ATP-competitive inhibitors of the mammalian target of rapamycin: Design and synthesis of highly potent and selective pyrazolopyrimidines

Article abstract of DOI:10.1021/jm900851f

The mammalian target of rapamycin (mTOR), a central regulator of growth, survival, and metabolism, is a validated target for cancer therapy. Rapamycin and its analogues, allosteric inhibitors of mTOR, only partially inhibit one mTOR protein complex. ATP-competitive, global inhibitors of mTOR that have the potential for enhanced anticancer efficacy are described. Structural features leading to potency and selectivity were identified and refined leading to compounds with in vivo efficacy in tumor xenograft models.

Full text of DOI:10.1021/jm900851f

J. Med. Chem. 2009, 52, 5013–5016 5013
DOI: 10.1021/jm900851f
inhibitors may be better tolerated, with the opportunity to
achieve a higher therapeutic index for enhanced clinical
efficacy. In addition, the clinical use of rapamycin analogues
in cancer treatment shows selective mTOR inhibition to be a
proven strategy. We describe here the first report of detailed
structure-activity relationships (SAR) of highly potent and
selective ATP-competitive mTOR inhibitors and the struc-
tural features that lead to selectivity over PI3K.
High throughput screening followed by hit to lead devel-
opment produced pyrazolopyrimidine 1, a potent inhibitor of
mTOR (IC50 = 9.6 nM) with low selectivity versus PI3KR
(IC50=47 nM) (Table 1).14 Phenols characteristically undergo
rapid metabolism, as evidenced by the short half-life of 1 in
nudemouse microsomes (T1/2=5 min). Therefore, initial SAR
efforts focused on a search for phenol bioisosteres that main-
tained potent mTOR inhibition while improving selectivity
and metabolic stability.
ATP-Competitive Inhibitors of the Mammalian
Target of Rapamycin: Design and Synthesis of
Highly Potent and Selective
Pyrazolopyrimidines†
Arie Zask,*,‡ Jeroen C. Verheijen,‡ Kevin Curran,‡
Joshua Kaplan,‡ David J. Richard,‡ Pawel Nowak,‡
David J. Malwitz,‡ Natasja Brooijmans,‡ Joel Bard,‡
Kristine Svenson,‡ Judy Lucas,§ Lourdes Toral-Barza,§
Wei-Guo Zhang,§ Irwin Hollander,§ James J. Gibbons,§
Robert T. Abraham,§ Semiramis Ayral-Kaloustian,‡
Tarek S. Mansour,‡ and Ker Yu§
‡Chemical Sciences and §Oncology Research, Wyeth Research,
401 N. Middletown Road, Pearl River, New York 10965
Received June 10, 2009
The pyrazolopyrimidine scaffold was constructed by cycliza-
tion of 1-benzyl-4-hydrazinylpiperidine with 2,4,6-trichloropyr-
imidine-5-carbaldehyde (Scheme 1). The resulting 4,6-dichlo-
ropyrazolopyrimidine was treated with morpholine to give 2.
Debenzylation of 2 with R-chloroethyl chloroformate gave 3.
The piperidine NH of 3 could be functionalized to 4 by reductive
amination to give alkylamines, by acylation to give amides, or by
treatment with triphosgene followed by addition of amines or
alcohols to give ureas or carbamates, respectively. Suzuki
coupling of 4 with the pinacol ester of 4-aminophenylboronic
acid gave an aniline that was converted to the corresponding
ureidophenyl or carbamoylphenyl analogue 5 by treatment with
triphosgene and an amine or alcohol, respectively. Alternatively,
5 could be prepared by protecting the piperidine NH of 3 as a
tert-butyl carbamate group and converting the chloride to the
arylmethylurea as described above. Removal of the tert-butyl
carbamate group with trifluoroacetic acid gave 6. The NH
group of 6 could be further functionalized to give 5 (-XR2=
-NHCH3) as described above.
Abstract: The mammalian target of rapamycin (mTOR), a central
regulator of growth, survival, and metabolism, is a validated target
for cancer therapy. Rapamycin and its analogues, allosteric inhibi-
tors of mTOR, only partially inhibit one mTOR protein complex.
ATP-competitive, global inhibitors of mTOR that have the poten-
tial for enhanced anticancer efficacy are described. Structural
features leading to potency and selectivity were identified and
refined leading to compounds with in vivo efficacy in tumor
xenograft models.
