine)palladium-
(0) (440 mg, 0.38 mmol), 1.0 M Na2CO3 (7.4 mL, 7.40 mmol), 1.0 M
’ EXPERIMENTAL SECTION
All chemicals were purchased from commercial suppliers and used as
received. Flash chromatography was carried out with prepacked silica
cartridges from either ISCO or SiliCycle on an ISCO Companion
chromatography system using gradient elution. NMR spectra were
recorded on a Bruker AV III 400 or 500 NMR spectrometer and
referenced to tetramethylsilane. Preparative HPLC was performed on
a Polaris C18 5 μm column (50 mm ꢁ 21 mm), eluting with mixtures of
waterꢀacetonitrile. Low-resolution mass spectra were recorded on a
Sciex 15 mass spectrometer in ESþ mode. High-resolution mass spectra
were recorded on Waters LCT Premiere XC in ESþ mode. Diagnostic
data are provided for intermediates that are mixtures of diastereomers.
All final compounds were purified to >95% chemical and optical purity,
as assayed by HPLC (Waters Acquity UPLC column, 21 mm ꢁ 50 mm,
1.7 μm) with a gradient of 0ꢀ90% acetonitrile (containing 0.038%
TFA) in 0.1% aqueous TFA, with UV detection at λ = 254 and 210 nm,
and with CAD detection with an ESA Corona detector. In cases where
enantiomers or diastereomers were separated by chiral stationary phase
HPLC, the final products were g98% ee or de.
3430
dx.doi.org/10.1021/jm200215y |J. Med. Chem. 2011, 54, 3426–3435

Journal of Medicinal Chemistry
ARTICLE
Table 3. Structure and Potency of Quaternary Subsitituted Analoguesa
a Assays described in ref 6. nd = not determined. b Assayed in NCI-PC3 cell line.
potassium acetate (7.4 mL, 7.40 mmol), and acetonitrile (15 mL) was
heated at 110 °C in a microwave reactor for 30 min. The mixture was
partitioned between saturated NH4Cl (100 mL) and ethyl acetate (50 mL).
The phases were separated and the aqueous phase was extracted with ethyl
3431
dx.doi.org/10.1021/jm200215y |J. Med. Chem. 2011, 54, 3426–3435

Journal of Medicinal Chemistry
ARTICLE
Table 4. PK Profiles of Quaternary Substituted Analogues
mouse PKa
iv (1 mg/kg)
po (5 mg/kg)
microsomes
b
Clhep
((mL/min)/kg) Cl ((mL/ Vss (L/ t1/2 AUC
PPB (%)
compd
h/r/mc
min)/kg) kg) (h) (μM h) F (%) h/r/mc
3
22b
22a
22c
22d
23c
29g
29h
29i
21/24/53
15/24/61
17/25/67
12/15/52
10/34/74
15/24/61
13/29/59
14/17/54
40
64
47
87
76
57
55
56
1.5 0.71 0.96
1.5 0.33 0.59
2.0 0.77 0.78
2.4 0.34 0.43
2.6 0.57 0.14
56
78/76/83
110 87/nd/94
50
65
73/74/89
86/79/94
6.0 59/67/78
1.3 1.1
1.0 0.27 0.66
2.7 1.3 0.65
0.63
53
55
73
82/78/91
77/79/86
66/88/85
a Compound dosed iv as a solution in 80% PEG/10% EtOH and dosed
po as a suspension in MCT. b Clhep predicted from incubation with
microsomes. c h/r/m = human/rat/mouse.
