Design and synthesis of benzoazepin-2-one analogs as allosteric binders targeting the PIF pocket of PDK1

Article abstract of DOI:10.1016/j.bmcl.2010.05.019

A novel series of benzoazepin-2-ones were designed and synthesized targeting the PIF pocket of AGC protein kinases, among which a series of thioether-linked benzoazepin-2-ones were discovered to bind to the PIF pocket of 3-phosphoinositide-dependent kinase-1 (PDK1), and to displace the PIF peptide with an EC₅₀ values in the lower micromolar range. The structure-activity relationships (SARs) of the linker region, tail region, and distal region were explored to further optimize these novel binders which target the PIF pocket of PDK1. When tested in an in vitro PDK1 enzymatic assay using a peptide substrate, the benzodiazepin-2-ones increased the activity of the enzyme in a concentration-dependent fashion, indicating these compounds act as PDK1 allosteric activators. These new compounds may be further developed as therapeutic agents for the treatment of diseases where the PDK1-mediated AGC protein kinases are dysregulated.

Full text of DOI:10.1016/j.bmcl.2010.05.019

Bioorganic & Medicinal Chemistry Letters 20 (2010) 3897–3902
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of benzoazepin-2-one analogs as allosteric binders
targeting the PIF pocket of PDK1
Linyi Wei a, , Xiaoqi Gao a, Robert Warne b, Xueshi Hao a, Dirksen Bussiere c, Xiang-ju Gu a,à
,
a,§,
Tetsuo Uno a, , Yi Liu
*
*
a Genomics Institute of the Novartis Research Foundation (GNF), 10675 John Jay Hopkins Dr. San Diego, CA 92121, USA
b Novartis Institutes for BioMedical Research, 4560 Horton Street, Emeryville, CA 94608, USA
c Global Discovery Chemistry, Novartis Institutes for BioMedical Research, 4560 Horton Street, Emeryville, CA 94608, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 10 March 2010
Revised 6 May 2010
Accepted 10 May 2010
Available online 12 May 2010
A novel series of benzoazepin-2-ones were designed and synthesized targeting the PIF pocket of AGC pro-
tein kinases, among which a series of thioether-linked benzoazepin-2-ones were discovered to bind to
the PIF pocket of 3-phosphoinositide-dependent kinase-1 (PDK1), and to displace the PIF peptide with
an EC50 values in the lower micromolar range. The structure–activity relationships (SARs) of the linker
region, tail region, and distal region were explored to further optimize these novel binders which target
the PIF pocket of PDK1. When tested in an in vitro PDK1 enzymatic assay using a peptide substrate, the
benzodiazepin-2-ones increased the activity of the enzyme in a concentration-dependent fashion, indi-
cating these compounds act as PDK1 allosteric activators. These new compounds may be further devel-
oped as therapeutic agents for the treatment of diseases where the PDK1-mediated AGC protein kinases
are dysregulated.
Keywords:
3-Phosphoinositide-dependent kinase-1
PDK1
Allosteric activators
AGC protein kinases
PIF pocket
Ó 2010 Elsevier Ltd. All rights reserved.
The activation of phosphatidylinositol 3-kinase (PI3K) by extra-
cellular insulin and growth factors initiates intracellular protein
phosphorylation cascades via the generation of the lipid second
messenger phosphatidylinositol 3,4,5-trisphosphate (PIP3).1 These
signal events lead to the activation of a structurally related group
of AGC (cAMP-dependent, cGMP-dependent, protein kinase C)
family protein kinases, including protein kinase B (PKB)/Akt, p70
ribosomal S6 kinase (S6K), p90 ribosomal S6 kinase (RSK), serum-
and glucocorticoid-inducible kinase (SGK), mitogen- and stress-
activated protein kinase (MSK), isoforms of protein kinase C
(PKC), protein kinase C-related kinase (PRK), and others. It is well
established that AGC protein kinases regulate a diverse array of
physiological processes relevant to cell metabolism, growth, prolif-
eration, and survival2 through distinct essential signaling path-
ways mediated by PI3K,3–6 mitogen-activated protein (MAP)
kinase7,8 and Rho GTPase9 and others. The deregulation of AGC
family protein kinases has been found to be associated with human
diseases such as cancer and diabetes.10,11
Despite the involvement in a variety of signaling pathways, the
AGC protein kinases share two key regulatory features that are crit-
ical for their activation. First, their activation requires phosphory-
lation of a serine or threonine residue in the activation loop
within the kinase domain, which results in a partial activation.