The mammalian target of rapamycin (mTORa) is a key
component of the phosphoinositide 3-kinase (PI3K) pathway
aberrant in a large proportion of human cancers and a target
of intense inhibitor development efforts.1-4 mTOR exists as
two protein complexes, mTORC1 and mTORC2, of which
rapamycin and its analogues selectively target mTORC1
through allosteric binding to the FKBP12-rapamycin binding
(FRB) domain. However, selective inhibition of mTORC1 by
rapamycin has been shown to cause enhancement of PI3K
signaling through a negative feedback mechanism5 and recent
reports indicate that rapamycin does not completely inhibit
mTORC1.6,7 The emerging role of mTORC2 in tumor growth
and survival, along with the lack of suppression of this path-
way by the rapamycins, makes the discovery of ATP-compe-
titive mTOR inhibitors that target both mTOR complexes
highly desirable.
Conversion of the 3-phenol group in 1 to the bioisosteric
4-acetamidophenyl group gave compound 7 which exhibited
equipotent mTOR activity (IC50=7 nM) to that of 1 but with
improved microsomal stability and an order of magnitude
greater selectivity versus PI3KR (5Â vs 58Â, respectively)
(Table 1). Interestingly, the 3-acetamidophenyl group gave
8 having a marked reduction in mTOR potency (IC50
=
2450 nM).15 The methyl 4-phenylcarbamoyl group gave
potent analogue 9 (mTOR IC50=4.6 nM) with a 3-fold in-
creased selectivity versus PI3KR (174Â) compared with 7.16
Preparation of the analogous urea gave 10 with more than
an order of magnitude greater enzyme potency (mTOR
The structural similarity of the ATP-binding sites of mTOR
and PI3K has made the search for selective mTOR inhibitors
challenging, and reports on the biology of selective mTOR
inhibitors have only recently appeared.8-11 Nonselective in-
hibitors of PI3K that potently inhibit mTOR and other
kinases have previously been reported.12,13 Selective mTOR
IC50=0.38 nM) and cell growth inhibition (LNCap IC50 =
3 nM) and comparable selectivity (108Â) and microsomal
stability.
†The atomic coordinates and the related experimental data for the
X-ray crystal structure of 26 bound to PI3Kγ has been deposited with
the Protein Data Bank: PDB code 3IBE.
*To whom correspondence should be addressed. Phone: 845-602-
2836. Fax: 845-602-5561. E-mail: zaska@wyeth.com.
a Abbreviations: mTOR, mammalian target of rapamycin; PI3K,
phosphoinositide 3-kinase; SAR, structure-activity relationship;
mTORC1, mammalian target of rapamycin complex 1; mTORC2,
mammalian target of rapamycin complex 2; FRB, FKBP12-rapamycin
binding.
The structural similarity of the ATP-binding sites of mTOR
and the PI3Ks enabled an mTOR homology model to be built
based on the X-ray crystal structure of PI3Kγ. Overall
sequence similarity between mTOR and PI3Kγ is 25% in
the kinase catalytic domain; however, the ATP-binding sites
are significantly more conserved with a sequence similarity of
68%. There are two residue differences in the hinge region,
namely, Trp2239 (mTOR) is Ile881 in PI3Kγ and Pro2241
r
2009 American Chemical Society
Published on Web 07/31/2009
pubs.acs.org/jmc

5014 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 16
Table 1. Bioisosteres of the Phenol Group
Zask et al.
selec-
PI3KRa tivityb LNCapa microsc
R3
Y
mTORa
compd
1
3-OH
N
N
9.6 ( 1.5
7.0 ( 0.9
2450 ( 450 2005 ( 162
4.6 ( 1.1
46 ( 6
47 ( 6
404 ( 96
5
53 ( 8
5
7
4-NHCOCH3
3-NHCOCH3
4-NHCO2CH3
4-NHCO2CH2-
CH3
58 66 ( 24
12
8
CH
N
9
801 ( 199 174 213 ( 13
631 ( 22
14
11
CH
14 3400 ( 306
189 11 ( 5
108 3 ( 2
Figure 1. Docking of 10 in an mTOR homology model based on a
PI3Kγ crystal structure.
12
10
4-NHCO2CH2-
CH2OH
N
N
1.5 ( 0.1
284 ( 84
6
(mTOR) is a lysine in PI3Kγ (Lys883). There are also
differences in the so-called specificity surface of the binding
site: His2241 and Lys2243 in mTOR are Asp884 and Ala885
in PI3Kγ, respectively. Finally, the “bottom” of the ATP-
binding site near the hinge region contains a leucine residue in
mTOR (residue 2354), while PI3Kγ has a phenylalanine at
this position (Phe961).
4-NHCONHCH3
0.38 ( 0.05 41 ( 2
12
a Average IC50 (nM) ( SEM. b PI3KR/mTOR. c Nude mouse micro-
somes T1/2 (min).