acetate (2 ꢁ 50 mL). The combined organic phases were dried (Na2SO4),
filtered, adsorbed onto Celite, and chromatographed using an 80 g column
and 0ꢀ75% ethyl acetate in heptane to afford 2.12 g of 13a as a colorless
solid with 90% purity. A portion of this material (0.50 g) was slurried in
i-PrOH (5 mL) at 50 °C for 1 h. The material was collected by filtration on
paper, washed with i-PrOH, and dried under vacuum to afford 325 mg of
pure material: 1H NMR (500 MHz, DMSO-d6) δ 8.67 (s, 1H), 8.20 (d, J =
8.7 Hz, 2H), 7.48 (d, J = 8.8 Hz, 2H), 6.18 (t, J = 5.5 Hz, 1H), 5.15 (d, J =
11.7 Hz, 1H), 5.08 (d, J= 11.7 Hz, 1H), 4.27 (s, 1H), 3.99 (s, 1H), 3.93 (dd,
J = 11.4, 3.3 Hz, 1H), 3.72 (d, J = 11.5 Hz, 1H), 3.65 (dd, J = 11.5, 2.8 Hz,
1H), 3.50 (ddd, J = 11.8, 11.8, 2.8 Hz, 1H), 3.38ꢀ3.32 (m, 1H), 3.17ꢀ3.04
(m, 2H), 1.40 (s, 3H), 1.39 (s, 3H), 1.26 (d, J = 6.8 Hz, 3H), 1.06 (d, J = 7.2
Hz, 3H); 13C NMR (126 MHz, DMSO-d6) δ 173.48, 162.60, 157.39,
154.82, 142.65, 130.21, 128.33, 116.83, 106.87, 82.90, 70.21, 68.52, 66.21,
47.22, 33.87, 26.28, 26.23, 15.33, 14.21; HRMS (ESþ) m/z 412.2272
(412.2349 calcd for C22H30N5O3 M þ Hþ).
(()-Ethyl 5-Allyl-4-amino-5-methyl-2,5-dihydrofuran-3-
carboxylate (16). To a suspension of sodium hydride (1.9 g, 49 mmol)
in THF (50 mL) was added ethyl 2-hydroxy-2-methylpent-4-enoate15
(7.0 g, 2.78 mol) at 0 °C under N2, and the brown solution was stirred at
0 °C for 30 min and at room temperature for 30 min. The solvent was
removed under reduced pressure to afford an oil, which was cooled at
0 °C. Ethyl acrylate (14 mL, 129 mmol) in DMSO (50 mL) was added
rapidly. The ice bath was removed after 15 min, and the solution was
stirred at room temperature for 1.5 h. The reaction mixture was poured
into3% H2SO4 (700 mL) slowly at0 °C, followed byextraction withether
(3 ꢁ 150 mL). The combined organic phases were washed with brine,
dried over MgSO4, and concentrated to afford 9.40 g of β-keto ester (15)
as a clear liquid, which was used without further purification.
The crude β-keto ester from above (9.40 g, 44 mmol) was combined
with NH4OAc (34 g, 443 mmol) and EtOH (200 mL) and heated at
85 °C overnight. The solvent was removed under reduced pressure, and
the residue was diluted with EtOAc (250 mL). The precipitate was
filtered and washed with EtOAc. The combined organics were washed
with 10% NaHCO3 (1 ꢁ 150 mL), brine (1 ꢁ 150 mL), dried over
MgSO4, filtered, and concentrated under reduced pressure. The residue
was purified by flash chromatography, 220 g column, 1ꢀ15% EtOAc/
heptane, to give 4.7 g (51%, two steps) of the title compound as a light-
yellow oil. 1H NMR (500 MHz, CDCl3) δ 5.82 (ddd, J = 17.3, 10.2, 7.2
Hz, 1H), 5.55ꢀ5.25 (m, 1H), 5.18ꢀ5.05 (m, 2H), 4.74ꢀ4.57 (m, 2H),
4.24ꢀ3.98 (m, 2H), 2.49ꢀ2.22 (m, 2H), 1.37 (s, 3H), 1.27 (t, J = 7.1 Hz,
3H); 13C NMR (126 MHz, CDCl3) δ 165.99, 161.04, 132.94, 118.85,
Figure 2. (A) Depiction of 22a (pink) in the active site of the mTOR
homology model with residues important for binding and selectivity vs PI3Kγ
shown in cyan. View is edge-on to the plane of the inhibitor bicyclic ring.
Hydrogen bonds to the protein are shown as thick hashed lines. The image
was generated using Benchware3DExplorer. (B) Same view and molecule as
in (A) but with a vdW surface color-coded by lipophilic potential (tan-brown
= lipophilic, cyan-blue = polar, green = between lipophilic and polar) added to
the protein and with a mesh surface also color-coded by lipophilic potential
added to the ligand to show shape and lipophilic complementarity. Note the
brown regions match up with each other. (C) Docked model of pyridone13i
showing conserved hydrogen bonds (Tripos Associates, St. Louis, MO).
91.98, 87.11, 70.85, 59.30, 43.95, 24.52, 14.70; HRMS (ESþ) m/z
212.1276 (212.1287 calcd for C11H18NO3 M þ Hþ).
