The discovery of small molecule chemical probes of Bcl-XL and Mcl-1

Article abstract of DOI:10.1016/j.bmc.2008.06.023

A tetrahydroaminoquinoline-based library was generated with the goals of finding small molecule modulators of protein-protein interactions. Several library members as well as other related intermediates were tested for their ability to bind to Bcl-X_L and Mcl-1 by in silico and ¹⁵N NMR studies. The NMR study led to the identification of the tetrahydroaminoquinoline-based nude scaffold, 7 as a weak binder (K_d = 200 μM for Bcl-X_L and K_d = 300 μM for Mcl-1) to both proteins. Using this scaffold as the starting material, we then synthesized a focused library of only 9 derivatives by applying the principles of a fragment-based approach. All these derivatives were then tested by NMR and this led to the discovery of a novel, small molecule (MIPRALDEN, 17) as a binder to Mcl-1 and Bcl-X_L (K_D = 25 and 70 μM). This finding is novel because to our knowledge there are not many small molecules known in the literature that bind to Mcl-1. Crown Copyright

Full text of DOI:10.1016/j.bmc.2008.06.023

Bioorganic & Medicinal Chemistry 16 (2008) 7443–7449
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
The discovery of small molecule chemical probes of Bcl-XL and Mcl-1
a,b,
Michael Prakesch a, Alexey Yu Denisov c, Marwen Naim d, Kalle Gehring c, , Prabhat Arya
*
*
a Ontario Institute for Cancer Research, MaRS Centre, South Tower, 101 College Street, Toronto, Ont., Canada M5G 1L7
b Steacie Institute for Molecular Sciences, National Research Council of Canada, 100 Sussex Drive, Ottawa, Ont., Canada K1A 0R6
c Department of Biochemistry, McGill University, 3655 Promenade Sir William Osler, Montréal, Que., Canada H3G 1Y6
d Computational Chemistry & Biology Group, Biotechnology Research Institute, National Research Council of Canada, 6100 Royalmount Avenue, Montréal, Que., Canada H4P 2R2
a r t i c l e i n f o
a b s t r a c t
Article history:
A tetrahydroaminoquinoline-based library was generated with the goals of finding small molecule mod-
ulators of protein–protein interactions. Several library members as well as other related intermediates
were tested for their ability to bind to Bcl-XL and Mcl-1 by in silico and 15N NMR studies. The NMR study
led to the identification of the tetrahydroaminoquinoline-based nude scaffold, 7 as a weak binder
Received 7 April 2008
Revised 31 May 2008
Accepted 6 June 2008
Available online 18 June 2008
(Kd = 200 lM for Bcl-XL and Kd = 300 lM for Mcl-1) to both proteins. Using this scaffold as the starting
material, we then synthesized a focused library of only 9 derivatives by applying the principles of a frag-
ment-based approach. All these derivatives were then tested by NMR and this led to the discovery of a
novel, small molecule (MIPRALDEN, 17) as a binder to Mcl-1 and Bcl-XL (KD = 25 and 70 lM). This finding
is novel because to our knowledge there are not many small molecules known in the literature that bind
Keywords:
Natural product-inspired probes
Mcl-1
Bcl-2 family
to Mcl-1.
Protein–protein interactions
Small molecule–protein interactions by
NMR
Crown Copyright Ó 2008 Published by Elsevier Ltd. All rights reserved.
Apoptosis
Chemical biology
High-throughput synthesis
Combinatorial chemistry
Diversity-oriented synthesis
1. Introduction
In cases where the structural information of the protein is known
and if the protein could be subjected to a low-throughput screening
The programmed cell death (apoptosis) machinery follows sev-
eral well-organized cell signaling networks where one of the two
major pathways (i.e., the mitochondrial process or intrinsic path-
way) involves a highly regulated series of proteins.1,2 Three sub-
families of the protein known as the Bcl-2 family interact together
and decide whether the cell should die or survive. One group is
prone to inducing apoptosis (e.g., Bax, Bak, and Bok), the second
one is also pro-apoptotic and is called ‘BH3-only pro-apoptotic’
(Bad, Bik, Bim, Bid, Hrk, Bcl-rambo, Bmf, Bcl-G, PUMA, and Noxa),
and the third group is known to be pro-survival (e.g., Bcl-2, Bcl-
XL, Bcl-w, Bcl-B, Mcl-1, and A1).3–5 It was shown that cancer cells
over express Bcl-2 family proteins around mitochondria and con-
tribute to tumor initiation, progression and resistance to ther-
apy.6,7 Thus, triggering apoptosis and inducing cancer cells to die
could be one therapeutic way to fight cancers and an especially
attractive one if apoptosis could be induced using small molecules.