For PKB/Akt,12,13 S6K,14,15 RSK,16,17 SGK5,6 and PRK,9 the phosphor-
ylation of their activation loops is catalyzed by 3-phosphoinosi-
tide-dependent kinase-1 (PDK1), an upstream master AGC kinase
which phosphorylates and activates other members of the AGC ki-
nase family. Second, most AGC family members also require the
phosphorylation of a serine or threonine residue in a hydrophobic
motif (HM) located in a conserved region at C-terminal of the ki-
nase domains which results in the full activation.12,13,18 In PKB/
Akt, conventional PKCs, S6K and others, the HM is characterized
by three aromatic amino acids surrounding the serine/threonine
residue: Phe-X–X-Phe-Ser/Thr-Phe/Tyr; whereas in PRK and atypi-
cal PKCs, the same sequence is contained except that a negatively
charged amino acid (aspartate or glutamate) mimics the phosphor-
ylated serine/threonine.19 With few exceptions, phosphorylation of
both the activation loop and HM is required for the maximal acti-
vation of AGC protein kinases.
* Corresponding authors.
E-mail addresses: tuno@gnf.org, unotetsuo@gmail.com (T. Uno), yi@intellikine.
com (Y. Liu).
Present address: SAIC-F, National Cancer Institute-Frederick/NIH, 1050 Boyles St,
Frederick, MD 21702-1201, USA.
à
Present address: China Novartis Institutes for BioMedical Research, Lane 898
The mechanism by which HM phosphorylation triggers activa-
tion relies on the docking of the phosphorylated HM to a particular
hydrophobic pocket in the N-terminal lobe of the kinase. Many
Halei Road, Shanghai, China.
§
Present address: Intellikine Inc., 10931 North Torrey Pines Road, Suite 103, La
Jolla, CA 92037, USA.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.05.019

3898
L. Wei et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3897–3902
AGC protein kinases contain this unique pocket which is formed by
the B and C helixes and the adjacent b-sheet formed by b4 and
LINKER
HM/PIF Pocket
a
a
b5. This HM binding pocket, referred to as the PIF (PDK1 Interact-
ing Fragment) pocket, was initially determined to be the binding
pocket by which PDK1 interacts with the phosphorylated HMs of
downstream substrates, such as PRK2,20 RSK2,21 S6K, and
SGK.22,23 In addition, the binding of HMs can activate PDK1 and en-
hance phosphorylation of downstream kinases.24 PDK1 harbors a
PIF pocket similar to other AGC kinases, but lacks a complementary
HM.25 The intermolecular docking of a phosphorylated C-terminal
hydrophobic motif (HM) of downstream kinases to the PIF pocket
of PDK1 allows for the specific and selective control of AGC family
protein kinases in response to extracellular events. Moreover, RSK,
S6K, PKB/Akt, MSK and SGK also contain a HM/PIF pocket in the N-
terminal lobe of the kinase, which are used to interact with their
own C-terminal phosphorylated HMs.26–29 The intramolecular
docking of the phosphorylated HMs to the PIF pocket results in
L
DISTAL
RING
N
O
TAIL
T
L
N
N
T
O
N
L
N
N
N
T
O
Docking Result to PDK1(1H1W)
Figure 2. Predicted binding mode of benzo/pyrimido (di)azepine-2-one scaffolds
the ordering of the
aB and aC helixes, which are disordered in
(PDK1 coordinates from PDB ID code 1H1W).
the unphosphorylated and monophosphorylated forms of the ki-
nase,30,31 and subsequently induces synergistic stimulation of the
kinase’s catalytic activity. The HM phosphorylation site acts in con-
cert with the activation loop phosphorylation site to stabilize the
active conformation. Sequence conservation suggests that this
mechanism is a key feature in the activation of more than 40 hu-
man AGC protein kinases.
matic groups and the PIF pocket and through the specific interac-
tion of the tail substitute with the residues that form the
phosphate-binding site which neighbors the PIF binding pocket.
