Morpholine derivatives greatly enhance the selectivity of mammalian target of rapamycin (mTOR) inhibitors

Article abstract of DOI:10.1021/jm901415x

Dramatic improvements in mTOR-targeting selectivity were achieved by replacing morpholine in pyrazolopyrimidine inhibitors with bridged morpholines. Analogues with subnanomolarmTORIC₅₀ values and up to 26000-fold selectivity versus PI3Kα were p

Full text of DOI:10.1021/jm901415x

7942 J. Med. Chem. 2009, 52, 7942–7945
DOI: 10.1021/jm901415x
Table 1. 2-Methylmorpholine and Bridged Morpholine Containing
Analogues
Morpholine Derivatives Greatly Enhance the
Selectivity of Mammalian Target of Rapamycin
(mTOR) Inhibitors
Arie Zask,*,† Joshua Kaplan,† Jeroen C. Verheijen,†
David J. Richard,† Kevin Curran,† Natasja Brooijmans,†
Eric M. Bennett,† Lourdes Toral-Barza,‡ Irwin Hollander,‡
Semiramis Ayral-Kaloustian,† and Ker Yu‡
†Chemical Sciences and ‡Oncology Research, Wyeth Research, 401
N. Middletown Road, Pearl River, New York 10965
compd R1
R2
Me
mTORa
PI3KRa
PI3KR/mTOR
Received September 22, 2009
2
1
7
8
4
1
7
8
4
7
1
7
8
4
0.46 ( 0.08 100 ( 17
0.22 ( 0.01 1803
0.22 ( 0.06 4271
2.3 ( 0.4
0.32 ( 0.06 490 ( 198
0.2 ( 0.02 5333
0.62 ( 0.10 5554
1.9 ( 0.6 721
217
8013
19414
116
24
28
16
19
25
29
17
27
20
26
30
18
Me
Me
Me
Et
Abstract: Dramatic improvements in mTOR-targeting selectivity
were achieved by replacing morpholine in pyrazolopyrimidine
inhibitors with bridged morpholines. Analogues with subnanomo-
lar mTOR IC50 values and up to 26000-fold selectivity versus PI3KR
were prepared. Chiral morpholines gave inhibitors whose enantio-
mers had different selectivity and potency profiles. Molecular
modeling suggests that a single amino acid difference between
PI3K and mTOR (Phe961Leu) accounts for the profound selectivity
seen by creating a deeper pocket in mTOR that can accommodate
bridged morpholines.
268 ( 24
1531
26665
8886
360
Et
Et
Et
c-propyl 0.62 ( 0.17 4320 ( 2341
6913
175
3-Pyr
3-Pyr
3-Pyr
3-Pyr
0.20 ( 0.01 35 ( 6
0.11 ( 0.01 613 ( 79
0.16 ( 0.02 1827
5577
11073
135
0.62 ( 0.06 81 ( 9
a Average IC50 ( SEM (nM).
The mammalian target of rapamycin (mTORa) is a mem-
ber of the phosphoinositide 3-kinase (PI3K) related kinases
(PIKKs), a family of unconventional high molecular mass
serine/threonine protein kinases. In cancer, mTOR is fre-
quently hyperactivated and is a clinically validated target for
therapy.1 While numerous reports of dual-pan PI3K/mTOR
inhibitors have appeared,2 few studies have identified selective
mTOR inhibitors.1,3-9 The development of specific mTOR
inhibitors is particularly challenging because of the extensive
conservation of the ATP-binding pockets of the PI3K family.
Nevertheless, identification of highly potent and specific
ATP-competitive mTOR inhibitors is highly desirable for
further validating mTOR as a disease target and exploiting
the therapeutic potentials of mTOR-targeting in cancer.
Selective mTOR inhibitors may be better tolerated, with the
opportunity to achieve a higher therapeutic index for en-
hanced clinical efficacy.
investigation of the effects of morpholine substitution on the
potency and selectivity of pyrazolopyrimidines incorporating
chiral(3 and 4) andachiral(5) methyl substituted morpholines
as well as chiral (6) and achiral (7 and 8) bridged morpholines.