(()-7-Allyl-7-methyl-5,7-dihydrofuran[3,4-d]pyrimidine-
2,4(1H,3H)-dione (17). Phosgene (20% solution in toluene, 50.7 mL,
3432
dx.doi.org/10.1021/jm200215y |J. Med. Chem. 2011, 54, 3426–3435

Journal of Medicinal Chemistry
ARTICLE
(300 mL), made basic (pH ∼ 9) by adding NaOH pellets (few at a time)
while cooling in an ice bath. The dark brown basic mixture was extracted
with CH2Cl2 (3 ꢁ 100 mL). The combined organic phases were dried
over MgSO4, filtered, and concentrated under reduced pressure to afford
5.8 g (85%) of the title compound as a dark brown solid: 1H NMR (500
MHz, CDCl3) δ 5.69 (ddd, J = 17.7, 10.4, 7.3 Hz, 1H), 5.10ꢀ5.03 (m,
4H), 2.59 (d, J = 7.3 Hz, 2H), 1.51 (d, J = 5.8 Hz, 3H); 13C NMR (101
MHz, DMSO-d6) δ 178.32, 158.87, 154.53, 132.34, 128.97, 119.09,
86.73, 67.21, 43.31, 24.22. LCꢀMS: m/z = 246 (M þ Hþ)
7-Allyl-2-chloro-7-methyl-4-((S)-3-methylmorpholino)-5,7-
dihydrofuro[3,4-d]pyrimidine (19a). Compound 19a was prepared
from intermediate18bythe procedure forcompound 11a: 1H NMR (500
MHz, CDCl3) δ 5.76ꢀ5.66 (m, 1H), 5.11ꢀ5.02 (m, 4H), 3.98 (dd, J =
11.6, 3.6 Hz, 1H), 3.73 (m, 2H), 3.56 (t, J = 11.8 Hz, 1H), 3.56 (t, J = 11.8
Hz, 1H), 3.37 (t, J = 12.7 Hz, 1H), 3.37 (t, J = 12.7 Hz, 1H), 2.60ꢀ2.47
(m, 2H), 1.45 (s, 3H), 1.34 (m, 3H). LCꢀMS: m/z = 310 (M þ Hþ).
1-(7-Allyl-7-methyl-4-((S)-3-methylmorpholino)-5,7-dihy-
drofuro[3,4-d]pyrimidin-2-yl)phenyl)-3-ethylurea (20a). Com-
pound 20a was prepared from intermediate 19a by the procedure for
compound 13a: 1H NMR (400 MHz, CDCl3) δ 8.37 (d, J = 8.6 Hz, 2H),
7.37 (d, J = 8.7 Hz, 2H), 6.38 (s, 1H), 5.83ꢀ5.71 (m, 1H), 5.19ꢀ4.99 (m,
4H), 4.74 (s, 1H), 4.24ꢀ4.12 (m, 1H), 4.07ꢀ4.00 (m, 2H), 3.82ꢀ3.74 (m,
2H), 3.62 (ddd, J = 11.9, 11.9, 2.8 Hz, 1H), 3.42 (ddd, J = 12.6, 12.6, 2.7 Hz,
1H), 3.37ꢀ3.29 (m, 2H), 2.65ꢀ2.53 (m, 2H), 1.50 (s, 3H), 1.36 (t, J = 6.7
Hz, 3H), 1.18 (t, J = 7.2 Hz, 3H). LCꢀMS: m/z = 438 (M þ Hþ).
1-Ethyl-3-(4-((S)-7-(2-hydroxyethyl)-7-methyl-4-((S)-3-methyl-
morpholino)-5,7-dihydrofuro[3,4-d]pyrimidin-2-yl)phenyl)urea
(22a) and 1-Ethyl-3-(4-((R)-7-(2-hydroxyethyl)-7-methyl-4-((S)-
3-methylmorpholino)-5,7-dihydrofuro[3,4-d]pyrimidin-2-yl)-
phenyl)urea (22d). Olefin 20a (100 mg, 0.2 mmol) was dissolved in
THF/water (3:1, 6 mL) and cooled in an ice bath. To this solution was
added N-methylmorpholine N-oxide (32 mg, 0.27 mmol) and OsO4 (a
couple crystals). The resulting solution was stirred at room temperature
for 48 h. It was quenched by the addition of Na2SO3 (340 mg, 2.7
mmol), stirred at room temperature for 30 min, diluted with water
(10 mL), and extracted with EtOAc (2 ꢁ 20 mL). The combined
organic phases were dried over MgSO4, filtered, and concentrated under
reduced pressure to afford a crude diol, which was carried on without
further purification. LCꢀMS: m/z = 472 (M þ Hþ).