After the in silico evaluation of these scaffolds, benzoazepin-2-
one, benzodiazepine-2-one, and pyrimidoazepin-2-one scaffolds
were chosen as the starting points for further chemical elaboration.
Based on their predicted binding modes (Fig. 2), SAR studies were
conducted on the tail region, the linker region and the distal ring
region.
The benzoazepin-2-one analogs were prepared by using com-
mercially available 3-bromo-benzoazepin-2-one as the starting
material (Fig. 3). Briefly, the treatment with sodium azide con-
verted the bromide to azide. The reaction of the resulting azide
and methyl bromoacetate afforded the N-alkylated intermediate.
The reduction of the azide to amine followed by reductive amina-
tion of the amine intermediate with aldehydes yielded the amine-
linked analogs. The corresponding amide- and sulfonamides-linked
analogs were also synthesized by coupling with carboxyl acid and
treatment with sulfonyl chlorides, respectively. Using a similar
procedure, more linker modifications were explored; thus, 3-bro-
mo benzoazepin-2-one was treated with phenol, thiophenol and
carboxylic acid potassium salt, followed by N-alkylation, affording
the corresponding ether-, thioether- and ester-linked analogs.
The synthesis of benzodiazepin-2-one analogs (Fig. 4) was
straightforward. Starting from commercially available 2-(2-nitro-
phenyl) acetonitrile, the cyano group was reduced to amine, fol-
lowed by the reductive amination with the corresponding
aldehydes. Then the nitro group was reduced to amine. The ring-
closure reaction was achieved by the treatment with triphosgene.
Then N-alkylation afforded the corresponding benzodiazepine
analogs.
It is possible that small molecules binding to the PIF pocket
would regulate both PDK1 catalytic activity as well as interaction
with substrates. Several small molecule allosteric activators of
PDK1 have been reported recently from virtual screening and
NMR-based fragment screening targeting the PIF pocket of PDK1
(Fig. 1).31–34 These compounds are structurally similar and all con-
tain a negatively charged carboxylate sandwiched by two aromatic
ring hydrophobic moieties. The negatively charged carboxylate
mimics the phosphate group of the phosphorylated HM and en-
ables the interaction with phosphate-binding site adjacent to the
PIF pocket; the two aromatic rings most likely mimic the two
phenylalanine residues on the HM of the substrate kinases. Crystal
structures have demonstrated that the identified small molecules
could modulate the PDK1 activity through binding to the PIF pock-
et and prompting the conformational change both in the crystal
and in solution. In an effort to identify selective small molecule
modulators of PDK1, we performed virtual docking on the PIF
pocket of PDK1 and a set of benzoazepin-2-one derivative scaffolds
were designed de novo (Fig. 2). Molecular docking was carried out
using GOLD [version 1.3 (Cambridge Crystallographic Data Center,
Cambridge, U.K.)]. PDK1 crystal structure (PDB ID code 1H1W) was
used as the protein template in docking. This molecular modeling
study showed that these new scaffolds could bind to the PIF pocket
mainly through hydrophobic interaction between the two aro-
The synthesis of the pyrimidoazepin-2-one scaffold began with
treating 5-bromo-2,4-dichloropyrimidine with glycine methyl es-
ter (Fig. 5). Stille coupling with allyltributylstannane, and oxidative
cleavage of the terminal alkene by ozonolysis followed by in situ
reductive amination afforded the key 4,5-di-substituted pyrimdine
intermediate. This intermediate was then cyclised by the treat-
ment with triphosgene and finally the amino groups were installed
by the Pd-catalyzed amination to provide the corresponding pyr-
imidodiazepine-2-one analogs.