study by NMR, the use of the fragment-based approach to identify
small molecule ligands is gaining momentum.8–11 An excellent exam-
ple of this approach is the work from Abbott Laboratories on the dis-
covery of ABT-737 as a small molecule binder to Bcl-XL. ABT-737 is a
nanomolar (nM) binder to Bcl-XL and the synthesis of this small mol-
ecule was achieved using the structural information of Bak:Bcl-XL
protein–protein (p–p) interaction.6 Although ABT-737 is a strong
and selective binder to Bcl-XL, it was shown not to target Mcl-1 pro-
tein, thereby conferring resistance to the drug. Further, it was shown
that the down-regulation of Mcl-1, when ABT-737 is used, removed
its lethality effect.12,13 Hence, finding a small molecule targeting
Mcl-1 only, or even both Bcl-XL and Mcl-1, could serve as a highly
desirable chemical probe to investigate its biological functions.
2. Results and discussion
2.1. Library generation of tetrahydroquinoline derivatives with
the goals of targeting protein–protein interactions
* Corresponding authors. Tel.: +1 613 993 7014; fax: +1 613 952 0068.
E-mail addresses: kalle.gehring@mcgill.ca (K. Gehring), prabhat.arya@nrc.ca
(P. Arya).
Targeting protein–protein (p–p) interactions by small
molecules remains
a
challenging undertaking since these
0968-0896/$ - see front matter Crown Copyright Ó 2008 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.06.023

7444
M. Prakesch et al. / Bioorg. Med. Chem. 16 (2008) 7443–7449
interactions cover a relatively large surface and involve multi-
ple hot-spots that may have extensive hydrophobic surfaces.14
With the goals of finding small molecule probes of p–p inter-
actions,15 we are developing high-throughput approaches lead-
ing to the library generation of tetrahydroquinoline alkaloid-
inspired compounds. With this objective, the synthesis of the
tetrahydroaminoquinoline scaffold, 1, was achieved as shown
in Scheme 1.16 To develop the solid-phase synthesis method,
compound 2 was obtained following several easy transforma-
tions. This included the introduction of a three-carbon spacer
and changing the N-Teoc to the N-Fmoc protecting group
giving derivative 2. These transformations allowed the com-
pound to be compatible with the solid-phase synthesis that
utilizes silylation as the mode of immobilization. The loading
of compound 2 was performed using the Broad Institute load-
ing-protocol and it was achieved in high yields (76% after
cleavage from the support). When carried out on a large scale,
the loss of the N-Fmoc protecting group was observed, yielding
3 as the free amino alcohol derivative. The loaded compound 3
2.2. In silico screening of the generic library
In search of small molecule binders to Bcl-2 proteins
a
family, the library members were then tested by virtual
screening against two anti-apoptotic proteins: Bcl-XL Protein
(1YSI and 1YSN) and Bcl-2 (1YSW). The screening studies re-
sulted in the identification of several compounds from the li-
brary showing good scores (see Fig. 1). The in silico identified
library members showing good scores were then re-synthe-
sized in solution to be tested further in NMR-binding experi-
ments with Bcl-XL protein. Due to their poor solubility, all
the leads identified by in silico were not compatible with
the NMR studies.
2.3. A search for the small molecule scaffold by NMR screening
Shown in Scheme 2 are two small molecule derivatives, 7 and 8,
that were obtained from 1 and 2 with the goal of validating our in
silico studies. To enhance their solubility in an aqueous media, the
protecting groups around the scaffold were removed to increase
the number of hydrogen-bonding donors. Compounds 7 and 8
were then subjected to 15N NMR studies for their ability to interact
with Bcl-XL.