Scheme 1. Synthesis of Pyrazolopyrimidine Analoguesa
The X-ray structures of our inhibitors bound to PI3Kγ
(vide infra) were used as the basis for docking studies with the
mTOR homology model. Thus, modeling studies of urea 10
with the mTOR homology model show that the urea group
makes three hydrogen bonds to the ATP-binding pocket, two
between the urea NHs and Asp2195 and one between the urea
carbonyl and Lys2187 (Figure 1). In contrast, in the less
potent analogue 1, the analogous phenol group makes only
a single hydrogen bond to Asp2195.14
SAR of the carbamoylphenyl group revealed that the larger
ethyl carbamate in 11 led to lower potency than the methyl
carbamateanalogue 9 (Table 1).15 However, conversion of the
ethyl carbamate group to a hydroxyethyl carbamate gave 12
that was appreciably more potent than 9 or 11. Molecular
modeling suggests that while the methyl and ethyl carbamate
groups have hydrogen bond interactions with Asp2195 and
Lys2187, the hydroxyethyl group of 12 can possibly form a
third hydrogen bond to Asp2191 which could account for its
increased potency.
While 10 was a potent mTOR inhibitor, it suffered from
poor metabolic stability in nude mouse microsomes (T1/2=12
min). Metabolite identification studies showed that the pico-
lylpiperidine group was metabolized by dealkylation of the
piperidine ring nitrogen, leading to the less active analogue 6
(Table 2). Therefore, other piperidine ring substituents were
explored. Substitution by a benzyl group led to analogue 13
with comparable potency to 10 but lower nude mouse micro-
somal stability (T1/2 = 3 min). An N-methyl group gave
analogue 14 with decreased potency and stability. Acylation
of the piperidine gave compound 15, with increased stability
(T1/2 = 27 min) but decreased cellular activity. Similarly,
methylurea analogues 16 and 17 had decreased potency.
Prolonged microsomal stability (T1/2 > 30 min) combined
with good potency and selectivity was achieved with the
methyl carbamate group (analogue 18).
a R1, R2, and X are as defined in the tables. Reagents and conditions:
(a) triethylamine; (b) morpholine; (c) R-chloroethyl chloroformate;
(d) triphosgene, then alcohol or amine; (e) acid chloride or chlorofor-
mate; (f) aldehyde/sodium triacetoxyborohydride; (g) 4-aminophenyl-
boronic acid or 4-isocyanatophenylboronic acid/amine or alcohol,
palladium (0), sodium carbonate; (h) di-tert-butyl dicarbonate; (i) tri-
fluoroacetic acid.

Letter
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 16 5015
Table 3. SAR of the Urea and Carbamate Groups
Table 2. Exploration of the Piperidine Ring Substituent
selec-
compd
R1
R2
mTORa
PI3KRa tivityb LNCapa microsc
23 ( 10 16 1367 ( 433 >30
-CH3 0.52 ( 0.10 14 ( 2
selec-
tivityb LNCapa microsc
compd
R2
X
mTORa
PI3KRa
6
-H
-Bn
-CH3 1.4 ( 0.4
13
14
15
16
17
18
26
27 1.5 ( 0.3
10 733 ( 133
107 817 ( 17
75 2567 ( 186
190 197 ( 3
217 31 ( 5
3
2
19
20
21
22
23
24
25
27
28
-CH3
-CH2CH2OH
-Et
-CH2CH2F
-CH(CH2)2
-3-Pyr
O
4.3 ( 0.6
1.0 ( 0.2
1026 ( 434
681 ( 124
239 355 ( 45
681 28 ( 1
1153 55
>30
>30
>30
>30
>30
24
-CH3
-COCH3
-CONHCH3 -CH3 2.0 ( 0.3
-CON(CH3)2 -CH3 1.1 ( 0.1
-CO2CH3
-CO-3-Pyr
-CH3 22 ( 7
-CH3 2.1 ( 0.4
211 (40
225 ( 18
149 ( 1
209 ( 2
O
27
NH 0.32 ( 0.06 490 ( 198
NH 0.61 ( 0.03 505 ( 225
NH 0.45 ( 0.04 661 ( 156
NH 0.20 ( 0.01 35 ( 6
NH 0.30 ( 0.02 18
828 52 ( 13
1469 42 ( 3
175 31 ( 4
60 60 ( 15
44 <1
>30
>30
22
-CH3 0.46 ( 0.08 100 ( 17
-4-Pyr 0.83 ( 0.29 548 ( 401 660 150 ( 29
-4-Pyr
9
a Average IC50 (nM) ( SEM. b PI3KR/mTOR. c Nude mouse micro-
somes T1/2 (min).