Cl
Cl
O
S
HOOC
PS48
COOH
Compound A
The ability (EC50) of the de novo designed compounds to dis-
place the PIFtide peptide35 from the PIF pocket of PDK1 was then
measured. Briefly, an HTRF assay measuring the binding of bio-
tin-labeled PIFtide to the His-tagged PDK1 kinase domain protein
was established using europim labeled anti His-tag antibody and
allophycocyanin-streptavidin. Compound activity was measured
HOOC
Compound B
HOOC
Compound C
O
OH
Figure 1. Small molecule allosteric activators of PDK1.

L. Wei et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3897–3902
3899
N3
Br
N3
a
b
c
N
N
H
N
H
O
O
O
R
NHCH2R1
d
e
N
O
O
R
R
NH2
NHCOR1
NHSO2R1
N
O
N
N
R
f
O
R
OR1
SR1
OR1
SR1
g
h
i
N
N
O
O
O
O
H
R
b
Br
N
H
N
N
N
H
O
O
R
R
R1
R1
O
O
O
O
N
O
H
j
_COOMe
_COOH
R:
Figure 3. Synthesis of benzoazepin-2-one series: (a) NaN3/DMSO, 80 °C; (b) methyl bromoacetate, (Bu4N)+BrÀ, KOH/THF; (c) Pd/C, H2/EtOH; (d) RCHO, NaBH4/MeOH; (e)
RCOOH, HATU, DIEA/DMF; (f) RSO2Cl, DIEA/DCM; (g) ROH, NaH/dioxane; (h) RSH, NaH/dioxane; (i) RCOOK, 18-crown-6/MeCN; (j) LiOH/MeOH.
H
N
R1
NH2
CN
a
b
c
NO2
NO2
NO2
H
N
R1
R1
R1
N
d
e
N
NH2
N
H
N
O
O
R
f
_COOH
_COOMe
R:
Figure 4. Synthesis of benzodiazepine-2-one series: (a) BH3–THF (1.0 M); (b) RCHO, NaBH4/MeOH; (c) Pd/C, H2 /EtOH; (d) triphosgene/THF; (e) methyl bromoacetate, NaH/
THF; (f) LiOH/MeOH.
in a competition assay and the compound concentration that re-
sulted in 50% decrease of HTRF signal was designated as EC50. Com-
pound A (Fig. 1) was used as a positive control, as it is known to
The terminal carboxylate (tail region) was required for the bind-
ing (Table 2). This observation could be explained by electrostatic
interaction of the negative charge with Arg 131 of PDK1 (Fig. 6),
consistent with the fact that the activation of PDK1 by the HMs
of substrates is dependent on the phosphorylation of the HM or
the acidic residue mimicking the phosphorylated site.20,22
bind to the PIF pocket of PDK1 with an KD value of 18
by isothermal titration calorimetry (ITC).32 In our competition as-
say, the EC50 of this compound is 133 M.
lM measured
l
Based on binding assay results, it was found that the benzoaze-
pin-2-one analog 2 bound to the PIF pocket of PDK1, whereas the
benzodiazepine-2-one and pyrimidoazepin-2-one analogs showed
weak or no detectable binding affinity (Table 1).
The nature of the linker was also crucial for the binding. The thi-
oether-linker showed the best binding affinity of all the linkers we
explored though the amine linker also showed weak affinity
(Table 2, 10 vs 2). For the distal ring region (Table 3), a small

3900
L. Wei et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3897–3902
Br
Cl
Br
N
N
a
b
c
N
N
OMe
OMe
N
H
Cl
Cl
N
N
N
Cl
N
H
N
Cl
O
O
H
N
Ar
OMe
Ar
Ar
d
e
N
N
N
N
N
H
N
RHN
N
f
N
N
Cl
O
O
O
CH2COOMe
CH2R
R: _COOMe
_COOH
Figure 5. Synthesis of pyrimidodiazepine-2-one series: (a) MeOOCCH2NH2ÁHCl, DIEA/THF; (b) allyltributylstannane, Pd(dba)2/tri-2-furylphosphine; (c) O3, DCM then
benzylamine, NaBH4; (d) triphosgene, LHMDS/THF; (e) RNH2, Pd(OAc)2, Xantphos, dioxane, MW, 120 °C; (f) LiOH/MeOH.