Figure 3. In vivo target modulation of mTOR pathway in NCI-PC3
human prostate tumor xenografts. Compounds13a (A) and 22a (B) were
administered as a single oral dose to tumor-bearing mice. Plasma and
tumor samples were collected at 1, 6, and 10 h postdose and analyzed for
drug exposure and PD biomarker modulation. Target inhibition was
calculated relative to the average time-matched vehicle control.
The crude diol from above was dissolved in THF/water (3:1, 6 mL),
and NaIO4 (73 mg, 0.34 mmol) was added in one portion. The mixture
was stirred at room temperature for 1.5 h. The reaction mixture was
diluted with brine (10 mL), and the aqueous layer was extracted with
EtOAc (2 ꢁ 20 mL). The combined organic phases were dried over
MgSO4, filtered, and concentrated under reduced pressure to afford
crude aldehyde, which was carried on without further purification.
LCꢀMS: m/z = 442 (M þ Hþ).
95.8 mmol) was added to a solution of enamine 16 (13.5 g, 63.9 mmol),
pyridine (20.7 mL, 256 mmol), and 1,2-dichloroethane (230 mL) at
0 °C over 10 min. After the addition, the ice bath was removed and the
mixture was stirred at room temperature for 2 h. The reaction mixture
was cooled in an ice bath, and NH4OH (89 mL, 639 mmol) was added
dropwise. After 15 min, the ice bath was removed and the mixture was
allowed to warm to room temperature and then heated in an oil bath at
50 °C overnight. The mixture was allowed to cool to room temperature,
and the phases were separated. The organic phase was extracted with 1%
NH4OH (2 ꢁ 100 mL). The combined aqueous phases were washed
with CH2Cl2 (2 ꢁ 100 mL). The aqueous phase was concentrated under
reduced pressure until solid precipitated. The solid was collected by
filtration, washed with a small amount of water (20 mL), and dried under
high vacuum to afford 5.9 g (44%) of the title compound as a light-yellow
solid: 1H NMR (500 MHz, DMSO-d6) δ 10.99 (s, 1H), 5.74ꢀ5.56 (m,
1H), 5.09ꢀ5.02 (m, 2H), 4.66 (d, J = 10.5 Hz, 1H), 4.61 (d, J = 10.5 Hz,
1H), 3.28 (s, 1H), 2.55ꢀ2.51 (m, 1H), 2.42 (dd, J = 14.3, 6.8 Hz, 1H),
1.38 (s, 3H); 13C NMR (126 MHz, DMSO-d6) δ 159.84, 156.04,
152.52, 132.74, 118.47, 105.50, 86.41, 68.88, 42.47, 24.34. LCꢀMS:
m/z = 209 (M þ Hþ).