-Ph-4-Pip-CH3 NH 0.34 ( 0.02 15 ( 2
-Ph-4-CH2OH NH 0.08 ( 0.01 6.0
>30
13
75 <1
a Average IC50 (nM) ( SEM. b PI3KR/mTOR. c Nude mouse micro-
somes T1/2 (min).
Consistent with expectations of an ATP-competitive, glo-
bal inhibitor of mTOR, 18 strongly suppressed mTORC1 and
mTORC2 biomarkers P-S6K(T389) and P-AKT(S473),
respectively, in nude mice bearing PTEN-null PC3MM2
xenograft tumors, upon dosing at 50 mg/kg ip, for at least
4 h. The mTOR selectivity of 18 was evidenced by the lack of
appreciable suppression of the PI3K biomarker P-AKT-
(T308). In vitro, 18 dose dependently suppressed these mTOR
biomarkers in U87MG and PC3MM2 cells and showed
selectivity versus the PI3K biomarker (Supporting Informa-
tion Figure S1). Compound 18 also suppressed tumor growth
in U87MG glioma bearing nude mice with 50 mg/kg ip once
daily dosing for 5 days with a treated over control (T/C)
value of 0.57 (P<0.05) on day 7.
The carbamoyl piperidine group also led to excellent micro-
somal stability when the methylurea in 18 was replaced with
carbamates, giving potent and selective analogues 19 and 20
(Table 3).16 Analogue 19 completely inhibited mTORC1
biomarker P-S6K(T389) for at least 6 h and substantially
inhibited mTORC2 biomarker P-AKT(S473) for 4-6 h in
nude mice bearing the PTEN-null PC3MM2 xenograft
tumors upon ip dosing at 50 mg/kg.11 As with 18, the PI3K
biomarker AKT(T308) was not significantly inhibited, show-
ing that 19 selectively inhibited mTOR. In vitro, 19 also
dose dependently suppressed these mTOR biomarkers in
U87MG and PC3MM2 cells and showed selectivity versus
the PI3K biomarker (Supporting Information Figure S1).11
Dosing at 50 mg/kg twice daily (b.i.d.) and once daily (q.d.)
for 5 days in U87MG glioma-bearing nude mice achieved
dose dependent suppression of tumor growth withT/C of 0.14
and 0.62 (P<0.05), respectively.11 Compounds with longer in
vivo biomarker suppression and/or greater inhibition of cell
growth than 18 and 19 may lead to stronger in vivo efficacy.
Compound 19 was selective for mTOR versus a panel of 24
kinases (Supporting Information Table 1) with IC50>50 μM
against these enzymes. It also showed high selectivity (1469Â)
versus PI3Kγ (IC50 = 6315 ( 1371 nM). Compound 19
(3 μmol/L) also did not block VP-16-induced PIKK substrate
phosphorylation of P-p53(S15) and P-Chk1(S345), indicating
that the underlying PIKKs were not inhibited.11
Figure 2. Movement of protein loops (green to purple) in PI3Kγ
to accommodate arylurea group of 26, leading to solvent exposed
4-position of pyridine ring.
The mTOR potency and selectivity of the urea analogues
could be tuned by modification of the urea substituent. While
alkyl groups increased selectivity, aryl groups increased
potency. Thus, conversion of the methylurea moiety in 18 to
larger alkyl groupssuch as ethyl, 2-fluoroethyl, or cyclopropyl
led to 21, 22, or 23, respectively, with ∼1000Â selectivity
versus PI3KR (Table 3). The pyridylureas 24 and 25 and other
arylureas (vide infra) maintained or improved enzyme po-
tency and cellular potency; however, selectivity versus PI3K
was generally lower than for the alkylureas.
An X-ray crystal structure of pyridylurea analogue 26
bound to PI3Kγ showed that peptide chains within the
enzyme shifted to accommodate the arylurea group and
allowed binding of arylurea analogues (Figure 2). This move-
ment within the enzyme exposes the 4-position of the aryl ring
to solvent. Water solubilizing groups are often employed in
drug development to enhance physicochemical properties and
are typically added to regions of the molecule not involved in
binding. The solvent accessibility of the 4-position of the aryl
group was therefore utilized to incorporate groups that could

5016 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 16
Zask et al.
further improve the compound properties. Thus, incorporation
of an N-methylpiperazine group at the 4-position of a phenyl-
urea led to an analogue (27) with increased water solubility at
pH 3 (>100 μg/mL) and greatly increased cellular potency
(LNCap IC50 <1 nM) (Table 3). Similar increases in cell
potency combined with potent mTOR inhibition were seen in
28 bearing a 4-hydroxymethylphenyl group.