Table 1
Table 2
Structures and binding affinities of compound 1–6
Structures and binding affinities of compound 1–2 and 7–16
a
Compd
Scaffolds
R
EC50 (lM)
L
1
2
–COOMe
–COOH
N/A
900
N
O
NH
R
N
O
R
a
Compd
L
R
EC50 (lM)
3
4
–COOMe
–COOH
N/A
N/A
1
2
–COOMe
–COOH
N/A
900
H
N
N
N
O
R
7
8
–COOMe
–COOH
N/A
N/A
O
S
9
10
–COOMe
–COOH
N/A
101
5
6
–COOMe
–COOH
N/A
N/A
N
11
12
–COOMe
–COOH
N/A
N/A
O
N
O
O
R
13
14
H
N
–COOMe
–COOH
N/A
N/A
a
Values are means of three experiments, (N/A = not active).
O
15
16
H
N
–COOMe
–COOH
N/A
N/A
neutral substitution at the para-position was favorable. Thus, 2-(3-
(4-chlorophenylthio)-2-oxo-2,3,4,5 tetrahydrobenzoazepin-1-yl)
acetic acid (Table 3, 17) was identified as the best binder with a
S
O
O
a
Values are means of three experiments, (N/A = not active).
EC50 value of 23
as compared to compound A, whose EC50 value was measured to
be 133 M in our assay as a reference.
lM, showing a fivefold increase in binding affinity
l
To gain a functional insight into these PIF pocket binders, we
tested compounds 17, 18 and 21 in in vitro PDK1 enzymatic assay
using a peptide substrate. As shown in Figure 7, these compounds
increased the PDK1 enzymatic activity in a concentration-depen-
by approximately fourfold at high concentration (>200 l
M).31,32
These results demonstrate that small molecules binding to PIF
pocket may induce slightly different conformational changes
which lead to differential activation of PDK1 activity.
dent fashion but to various degrees. At 25
l
M, compounds 17
In summary, a series of novel benzoazepin-2-ones were de-
signed to target the PIF pocket of PDK1 by de novo structure-based
design. The SARs of the tail region, unlinker region, and distal phe-
nyl ring region were explored and a sub-series of 2-thioether-ben-
zoazepin-2-ones were identified to bind to the PIF pocket of PDK1.
The designed compounds contain the common moieties identified
in the known PIF pocket binders, that is, two aromatic rings and
one carboxyl group. However, compared to known PIF pocket bind-
ers, the molecules presented here used a rigid cyclic benzoazepin-
2-one scaffold to orient the aromatic rings and the carboxyl group
that offers favorable binding entropy. This could explain the higher
binding affinity of compounds 17 and 18 compared to compound
A. In addition, the closely related molecules 18 and 21 that bind
to PIF pocket with similar affinity caused different levels of activa-
tion; the further understanding of the molecular mechanism will
require more structural studies including the co-crystallization of
and 18 increased the PDK1 activity by 14-fold and 10-fold, respec-
tively, whereas compound 21 showed less than twofold activation
of the enzyme under the same conditions (Fig. 7). It does not ap-
pear there is a strong correlation between the fold increase of
PDK1 activity and the EC50 value. Compound 18 binds to PIF pocket
with the same affinity as compound 21. Compound 18, however,
increased PDK1 activity more than fivefold higher than compound
21. The 14-fold increase of PDK1 activity by compound 17 is the
highest maximum activation reported by PDK1 PIF pocket activa-
tors. The highest concentration tested in the assay, 25 lM, was
around the EC50 value of 17 binding to PDK1. We expect that 17
will increase PDK1 activity furthermore and the activation curve
will reach a plateau at higher concentration. For comparison, the
natural PDK1 activator, PIFtide, increases PDK1 activity up to about
sevenfold.32 Compound A was reported to increase PDK1 activity

L. Wei et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3897–3902
3901
Figure 6. The proposed electrostatic interaction of the terminal carboxylate
(negative charge) with Arg 131 (PDK1 coordinates from PDB accession code 1H1W).
Figure 7. The PDK1 enzymatic assay. Compound 17 (circle), compound 18 (square)
and compound 21 (triangle). The catalytic activity of PDK1 without compounds was
set as 100%.