Profound effects on potency and selectivity were achieved
with compounds incorporating these morpholine derivatives,
including enantiomeric differentiation, leading to analogues
with unprecedented potency and selectivity. Enzyme docking
studies indicated that a single amino acid difference between
mTOR and PI3K in the vicinity of the hinge region was
responsible for the mTOR selectivity of bridged morpholine
containing inhibitors.
We previously reported that morpholine 1 containing
pyrazolopyrimidines (e.g., 2, Table 1) were potent and selec-
tive ATP-competitive inhibitors of mTOR with efficacy in
cancer xenograft models in nude mice.6-8 The binding mode
of these inhibitors in a mTOR homology model based on a
PI3Kγ X-ray cocrystal structure revealed that morpholine
forms the critical kinase hinge-region binding interaction to
Val88210 in mTOR.7 Morpholine has also been shown or
proposed to be the hinge-region binding group in other
mTOR/PI3K inhibitors.9,11,12 We therefore embarked on an
Analogues were synthesized by modificationofa previously
described route,7,8 substituting a morpholine derivative 3-8
for morpholine 1 (Scheme 1). Thus, cyclization of 1-benzyl-4-
hydrazinylpiperidine with 2,4,6-trichloropyrimidine-5-car-
baldehyde gave an intermediate 4,6-dichloropyrazolopyrimi-
dine 9 which was treated with a morpholine derivative to give
10. Debenzylation of 10 with R-chloroethyl chloroformate
followed by functionalization of the piperidine NH by treat-
ment with a carbamoyl chloride gave the corresponding
carbamate 11. Suzuki coupling of 11 with the pinacol ester
of 4-aminophenylboronic acid gave aniline 12 that was
*To whom correspondence should be addressed. Phone: 845-602-2836.
Fax: 845-602-5561. E-mail: zaska@wyeth.com. Cell phone: 646-709-7801.
a Abbreviations: mTOR, mammalian target of rapamycin; PIKK,
phosphoinositide 3-kinase related kinase; PI3K, phosphoinositide 3-ki-
nase; mTORC1, mammalian target of rapamycin complex 1; mTORC2,
mammalian target of rapamycin complex 2; SAR, structure-activity
relationship.
r
pubs.acs.org/jmc
Published on Web 11/16/2009
2009 American Chemical Society

Letter
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 24 7943
Scheme 1. Synthesis of Pyrazolopyrimidine Analoguesa
Figure 1. Docking of 15 (cyan ball and stick representation), and 23
(red tube representation) in the mTOR homology model. Tyr867
and Cys885, which define the pocket width, are shown in space
filling mode. The backbone is rendered as a ribbon. Yellow lines
indicate distances from the hinge NH to the morpholine O (text
labels colored the same as the inhibitors are in A). Residues are
numbered according to their positions in PI3Kγ.
a R1 and R2 are as defined in the tables. R3 is Me or Et. Reagents and
conditions: (a) 3-8; (b) R-chloroethyl chloroformate; (c) methyl or ethyl
chloroformate; (d) 4-aminophenylboronic acid, palladium(0), sodium
carbonate; (e) triphosgene, triethylamine, then R2NH2.
The excessive width of the cis-2,6-dimethylmorpholine
group in 15 causing the morpholine displacement in the
enzyme pocket of mTOR and PI3K (vide supra) could be
designed out by constraining the equatorial methyl groups in
an axial conformation through formation of an ethylene
bridge as in 2,6-bridged morpholine 7. Thus, replacement of
cis-2,6-dimethylmorpholine in 15 with 2,6-bridged morpho-
line 7 gave 23 that showed a dramatic increase in mTOR
potency relative to 15 with subnanomolar mTOR activity
comparable to morpholine 14 (Table 2). Molecular modeling
showed that the width of the 2,6-bridged morpholine in 23 is
readily accommodated by the mTOR morpholine binding
pocket (Figure 1). Remarkably, the PI3KR activity of 23
was profoundly reduced relative to 14 leading to a 13-fold
increase in mTOR selectivity for 23 (PI3KR/mTOR = 1410).