In summary, lead optimization led to the discovery of
analogues with potent, subnanomolar ATP-competitive
mTOR kinase inhibition and>1000Â selectivity versus PI3KR.
Through the exploration of different phenol bioisosteres, urea
and hydroxyethyl carbamate groups were found to provide
excellent potency because of their formation of three hydrogen
bonds to the enzyme. Selectivity versus PI3K could be con-
trolled through variation of the substituents on the ureidophe-
nyl moiety. Metabolite identification was used to identify and
remove metabolically labile groups, thus leading to analogues
with prolonged in vivo suppression of mTOR biomarkers and
efficacy in a nude mouse xenograft model. Further refinement
of analogues is ongoing and will be reported in due course.
(6) Choo, A. Y.; Yoon, S. O.; Kim, S. G.; Roux, P. P.; Blenis, J.
Rapamycin Differentially Inhibits S6Ks and 4E-BP1 To Mediate
Cell-Type-Specific Repression of mRNA Translation. Proc. Natl.
Acad. Sci. U.S.A. 2008, 105, 17414–17419.
(7) Guertin, D. A.; Sabatini, D. M. The Pharmacology of mTOR
Inhibition. Sci. Signalling 2009, 2 (67), 1–6.
(8) 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.
(9) 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.
(10) 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.
(11) 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., in press.
(12) Maira, S.-M.; Stauffer, F.; Brueggen, J.; Furet, P.; Schnell, C.;
Fritsch, C.; Brachmann, S.; Chene, P.; De Pover, A.; Schoemaker,
K.; Fabbro, D.; Gabriel, D.; Simonen, M.; Murphy, L.; Finan, P.;
Sellers, W.; Garcia-Echeverria, C. Identification and Characteriza-
tion of NVP-BEZ235, a New Orally Available Dual Phosphatidy-
linositol 3-Kinase/Mammalian Target of Rapamycin Inhibitor
with Potent in Vivo Antitumor Activity. Mol. Cancer Ther. 2008,
7 (7), 1851–1863.
(13) 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. The Identification of 2-(1H-Indazol-4-
yl)-6-(4-methanesulfonyl-piperazin-1-ylmethyl)-4-morpholin-4-yl-
thieno[3,2-d]pyrimidine (GDC-0941) as a Potent, Selective, Orally
Bioavailable Inhibitor of Class I PI3 Kinase for the Treatment of
Cancer. J. Med. Chem. 2008, 51 (18), 5522–5532.
Acknowledgment. The authors thank Dr. Joseph Marini
and Angela Bretz for nude mouse microsome assays and
Drs. Inder Chaudhary and Jack Wang for metabolite identi-
fication determinations.
Supporting Information Available: Synthetic procedures and
characterization data for all compounds, biological methods,
mTOR homology model construction, X-ray crystallographic
study information on 26 bound to PI3Kγ, cellular biomarker
inhibition data for 18 and 19, and kinase selectivity panel testing
data for 19. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) Guertin, D. A.; Sabatini, D. M. Defining the Role of mTOR in
Cancer. Cancer Cell 2007, 12, 9–22.
(2) Chiang, G. G.; Abraham, R. T. Targeting the mTOR Signaling
Network in Cancer. Trends Mol. Med. 2007, 13, 433–442.
(3) Yuan, T. L.; Cantley, L. C. PI3K Pathway Alterations in Cancer:
Variations on a Theme. Oncogene 2008, 27, 5497–5510.
(4) Verheijen, J.; Yu, K.; Zask, A. mTOR Inhibitors in Oncology.
Annu. Rep. Med. Chem. 2008, 43, 189–199.
(14) Nowak, P.; Ayral-Kaloustian, S.; Brooijmans, N.; Cole, D. C.;
Curran, K. J.; Ellingboe, J.; Gibbons, J. J.; Hu, B.; Hollander, I.;
Kaplan, J.; Malwitz, D. J.; Mansour, T. S.; Toral-Barza, L.;
Verheijen, J.; Zask, A.; Zhang, W.-G.; Yu, K. Unpublished results.
(15) SAR reveals that benzyl and 3-picolyl groups on the piperidine ring
give compounds with similar potency and selectivity.
(5) Carracedo, A; Pandolfi, P. P. The PTEN-PI3K Pathway: Of
(16) Compounds 9 and 19 are WYE-687 and WYE-354, respectively, in
ref 11.
Feedbacks and Cross-Talks. Oncogene 2008, 27, 5527–5541.