The aldehyde from above was dissolved in THF (2 mL) with a few
drops of MeOH added. The mixture was cooled in an ice bath, and
NaBH4 (17 mg, 0.46 mmol) was added. The resulting mixture was
stirred at room temperature for 30 min. The reaction was quenched with
the addition of saturated aqueous NH4Cl (5 mL), and the aqueous layer
was extracted with EtOAc (2 ꢁ 20 mL). The combined organic phases
were dried over MgSO4, filtered, and concentrated under reduced
pressure. The residue was purified by Prep HPLC to afford 60 mg of
as white solid. A 40 mg portion of this material was purified by SFC
Chiralpak As-H, 25% MeOH in CO2, to afford 23.3 mg (60%) of 22d as
a white solid and 16.3 mg (40%) of 22a as a white solid: (22d) 1H NMR
(400 MHz, DMSO-d6) δ 8.65 (s, 1H), 8.20 (d, J = 8.8 Hz, 2H), 7.48 (d,
J = 8.8 Hz, 2H), 6.14 (t, J = 5.6 Hz, 1H), 5.16 (d, J = 11.7 Hz, 1H), 5.06
(d, J = 11.7 Hz, 1H), 4.33 (t, J = 5.3 Hz, 1H), 4.24 (s, 1H), 3.94 (d, J = 8.3
Hz, 2H), 3.68 (dt, J = 11.7, 7.2 Hz, 2H), 3.53ꢀ3.48 (m, 2H), 3.33 (s,
1H), 3.17ꢀ3.07 (m, 2H), 2.01ꢀ1.91 (m, 2H), 1.37 (s, 3H), 1.25 (d, J =
6.7 Hz, 3H), 1.06 (t, J = 7.2 Hz, 3H); 13C NMR (126 MHz, DMSO-d6) δ
(()-7-Allyl-2,4-dichloro-7-methyl-5,7-dihydrofuro[3,4-d]-
pyrimidine (18). A suspension of 17 (5.9 g, 28 mmol), 1,2-dichloro-
ethane (20 mL), and POCl3 (35 mL, 375 mmol) was heated in a glass
tube sealed with a Teflon screw-cap at 90 °C overnight. After cooling to
room temperature, the reaction mixture was poured into crushed ice
3433
dx.doi.org/10.1021/jm200215y |J. Med. Chem. 2011, 54, 3426–3435

Journal of Medicinal Chemistry
ARTICLE
172.42, 162.48, 157.34, 154.77, 142.65, 130.13, 128.32, 116.80, 108.84,
107.32, 84.31, 70.16, 69.26, 66.17, 56.76, 47.21, 42.31, 33.85, 25.34,
15.32, 14.17; HRMS (ESþ) m/z 442.2493 (442.2454 calcd for
’ ASSOCIATED CONTENT
S
Supporting Information. Structures of additional mTOR
b
1
1
C23H31N5O4 M þ Hþ); (22a) H NMR (400 MHz, DMSO-d6) δ
inhibitors, additional experimental procedures, H NMR spectra,
and X-ray data collection and analysis results. This material is
available free of charge via the Internet at http://pubs.acs.org.
8.65 (s, 1H), 8.20 (d, J = 8.8 Hz, 2H), 7.48 (d, J = 8.8 Hz, 2H), 6.14 (t, J =
5.7 Hz, 1H), 5.11 (q, J = 11.7 Hz, 2H), 4.33 (t, J = 5.3 Hz, 1H), 4.24 (s,
1H), 3.94 (d, J = 8.0 Hz, 2H), 3.72 (d, J = 10.9 Hz, 1H), 3.65 (d, J = 8.8
Hz, 1H), 3.58ꢀ3.47 (m, 2H), 3.32 (s, 1H), 3.11 (dd, J = 13.5, 6.4 Hz,
2H), 1.95 (d, J = 9.2 Hz, 2H), 1.37 (s, 3H), 1.25 (d, J = 6.7 Hz, 3H), 1.06
(t, J = 7.1 Hz, 3H); HRMS (ESþ) m/z 442.2462 (442.2454 calcd for
C23H32N5O4 M þ H).
’ AUTHOR INFORMATION
Corresponding Author
*Phone: 650-225-1000. Fax: 650-467-8922. E-mail: fcohen@
gene.com.
Molecular Modeling. A homology model of the mTOR kinase
domain was built from publicly available PI3K γ crystal structures (PDB
code 3IBE) using the Fugue and Orchestrar modules of Sybyl, version
8.1 (Tripos Associates, St. Louis, MO), and further optimized using the
Maximin module within Sybyl (all atoms, AmberFF02 charges on
protein, MMFF94 charges on ligand) including the bound ligand from
3IBE. Docking studies were conducted using Gold, version 4.1, with a
single hydrogen bond constraint to the hinge Val backbone NH,
allowing flexible Lys and Asp side chains so appropriate hydrogen bond
geometry between the ligand urea moiety and protein could be attained.
A 10 Å sphere around the centroid of the ligand was used to define the
active site region.
’ ACKNOWLEDGMENT
We thank members of the DMPK and Purification groups
within Genentech Small Molecule Drug Discovery for analytical
support.