Similarly, replacement of morpholine 1 with 7 in 2, 19, and 20
gave the corresponding 24, 25, and 26 having comparable or
lower subnanomolar mTOR IC50 and greatly decreased PI3K
activity, leading to marked increases in selectivity (Table 1).
For example, 25 has a potent IC50 of 0.2 nM versus mTOR
and an extremely high selectivity of 26665 versus PI3KR.
Potent inhibition of cellular proliferation was also retained
upon replacement of morpholine with 7. Thus, bridged mor-
Table 2. cis-2,6-Dimethylmorpholine and Bridged Morpholine Con-
taining Analogues
compd
R1
mTORa
PI3KRa
41 ( 2
3877 ( 971
677
PI3KR/mTOR
14
15
23
1
5
7
0.38 ( 0.05
84 ( 21
0.48 ( 0.01
108
46
1410
a Average IC50 ( SEM (nM).
converted to the corresponding ureidophenyl 13 by treatment
with triphosgene and an amine. Trifluoroethyl substituted
analogues were prepared according to a literature route.13
Replacement of morpholine 1 in the pyrazolopyrimidine
mTOR inhibitor 14 with cis-2,6-dimethylmorpholine 5 gave
15 that was 90- and 200-fold less potent versus PI3KR and
mTOR, respectively, than 14 (Table 2). Examination of the
hinge-region/morpholine interaction in the mTOR homology
model showed that the width of the morpholine containing
pocket is partially defined by Tyr867 and Cys885 (Figure 1).
In 15, the methyl groups of cis-2,6-dimethylmorpholine 5
adopt an equatorial conformation making the morpholine
moiety wider than the binding pocket, thus resulting in
displacement of the morpholine away from Val882 and
accounting for its loss of potency relative to morpholine
analogue 14. An equatorial methyl group giving a wider
morpholine group in racemic 2-methylmorpholine 4(R/S)
analogues 16-18 may also account for their lower potency
versus mTOR and PI3KR than the corresponding morpholine
containing analogues 2, 19, and 20 (Table 1). Similarly,
replacement of morpholine with (R)-3-methylmorpholine
3(R) (vide infra) gave 21 that was less potent versus mTOR
and PI3KR than the corresponding morpholine 1 containing
22 (Table 3).
pholine analogue 25inhibited LNCapcellgrowthwithIC50
=
9 nM while the corresponding morpholine containing analo-
gue 19 had IC50 = 55 nM.7 Compounds in this manuscript
also showed similar or greater selectivity versus other class I
PI3K isoforms. All analogues had IC50 values for PI3Kγ that
were equal to or greater than their IC50 values for PI3KR. In
addition, 27 was shown to be selective for mTOR versus
PI3Kδ (IC50 = 2629 nM) and PI3Kβ (IC50 > 10 000 nM).
Selectivity versus PIKKs ATR and hSMG1 has been demon-
strated for a related pyrazolopyrimidine containing 7.14
Docking of the bridged morpholine analogues suggests
that a single amino acid difference between mTOR and
PI3KR/PI3Kγ causes a difference in the depth of the morpho-
line binding pockets that is responsible for the increased
selectivity observed for these analogues (Figure 2).10 Model-
ing indicates that Phe961 of PI3K is too large to comfor-
tably accommodate the ethylene-bridged morpholine of
23, causing displacement of the morpholine oxygen away
from its hydrogen bonding partner, the backbone NH of
Val882 (Figure 2A). However, in mTOR, the smaller amino
acid substitution leucine (Phe961Leu) creates a deeper pocket,
which accommodates the bridged morpholine 7 without

7944 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 24
Table 3. 3-Methylmorpholine Containing Analogues
Zask et al.
compd
R1
R2
mTORa
0.17 ( 0.05 704 ( 64
1.8 ( 0.1 553
0.87 ( 0.02 4362
PI3KRa PI3KR/mTOR
22
47
21
48
49
38
50
1
Et
4141
307
3(S) Et
3(R) Et
4985
289
3(S) -(CH2)2F 1.8
3(R) -(CH2)2F 1.2 ( 0.3
3(S) 4-Pyr
3(R) 4-Pyr
521
4546
3666
32
0.58 ( 0.1
18 ( 0.5
0.28 ( 0.06 517
1846
a Average IC50 ( SEM (nM).
causing significant displacement of 23 relative to the corre-
sponding morpholine containing 14 (Figures 1 and 2B).