Interestingly, scaffold 8 was found to be a poor binder to
the hydrophobic cleft of Bcl-XL with a KD of 10 mM. While
7 was still a weak binder, it had a better interaction with
could then be used in the generation of
a 105-membered
library by selectively introducing the first diversity using
a
DIC coupling with acids, R1CO2H. Under these conditions, there
was no sign of the amide coupling using the free secondary
amine. This group appears to be involved in an intra-molecular
hydrogen bonding with the carboxyl ester group (see Support-
ing material for more details). Following this, the second diver-
sity was then introduced using acid chloride (R2COCl) coupling
to give compound 4. After the N-Alloc removal and the cou-
pling of the third diversity member, using R3COCl, cleavage
of the all the library members gave 105 discrete compounds
with the general structure 6, in good yields. No purification
was required since the HPLC of each library member showed
an average purity greater than 90%.
the protein: KD = 200
lM. The perturbed residues were local-
ized in the helices -2, -5, and -7 with the most strongly af-
a
fected amino acid residues being 96–102, which is a typical
site for the earlier known small molecule drugs (ABT-737
and Gossypol). Further, NMR studies combined with computa-
tional experiments also confirmed the location of
7 in the
upper section of the Bcl-XL hydrophobic cleft. Scaffold 7 was
OH
Fmoc
N
CO2Et
OH
Teoc
N
CO2Et
OH
O
O
CO2Et
OH
(i) pTsOH, 57%
(ii) TsO(CH2)3OTHP,
CsCO3, 77%
H
N
alkylsilyl macrobeads,
trifluoromethanesulfonate,
O
MEMO
(iii) TBAF, 84%
2,6-lutidine, 76%
(iv) FmocCl, NaHCO3, 88%
(v) PPTS, 86%
NH
NH
NH
Alloc
Alloc
Alloc
O
2
1
3
O
R2
O
R2
R2
O
O
O
(i) Pd(PPh3)4, PPh3,
4-NMM, CH3CO2H
OH
(i) R1CO2H, DMAP,
DIC
N
N
HF-py
N
COOEt
OCOR1
COOEt
OCOR1
COOEt
OCOR1
(ii) R3COCl,
2,4,6-collidine
(ii) R2COCl,
O
O
2,4,6-collidine
AllocHN
R3OCHN
R3OCHN
5
6
4
105 membered library
CO2H
R1
COCl
R2
O
O
CO2H
CO2H
MeO
O
CO2H
O
COCl
CO2H
O
N
COCl
COCl
COCl
COCl
R3
N
COCl
COCl
MeO
COCl
COCl
Scheme 1. The library generation from the tetrahydroaminoquinoline scaffold.

M. Prakesch et al. / Bioorg. Med. Chem. 16 (2008) 7443–7449
7445
Figure 1. In silico Studies with Bcl-XL and Bcl-2 using the tetrahydroaminoquinoline library. (A–C) The library members.
3D minimized structure of 7
also tested with Mcl-1 protein and it showed similar interac-
tions with the BH3-peptide-binding site formed by helixes
-3, -4, and -5 with a KD of 300 M. The in silico docking of
compound
was then performed using FredÒ with both
proteins, Bcl-XL and Mcl-1, and 100 best poses are shown in
Figure 2A and B. The best pose is represented in Figure 2C
and D, where compound 7 is interacting at the same site of
the original Bak-peptide.
a-2,
Teoc
N
l
CO2H
OH
LiOH
7
1
MEMO
NH
Alloc
7
Although identified as a weak binder in our NMR experiments,
these findings were very attractive and offered an excellent oppor-
tunity to derivatize this ‘nude’ scaffold further using a fragment-
based type approach. At this stage, it was decided to synthesize
only a small set of compounds comprising the functionalization
of both amines of the nude scaffold, 7. Due to the presence of the
hydroxyl and carboxylic acid groups in each derivative, it was
anticipated that the solubility in an aqueous media would not be
the limiting factor.
3D minimized structure of 8
OH
O
CO2Et
OH
H
N
N
-Fmoc
2
removal
NH
Alloc
8
Scheme 2. Two derivatives of the tetrahydroaminoquinoline scaffold.

7446
M. Prakesch et al. / Bioorg. Med. Chem. 16 (2008) 7443–7449
Figure 2. In silico docking of 7 with Bcl-XL and Mcl-1.