’ ABBREVIATIONS USED
mTOR, mammalian target of rapamycin; PI3K, phosphatidyli-
nositol 3 kinase; mTORC, mammalian target of rapamycin
complex; LCꢀMS, liquid chromatographyꢀmass spectrometry;
TPSA, topological polar surface area; PPB, plasma protein
binding; PTEN, phosphatase and tensin homologue; CYP,
cytochrome P450
In-Life PKꢀPD Study. Human prostate cancer NCI-NCI-PC3
cells (National Cancer Institute, Frederick, MD) were implanted sub-
cutaneously into the right hind flanks of female NCR.nude mice (5 ꢁ
106 cells in 100 μL of Hank’s balanced salt solution). Tumors were
monitored until they reached a mean tumor volume of approximately
500 mm3. Then similarly sized tumors were randomly assigned to
groups (n = 4). Compounds 13a and 22a were formulated as suspen-
sions in 0.5% methylcellulose/0.2% Tween 80 (MCT) and dosed orally
at 25, 50, and 100 mg/kg (100 μL dose/25 g animal). Tumor and plasma
samples were harvested at 1, 6, and 10 h postdose.
’ REFERENCES
(1) Wullschleger, S.; Loewith, R.; Hall, M. N. TOR signaling in
growth and metabolism. Cell 2006, 124, 471–484.
(2) Chiang, G.; Abraham, R. T. Targeting the mTOR signaling
network in cancer. Trends Mol. Med. 2007, 13, 433–442.
Blood samples were harvested by terminal cardiac puncture and
collected into tubes containing K2EDTA as an anticoagulant. Samples
were kept chilled on ice until centrifugation at (1500ꢀ2000)g for 5 min;
plasma was transferred to a new vial and snap frozen. Tumor samples
were transferred to a 1.5 mL snap-cap polypropylene tube, weighed, and
then flash frozen on dry ice. Samples were stored at ꢀ80 °C prior to PD
biomarker evaluation.
Xenograft Tumor Lysates. Frozen tumors were lysed in ice-cold
extraction buffer by mechanical disruption and agitation with garnet/
ceramic sphere matrix in a FastPrep-24 (MP Biomedicals, Solon, OH;
five 20 s cycles with rest periods on wet ice). Extraction buffer
(Biosource; Carlsbad, CA) was supplemented with protease inhibitors
(F. Hoffman-La Roche; Mannheim, Germany), 1 mM phenylmethysu-
fonyl fluoride, and phosphatase inhibitor cocktails 1 and 2 (Sigma-
Aldrich; St. Louis, MO). After thorough disruption, tumor lysates were
clarified by centrifugation at 4 °C. Protein concentrations were deter-
mined using the BCA protein assay kit (Pierce; Rockford, IL) and
normalized to a standard concentration of 5 mg/mL.
PD Biomarker Assays. The Meso Scale Discovery Multi-Spot
biomarker detection system (Meso Scale Discovery; Gaithersburg, MD)
was used to determine the levels of Akt phosphorylated at serine 473
(pAkt) and S6RP phosphorylated at serine 235/236 (pS6RP). These
double-determinant immunoassays quantify protein levels on the basis
of measurements of electrochemoluminescence intensity. In the case of
pS473 and total Akt, both of these were measured in the same well using
a duplex assay plate. All other markers required separate plates to
measure phosphorylated and total protein. Protein was loaded at 20 μg/
well for Akt and 10 μg/well for S6RP.
(3) Liu, Q.; Thoreen, C.; Wang, J.; Sabatini, D.; Gray, N. S. mTOR
mediated anti-cancer drug discovery. Drug Discovery Today: Ther.
Strategies 2009, 6, 47–55.
(4) Guertin, D. A.; Sabatini, D. M. The pharmacology of mTOR
inhibition. Sci. Signaling 2009, 2, pe24.
(5) Sudarsanam, S.; Johnson, D. E. Functional consequences of
mTOR inhibition. Curr. Opin. Drug Discovery Dev. 2010, 13, 31–40.
(6) 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.; Salphati, L.; Sideris., S.; Tian, Q.; Tsui, V.;
Wan, N. C.; Wang, S.; Wiesmann, C.; Wong, S.; Zhu, B.-Y. Discovery of
(thienopyrimidin-2-yl)aminopyrimidines as potent, selective, and orally
available pan-PI3-kinase and dual pan-PI3-kinase/mTOR inhibitors for
the treatment of cancer. J. Med. Chem. 2010, 53, 1086–1097.
(7) Gianturco, M. A.; Friedel, P. A synthesis of 2,2,4-trimethyl-
4-hydroxytetrahydrofuran, a product of the photolysis of 4-methyl-
4-pentanone. Can. J. Chem. 1965, 43, 2121–2122.