Superposition of the mTOR plus 23 complex on the PI3Kγ
crystal structure shows that unfavorable steric contacts with
Phe961 would be present if 23 adopted its mTOR binding
mode in PI3K (Figure 2C). Sequence alignments of other
PIKK family members show that while DNA-PK is most
similar to mTOR, with Ile at 961, other PIKKs (e.g., ATM,
ATR, and hSMG1) contain a smaller Val at position 961.
The above modeling analysis similarly applies to analogues
with 3,5-ethylene bridged morpholine 8leading to highly potent
and selective mTOR inhibitors 28-30 (Table 1). For
example, replacing the morpholine in 2 with 8 gave 28, a
subnanomolar mTOR inhibitor (IC50 = 0.22 nM) with
19414-fold selectivity versus PI3KR. In an overlay of bridged
morpholines 7 and 8 in 23 and 28, respectively, both bridging
ethylene groups are directed toward Leu961 in the deeper
mTOR morpholine binding pocket (Supporting Information
Figure S1).
Replacement of morpholine with bridged morpholines also
greatly increased the selectivity of pyrazolopyrimidine inhibi-
tors with substituents other than piperidine on the pyra-
zole ring. For example, potent and selective trifluoroethyl
analogues were prepared (Table 4). The urea SAR previously
reported for morpholine containing pyrazolopyrimidine inhi-
bitors7,8 was also seen with the bridged morpholine analogues.
Thus, larger alkylurea substituents such as 2-fluoroethyl gave
increased selectivity (i.e., 31 and 32) while arylureas gave potent
inhibitors at the expense of some selectivity (i.e., 33-35).
Replacement of morpholine 1 with bridged morpholines 7
and 8 could convert even an analogue such as 33 having potent
PI3KR activity (IC50 = 10 nM) and moderate selectivity
(43-fold) into potent mTOR inhibitors (34 and 35, respectively)
with high (>1000-fold) selectivity versus PI3KR.
Intriguingly, the chiral 2,5-methylene bridged morpho-
lines 6(RR) and 6(SS) gave pairs of enantiomers that
showed differential binding affinity for mTOR and PI3KR
(Table 4). Compounds containing bridged morpholine 6(SS)
were >14-fold more potent versus mTOR than those with
6(RR). In contrast, the difference in potency versus PI3KR
was only 2- to 3-fold. While less potent than the analogous
ethylene bridged analogues, the analogues containing 6(SS)
retain single digit mTOR potency and high selectivity versus
Figure 2. Differences in binding of the bridged morpholine in 23
to mTOR and PI3Kγ. (A) Compounds 14 (color by element) and 23
(green) docked to PI3Kγ. Yellow dotted lines indicate hydrogen
bonds, with corresponding distance measurements colored the same
as the acceptor oxygen. (B) Compounds 14 (color by element) and
23 (green) docked to mTOR. (C) Superposition of the complex
between 23 and mTOR (red) with PI3Kγ (colored by element) with
close steric contacts indicated by blue dashed lines. Distances in
Angstroms.
PI3K. For example, the 3-pyridylurea 36 has mTOR IC50
1.2 nM and 482-fold selectivity versus PI3K.