O
O
O
R1:
R2:
Si
Si
Si
O
O
O
14 a: 100
11 a: 100
12 a: 300
13 a: 200
9 a: ∼4000
b: n.a.
10 a: ∼1000
b: 100
b: 300
b: n.a.
b: 200
b: 500
R1
O
R1:
N
CO2H
OH
R2:
MEMO
O
NH
R2
16 a: 1000
15 a: >5000
17 a: 70
3D minimized
structure of 17
b: 300
b: 25
b: n.a.
a: Binding with Bcl-XL protein in μM; b: Binding with Mcl-1 protein in μM; n.a.: not available
Figure 3. A focused, nine-membered library by solution-phase synthesis.
2.4. The use of a fragment-based approach in designing focused
nine-membered library
amines because it has been shown in the past to be an excellent
fragment when it comes to protein binding.17
Shown in Figure 3 are nine compounds that were then synthe-
sized in solution. To explore the potential of the chiral scaffold 7 to
expand the distal hydrophobic sites, the plan was to systematically
introduce several hydrophobic groups (i.e., Ph, Bn and 4-biphenyl)
at both amines via an amide bond. In particular, we were very
interested in exploring the use of the 4-biphenyl groups at both
2.5. NMR-binding experiments with Bcl-XL and Mcl-1
Compounds 9–17 were tested for their ability to interact with
Bcl-XL and Mcl-1 proteins and the results are shown in Figure 3.
To date, as we expected, our best result with Mcl-1 is the MIPRAL-
DEN (compound 17), which binds with a KD = 25 lM. Shown in yel-

M. Prakesch et al. / Bioorg. Med. Chem. 16 (2008) 7443–7449
7447
Figure 4. Interactions of 17 with Bcl-XL and Mcl-1 by NMR. (A and B) Proximity of residues (in yellow) of Bcl-XL and Mcl-1 with 17 (MIPRALDEN). (C and D) 15N–1H HSQC for
Bcl-XL (C) and Mcl-1. (D) Without (black) and with (red) 17. (E) Magnitude of the amide chemical shift changes [(
binding. The positions of helices as well as BH domains in Bcl-XL are shown. (F) Magnitude of the amide chemical shift changes [(
1 upon 17 binding. The positions of helices as well as BH domains in Mcl-1 are shown.
D
1H shift)2 + (
D
15N shift ꢀ 0.2)2]1/2 in Bcl-XL upon 17
a
D
1H shift)2 + (
D
15N shift ꢀ 0.2)2]1/2 in Mcl-
a
low in Figure 4A and B are the amino acid residues from Bcl-XL and
Mcl-1 proteins that were affected by 17 (see also Figure 4C and D
for 15N–1H HSQC). Magnitudes of amide chemical shift changes in
15N–1H HSQC spectra are also shown in Figure 4E and F. The in sil-
ico docking with compound 17 using FredÒ was further studied
with Bcl-XL and Mcl-1 proteins and the best poses are shown in
Figure 5. This is an interesting finding because although several
small molecule binders to Bcl-XL appeared in the literature in re-
cent years18,19,18,20–23 but to our knowledge there are not many
examples of their effective binding with Mcl-1.23,13,20,24,25 The dis-
covery of 17 also opens up tremendous opportunities for improv-
ing the binding, leading to the next generation of analogs.
Furthermore, work is also ongoing at several fronts to test the
scope of 17 in various biological systems and these findings will
be reported as they become available.
3.2. In silico procedure and tools
In the present method, an in-house made scoring function, ‘Sol-
vated Interaction Energy’ SIE, was coupled with the commercial
docking program suit (Omega-Fred from OpenEye Inc., Santa Fe,
CA, USA) to produce a virtual screening platform. The scoring func-
tion combines molecular mechanics force-fields with a continuum
treatment of solvation. It was carefully parameterized using a cu-
rated experimental-binding set that consists of diverse protein–
small molecule or protein–peptide complexes with known binding
affinities and X-ray structures.26 Moreover, SIE-scoring function
was subjected to two validation tests. First, its ability to discrimi-
nate the native-binding mode or a close conformation from an
ensemble of 100 decoy poses was determined. Second, its ability
to identify known binders embedded in a random chemical library
in a virtual screening context was tested. Overall, SIE scoring func-
tion performed extremely well in both tests.