Table 3
Structures and binding affinities of compound 10 and 17–21
Acknowledgments
S
R
The authors thank Drs Charles Cho and Jianwei Che for their
helpful discussions.
N
O
COOH
a
References and notes
Compd
10
R
EC50
101
23
(lM)
1. Cantley, L. C. Science 2002, 296, 1655.
2. Engelman, J. A.; Luo, J.; Cantley, L. C. Nat. Rev. Genet. 2006, 7, 606.
3. Ming, X. F.; Burgering, B. M.; Wennstrom, S.; Claesson-Welsh, L.; Heldin, C. H.;
Bos, J. L.; Kozma, S. C.; Thomas, G. Nature 1994, 371, 426.
4. Franke, T. F.; Yang, S. I.; Chan, T. O.; Datta, K.; Kazlauskas, A.; Morrison, D. K.;
Kaplan, D. R.; Tsichlis, P. N. Cell 1995, 81, 727.
5. Kobayashi, T.; Cohen, P. Biochem. J. 1999, 339, 319.
6. Park, J.; Leong, M. L.; Buse, P.; Maiyar, A. C.; Firestone, G. L.; Hemmings, B. A.
EMBO J. 1999, 18, 3024.
7. Sturgill, T. W.; Ray, L. B.; Erikson, E.; Maller, J. L. Nature 1988, 334, 715.
8. Deak, M.; Clifton, A. D.; Lucocq, L. M.; Alessi, D. R. EMBO J. 1998, 17, 4426.
9. Flynn, P.; Mellor, H.; Casamassima, A.; Parker, P. J. J. Biol. Chem. 2000, 275,
11064.
10. Luo, J.; Manning, B. D.; Cantley, L. C. Cancer Cell 2003, 4, 257.
11. Taniguchi, C. M.; Emanuelli, B.; Kahn, C. R. Nat. Rev. Mol. Cell Biol. 2006, 7, 85.
12. Alessi, D. R.; James, S. R.; Downes, C. P.; Holmes, A. B.; Gaffney, P. R.; Reese, C.
B.; Cohen, P. Curr. Biol. 1997, 7, 261.
17
Cl
18
OMe
COOH
53
19
187
Cl
Cl
20
21
75
45
Cl
13. Stokoe, D.; Stephens, L. R.; Copeland, T.; Gaffney, P. R.; Reese, C. B.; Painter, G.
F.; Holmes, A. B.; McCormick, F.; Hawkins, P. T. Science 1997, 277, 567.
14. Alessi, D. R.; Kozlowski, M. T.; Weng, Q. P.; Morrice, N.; Avruch, J. Curr. Biol.
1998, 8, 69.
15. Pullen, N.; Dennis, P. B.; Andjelkovic, M.; Dufner, A.; Kozma, S. C.; Hemmings, B.
A.; Thomas, G. Science 1998, 279, 707.
Cl
22
A
Unlabeled PIFtide
10
133
16. Jensen, C. J.; Buch, M. B.; Krag, T. O.; Hemmings, B. A.; Gammeltoft, S.; Frodin,
M. J. Biol. Chem. 1999, 274, 27168.
a
Values are means of three experiments.
17. Richards, S. A.; Fu, J.; Romanelli, A.; Shimamura, A.; Blenis, J. Curr. Biol. 1999, 9,
810.
these activators to PDK1. One of the designed compounds, com-
pound 17, showed comparable binding affinity to the unlabled PIF-
tide, a peptide derived from the natural PDK1 PIF pocket binding
sequence.24,33,35 The designed molecules activate the PDK1 activity
in an in vitro enzymatic activity presumably by binding to the PIF
pocket and allosterically stabilizing the active conformation. Com-
pound 17 is the most potent activator of PDK1 reported so far in
terms of the maximum fold increase of the PDK1 activity. A potent
activator of PDK1 could be a valuable tool in studying the cellular
signaling pathways involving PDK1. Activation of most down-
stream substrates of PDK1 such as S6K, RSK and SGK requires the
binding of the substrates’ phosphorylated HM motif to the PIF
pocket of PDK1. The compounds identified here could potentially
inhibit the activation of the downstream PDK1 substrates in the
cell, and, therefore, they may have applications in inhibiting
essential signaling pathways mediated by PI3K and provides a
new strategy for drug discovery targeting this important pathway
in cancer.