=
Similarly, (R)- and (S)-3-methylmorpholines 3(R) and 3(S)
gave pairs of enantiomers with different binding affinities for
mTOR and PI3KR (Table 3). While 3(R) containing com-
pounds were approximately 2-fold more potent versus mTOR
than the corresponding ones with 3(S), 3(S) containing com-
pounds were >8-fold more potent versus PI3KR than those
with 3(R). The different activity profiles of these enantiomers
allow for the design of mTOR inhibitors with high selectivity
(e.g., 21) or dual mTOR/PI3K inhibitors (i.e., 38) with these
morpholine derivatives.

Letter
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 24 7945
Table 4. Trifluorethyl Substituted Pyrazolopyrimidine Analogues
(2) Nuss, J. M.; Tsuhako, A. L.; Anand, N. K. Emerging Therapies
Based on Inhibitors of Phosphatidyl-Inositol-3-Kinases. Annu.
Rep. Med. Chem. 2009, 44, 339–356.
(3) Feldman, M. E.; Apsel, B.; Uotila, A.; Loewith, R.; Knight, Z. A.;
Ruggero, D.; Shokat, K. M. Active-Site Inhibitors of mTOR
Target Rapamycin-Resistant Outputs of mTORC1 and mTORC2.
PLoS Biol. 2009, 7 (2), No. e1000038; DOI: 10.1371/journal.
pbio.1000038.
(4) Thoreen, C. C.; Kang, S. A.; Chang, J. W.; Liu, Q.; Zhang, J.; Gao,
Y.; Reichling, L. J.; Sim, T.; Sabatini, D. M.; Gray, N. S. An ATP-
Competitive Mammalian Target of Rapamycin Inhibitor Reveals
Rapamycin-Resistant Functions of mTORC1. J. Biol. Chem. 2009,
284 (12), 8023–8032.
(5) Garcıa-Martınez, J. M.; Moran, J.; Clarke, R. G.; Gray, A.;
´ ´
Cosulich, S. C.; Chresta, C. M.; Alessi, D. R. Ku-0063794 Is a
Specific Inhibitor of the Mammalian Target of Rapamycin
(mTOR). Biochem. J. 2009, 421, 29–42.
(6) Yu, K.; Toral-Barza, L.; Shi, C.; Zhang, W.-G.; Lucas, J.; Shor, B.;
Kim, J.; Verheijen, J.; Curran, K.; Malwitz, D. J.; Cole, D. C.;
Ellingboe, J.; Ayral-Kaloustian, S.; Mansour, T. S.; Gibbons, J. J.;
Abraham, R. T.; Nowak, P.; Zask, A. Biochemical, Cellular and in
Vivo Activity of Novel ATP-Competitive and Selective Inhibitors
of the Mammalian Target of Rapamycin. Cancer Res. 2009, 69,
6232–6240.
(7) Zask, A; Verheijen, J. C.; Curran, K.; Kaplan, J.; Richard, D. J.;
Nowak, P.; Malwitz, D. J.; Brooijmans, N.; Bard, J.; Svenson,
K.; Lucas, J.; Toral-Barza, L.; Zhang, W.-G.; Hollander, I.;
Gibbons, J. J.; Abraham, R. T.; Ayral-Kaloustian, S.; Mansour,
T. S.; Yu, K. ATP-Competitive Inhibitors of the Mammalian
Target of Rapamycin: Design and Synthesis of Highly Potent
and Selective Pyrazolopyrimidines. J. Med. Chem. 2009, 52,
5013–5016.
(8) Verheijen, J.; Richard, D.; Curran, K.; Kaplan, J.; Lefever, M.;
Nowak, P.; Malwitz, D.; Brooijmans, N.; Toral-Barza, L.; Zhang,
W.-G.; Lucas, J.; Hollander, I.; Ayral-Kaloustian, S.; Mansour, T.;
Yu, K.; Zask, A. Discovery of 4-Morpholino-6-aryl-1H-pyrazolo-
[3,4-d]pyrimidines as Highly Potent and Selective, ATP-Competi-
tive Inhibitors of the mammalian Target of Rapamycin (mTOR):
Optimization of the 6-Aryl Substituent. J. Med. Chem., DOI:
10.1021/jm9013828.