3. Experimental procedures
3.1. Synthesis
3.3. Protein structure preparation (Bcl-XL, Bcl-2 and Mcl-1)
All the information on the experimental procedures, character-
ization data, and 1H, 13C NMR spectra is provided as the supple-
ment data (see Supporting information).
In order to explore in silico compound binding onto Bcl-2 pro-
tein family, 2 X-ray complexes of Bcl-XL and Bcl-2 and solution
structure of Mcl-1 were considered, that is, (i) BclXL (pdb entry

7448
M. Prakesch et al. / Bioorg. Med. Chem. 16 (2008) 7443–7449
0.4
0.35
0.3
E
α1
BH4
α2
α3
α4
α5
BH1
α6
BH2
α7
BH3
0.25
0.2
0.15
0.1
0.05
0
0
10
20
30
40
90 100 110 120 130 140 150 160 170 180 190
Residue number
0.45
F
0.4
0.35
0.3
α3
α1
α2
α4
α5
α6
α7
BH3
BH1
BH2
0.25
0.2
0.15
0.1
0.05
0
150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300
Residue number
Figure 4. (continued)
Figure 5. In silico docking of 17 with Bcl-XL and Mcl-1.
1YSI and 1YSN),6 (ii) Bcl-2 (pdb entry 1YSW),6 and (iii) Mcl-1 (pdb
entry 1WSX).27 Blocking groups are added to C- and N-termini
dine residues, any other ionizable residue side-chains were sys-
tematically charged (Lysine, Aspartate and Glutamate). For the
Bcl-XL and Bcl-2 structures, the small molecule inhibitor was re-
moved from the structure to allow docking program to explore
the whole protein including the active site. Each protein was care-
ends of both proteins, NH3 and COOꢁ, respectively. Hydrogen
þ
atoms are also added for whole proteins. As binding assay occurs
at neutral pH, histidine side-chains are kept neutral. Besides histi-

M. Prakesch et al. / Bioorg. Med. Chem. 16 (2008) 7443–7449
7449
fully prepared for force-fields and scoring function compatibility,
that is, partial atomic charges, atom-typing, bond-stretching, an-
gle-bending and torsion angles. Only hydrogen atoms are submit-
ted to a short minimization to optimize their orientation and
remove VdW clashes.
Edna Matta-Camacho for preparations of Bcl-XL and Mcl-1
proteins, Mark Hinds (WEHI, Australia) for NMR assignments of
Mcl-1, and Gordon Shore (GeminX, Canada) for Mcl-1 plasmid
preparation.
Supplementary data
3.4. Small molecules preparation
Experimental details and full characterization data for all new
compounds are provided. This material is available free of charge.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2008.06.023.
The ligands were built in Isis Base and converted from 2D to 3D
using concord program within sybyl (Tripos Inc.). The structures
were then used to generate GAFF atom types using the antecham-
ber module of AMBER.28 This facility automatically generates
parameters that are compatible with the AMBER force-field (atom
types, bond stretch, and angle bend, torsional and improper tor-
sional parameters). The automatically assigned GAFF atom types
were then manually checked and corrected as necessary. Partial
charges for the ligands were calculated using the AM1-BCC method
of assigning partial charges.29
References and notes
1. Fesik, S. W. Nat. Rev. Cancer 2005, 5, 876.
2. Fischer, U.; Schulze-Osthoff, K. Cell Death Differ. 2005, 12, 942.
3. Walensky, L. D. Cell Death Differ. 2006, 13, 1339.
4. Letai, A.; Scorrano, L. Cell Death Differ. 2006, 13, 1245.
5. Letai, A. Trends Mol. Med. 2005, 11, 442.
6. Oltersdorf, T.; Elmore, S. W.; Shoemaker, A. R.; Armstrong, R. C.; Augeri, D. J.;
Belli, B. A.; Bruncko, M.; Deckwerth, T. L.; Dinges, J.; Hajduk, P. J.; Joseph, M. K.;
Kitada, S.; Korsmeyer, S. J.; Kunzer, A. R.; Letai, A.; Li, C.; Mitten, M. J.;
Nettesheim, D. G.; Ng, S.; Nimmer, P. M.; O’Connor, J. M.; Oleksijew, A.; Petros,
A. M.; Reed, J. C.; Shen, W.; Tahir, S. K.; Thompson, C. B.; Tomaselli, K. J.; Wang,
B. L.; Wendt, M. D.; Zhang, H. C.; Fesik, S. W.; Rosenberg, S. H. Nature 2005, 435,
677.