18. Testa, J. R.; Bellacosa, A. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 10983.
19. Parekh, D. B.; Ziegler, W.; Parker, P. J. EMBO J. 2000, 19, 496.
20. Balendran, A.; Biondi, R. M.; Cheung, P. C.; Casamayor, A.; Deak, M.; Alessi, D. R.
J. Biol. Chem. 2000, 275, 20806.
21. Frodin, M.; Jensen, C. J.; Merienne, K.; Gammeltoft, S. EMBO J. 2000, 19, 2924.
22. Biondi, R. M.; Kieloch, A.; Currie, R. A.; Deak, M.; Alessi, D. R. EMBO J. 2001, 20,
4380.
23. Collins, B. J.; Deak, M.; Arthur, J. S.; Armit, L. J.; Alessi, D. R. EMBO J. 2003, 22,
4202.
24. Biondi, R. M.; Cheung, P. C.; Casamayor, A.; Deak, M.; Currie, R. A.; Alessi, D. R.
EMBO J. 2000, 19, 979.
25. Balendran, A.; Casamayor, A.; Deak, M.; Paterson, A.; Gaffney, P.; Currie, R.;
Downes, C. P.; Alessi, D. R. Curr. Biol. 1999, 9, 393.
26. Frodin, M.; Antal, T. L.; Dummler, B. A.; Jensen, C. J.; Deak, M.; Gammeltoft, S.;
Biondi, R. M. EMBO J. 2002, 21, 5396.
27. Yang, J.; Cron, P.; Thompson, V.; Good, V. M.; Hess, D.; Hemmings, B. A.;
Barford, D. Mol. Cell 2002, 9, 1227.
28. Huang, X.; Begley, M.; Morgenstern, K. A.; Gu, Y.; Rose, P.; Zhao, H.; Zhu, X.
Structure 2003, 11, 21.
29. Smith, K. J.; Carter, P. S.; Bridges, A.; Horrocks, P.; Lewis, C.; Pettman, G.; Clarke,
A.; Brown, M.; Hughes, J.; Wilkinson, M.; Bax, B.; Reith, A. Structure 2004, 12,
1067.

3902
L. Wei et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3897–3902
30. Yang, J.; Cron, P.; Good, V. M.; Thompson, V.; Hemmings, B. A.; Barford, D. Nat.
Struct. Biol. 2002, 9, 940.
31. Hindie, V.; Stroba, A.; Zhang, H.; Lopez-Garcia, L. A.; Idrissova, L.; Zeuzem, S.;
Hirschberg, D.; Schaeffer, F.; Jorgensen, T. J.; Engel, M.; Alzari, P. M.; Biondi, R.
M. Nat. Chem. Biol. 2009, 5, 758.
32. Engel, M.; Hindie, V.; Lopez-Garcia, L. A.; Stroba, A.; Schaeffer, F.; Adrian, I.;
Imig, J.; Idrissova, L.; Nastainczyk, W.; Zeuzem, S.; Alzari, P. M.; Hartmann, R.
W.; Piiper, A.; Biondi, R. M. EMBO J. 2006, 25, 5469.
33. Stockman, B. J.; Kothe, M.; Kohls, D.; Weibley, L.; Connolly, B. J.; Sheils, A. L.;
Cao, Q.; Cheng, A. C.; Yang, L.; Kamath, A. V.; Ding, Y. H.; Charlton, M. E. Chem.
Biol. Drug Des. 2009, 73, 179.
34. Liu, Y. Nature 2009, 461, 484.
35. Labeled PIF tide: Biotin-KTFCGTPEYLAPEVRREPRILSEEEQEMFRDFDYIADWK-
CONH2, was obtained from SynPep Corp. (Dublin, CA, USA). Unlabeled PIF
peptide: QEMFRDFDYIADW, was synthesized by us and the purity verified by
HPLC.