(9) Malagu, K.; Duggan, H.; Menear, K.; Hummersone, M.; Gomez,
S.; Bailey, C.; Edwards, P.; Drzewiecke, J.; Leroux, F.; Quesada,
M. J.; Hermann, G.; Maine, S.; Molyneaux, C.-A.; Le Gall, A.;
Pullen, J.; Hickson, I.; Smith, L.; Maguire, S.; Martin, N.;
Smith, G.; Pass, M. The Discovery and Optimization of Pyrido-
[2,3-d]pyrimidine-2,4-diamines as Potent and Selective Inhibitors
of mTOR Kinase. Bioorg. Med. Chem. Lett. 2009, 19, 5950–
5953.
a Average IC50 ( SEM (nM).
In summary, morpholine derivatives can profoundly
influence the mTOR and PI3K binding affinity of
pyrazolopyrimidine analogues containing them, resulting in
potent subnanomolar mTOR inhibitors with unprecedented
selectivity for mTOR (>20000-fold). Molecular modeling
suggests that increased mTOR selectivity in bridged morpho-
line containing inhibitors is caused by a leucine for phenylala-
nine substitution in mTOR versus PI3K that creates a deeper
pocket that can better accommodate bridged morpholines.
Chiral morpholine derivatives gave inhibitors whose enantio-
mers had differential binding affinity for mTOR and PI3K
resulting in different selectivity and potency profiles. Inclusion
of morpholine derivatives in pyrazolopyrimidines and other
scaffolds to produce mTOR selective analogues with potent in
vivo anticancer efficacy will be reported in due course.
(10) Amino acid residues are numbered according to their positions in
PI3Kγ. Val882 in PI3Kγ is Val2240 in mTOR.
(11) Walker, E. H.; Pacold, M. E.; Perisic, O.; Stephens, L.; Hawkins,
P. T.; Wymann, M. P.; Williams, R. L. Structural Determinants of
Phosphoinositide 3-Kinase Inhibition by Wortmannin, LY294002,
Quercetin, Myricetin, and Staurosporine. Mol. Cell 2000, 6, 909–
919.
(12) Knight, Z. A.; Gonzalez, B.; Feldman, M. E.; Zunder, E. R.;
Goldenberg, D. D.; Williams, O.; Loewith, R.; Stokoe, D.;
Balla, A.; Toth, B.; Balla, T.; Weiss, W. A.; Williams, R. L.;
Shokat, K. M. A Pharmacological Map of the PI3-K Family
Defines a Role for p110R in Insulin Signaling. Cell 2006, 125,
733–747.
(13) Richard, D. J.; Verheijen, J. C.; Curran, K.; Kaplan, J.; Brooijmans,
N.;Toral-Barza, L.;Hollander, I.;Yu, K.; Zask, A. Incorporationof
Water-Solubilizing Groups in Pyrazolopyrimidine mTOR Inhibi-
tors; Discovery of Highly Potent and Selective Analogs with Im-
proved Human Microsomal Stability. Bioorg. Med. Chem. Lett.
2009, 19, 6830-6835.
(14) Yu, K.; Shi, C.; Toral-Barza, L.; Lucas, J.; Shor, B.; Kim, J. E.;
Zhang, W.-G.; Mahoney, R.; Gaydos, C.; Tardio, L.; Kim, K.;
Curran, K.; Ayral-Kaloustian, S.; Mansour, T. S.; Abraham, R. T.;
Zask, A.; Gibbons, J. J. Beyond Rapalog Therapy: Preclinical
Pharmacology and Antitumor Activity of WYE-125132, an ATP-
Competitive and Specific Inhibitor of mTORC1 and mTORC2.
Cancer Res. In Press.
Acknowledgment. The authors thank Drs. Tarek Mansour,
Robert Abraham, and James Gibbons for program support
and contributions.
Supporting Information Available: Experimental, biological,
and molecular modeling methods. This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) Verheijen, J.; Yu, K.; Zask, A. mTOR Inhibitors in Oncology.
Annu. Rep. Med. Chem. 2008, 43, 189–199.