(8) Nakai, H.; Saito, T.; Obitsu, T. CRF Antagonists and Hetero-
bicyclic Compounds. EP 1666468, 2006.
(9) Danel, K.; Larsen, E.; Pedersen, E. B. An easy synthesis of 5,
6-disubstituted acyclouridine derivatives. Synthesis 1995, 934–936.
(10) Wang, J.; Zhang, Y.; Gonzales, J. E., III; Martinborough, E.;
Zimmerman, N. Condensed Pyrimidine Compounds as Inhibitors of
Voltage-Gated Ion Channels. WO 200514558, 2005.
(11) Tang, W.; Wu, S.; Zhang, X. Enantioselective hydrogenation of
tetrasubstituted olefins of cyclic β-(acylamino)acrylates. J. Am. Chem.
Soc. 2003, 125, 9570–9571.
(12) Zask, A.; Kaplan, J.; Verheijen, J. C.; Richard, D. J.; Curran, K.;
Brooijmans, N.; Bennett, E. M.; Toral-Barza, L.; Hollander, I.; Ayral-
Kaloustian, S.; Yu, K. Morpholine derivatives greatly enhance the
selectivity of mammalian target of rapamycin (mTOR) inhibitors.
J. Med. Chem. 2009, 52, 7942–7945.
Relative levels of phosphoprotein were determined by comparison of
treated tumor lysates with an average of the time-matched vehicle
controls.
3434
dx.doi.org/10.1021/jm200215y |J. Med. Chem. 2011, 54, 3426–3435

Journal of Medicinal Chemistry
ARTICLE
(13) Cruciani, G.; Carosati, E.; De Boeck, B.; Ethirajulu, K.; Mackie,
C.; Howe, T.; Vianello, R. MetaSite: understanding metabolism in
human cytochromes from the perspective of the chemist. J. Med. Chem.
2005, 48, 6970–6979.
(14) King, R. S. Biotransformations in Drug Metabolism. In Drug
Metabolism Handbook; Nassar, A. F., Hollenberg, P. F., Scatina., J., Eds.;
John Wiley & Sons, Inc.: Hoboken, NJ, 2009; pp 17ꢀ38.
(15) Ojima, I.; Miyazawa, Y.; Kumagi, M. Asymmetric addition
of allyltrimethylsilane to (ꢀ)-menthyl pyruvate and phenylglyoxylate.
J. Chem. Soc., Chem. Commum. 1976, 927–928.
(16) Nowak, P.; Cole, D. C.; Brooijmans, N.; Bursavich, M. G.;
Curran, K. J.; Ellingboe, J. W.; Gibbons, J. J.; Hollander, I.; Hu, Y.;
Kaplan, J.; Malwitz, D. J.; Toral-Barza, L.; Verheijen, J. C.; Zask, A.;
Zhang, W.-G.; Yu, K. Discovery of potent and selective inhibitors of the
mammalian target of rapamycin (mTOR) kinase. J. Med. Chem. 2009,
52, 7081–7089.
(17) Grunwald, V.; DeGraffenried, L.; Russel, D.; Friedrichs, W. E.;
Ray, R. B.; Hidlago, M. Inhibitors of mTOR reverse doxorubicin
resistance conferred by PTEN status in prostate cancer cells. Cancer
Res. 2002, 62, 6141–6145.
(18) Javle, M. M.; Shroff, R. T.; Xiong, H.; Varadhachary, G. A.;
Fogelman, D.; Reddy, S. A.; Davis, D.; Zhang, Y.; Wolff, R. A.;
Abbruzzese, J. L. Inhibition of the mammalian target of rapamycin
(mTOR) in advanced pancreatic cancer: results of two phase II studies.
BMC Cancer 2010, 368–375.
(19) Breuleux, M.; Klopfenstein, M.; Stephan, C.; Doughty, C. A.;
Barys, L.; Maira, S. M.; Kwiatkowski, D.; Lane, H. A. Increased AKT
S473 phosphorylation after mTORC1 inhibition is rictor dependent and
does not predict tumor cell response to PI3K/mTOR inhibition. Mol.
Cancer Ther. 2009, 8, 742–753.
3435
dx.doi.org/10.1021/jm200215y |J. Med. Chem. 2011, 54, 3426–3435