7. Reed, J. C. Nat. Rev. Drug Disc. 2002, 1, 111.
8. Rees, D. C.; Congreve, M.; Murray, C. W.; Carr, R. Nat. Rev. 2004, 3, 660.
9. Erlanson, D. A. Curr. Opin. Biotechnol. 2006, 17, 643.
10. Hajduk, P. J.; Greer, J. Nat. Rev. Drug Disc. 2007, 6, 211.
11. Hajduk, P. J. Nat. Chem. Biol. 2006, 2, 658.
12. Tahir, S. K.; Yang, X.; Anderson, M. G.; Morgan-Lappe, S. E.; Sarthy, A. V.; Chen,
J.; Warner, R. B.; Ng, S. C.; Fesik, S. W.; Elmore, S. W.; Rosenberg, S. H.; Tse, C.
Cancer Res. 2007, 67, 1176.
3.5. Docking and scoring
All the members of the library were used in a virtual screening
on two anti-apoptotic proteins: Bcl-XL Protein protein (1YSI and
1YSN) and Bcl-2 (1YSW). We were pleased to find several com-
pounds from the library showing good scores. For Bcl-XL (1YSI),
the two compounds A and B as well as for Bcl-XL (1YSN), the com-
pound C was found as lead compounds (Fig. 1). Concerning the pro-
tein Bcl-2 (1YSW) the same compound C was found as a lead
compound but with a lower-binding affinity (Fig. 2).
13. Chen, S.; Dai, Y.; Harada, H.; Dent, P.; Grant, S. Cancer Res. 2007, 67, 782.
14. Arkin, M. R.; Wells, J. A. Nat. Rev. Drug Disc. 2004, 3, 301.
15. Fry, D. C. Biopolymers 2006, 84, 535.
16. Prakesch, M.; Sharma, U.; Sharma, M.; Khadem, S.; Leek, D. M.; Arya, P. J. Comb.
Chem. 2006, 8, 715.
3.6. NMR studies: protein sample preparation and NMR
spectroscopy
17. Hajduk, P. J.; Bures, M.; Praestgaard, J.; Fesik, S. W. J. Med. Chem. 2000, 43, 3443.
18. Bruncko, M.; Oost, T. K.; Belli, B. A.; Ding, H.; Joseph, M. K.; Kunzer, A.;
Martineau, D.; McClellan, W. J.; Mitten, M.; Ng, S. C.; Nimmer, P. M.; Oltersdorf,
T.; Park, C. M.; Petros, A. M.; Shoemaker, A. R.; Song, X.; Wang, X.; Wendt, M.
D.; Zhang, H.; Fesik, S. W.; Rosenberg, S. H.; Elmore, S. W. J. Med. Chem. 2007,
50, 641.
19. Lee, E. F.; Czabotar, P. E.; Smith, B. J.; Deshayes, K.; Zobel, K.; Colman, P. M.;
Fairlie, W. D. Cell Death Differ. 2007, 14, 1711.
20. Nguyen, M.; Marcellus, R. C.; Roulston, A.; Watson, M.; Serfass, L.; Murthy
Madiraju, S. R.; Goulet, D.; Viallet, J.; Belec, L.; Billot, X.; Acoca, S.; Purisima, E.;
Wiegmans, A.; Cluse, L.; Johnstone, R. W.; Beauparlant, P.; Shore, G. C. Proc.
Natl. Acad. Sci. U.S.A. 2007, 104, 19512.
21. Wang, L.; Kong, F.; Kokoski, C. L.; Andrews, D. W.; Xing, C. Bioorg. Med. Chem.
Lett. 2008, 18, 236.
22. Wang, G.; Nikolovska-Coleska, Z.; Yang, C. Y.; Wang, R.; Tang, G.; Guo, J.;
Shangary, S.; Qiu, S.; Gao, W.; Yang, D.; Meagher, J.; Stuckey, J.; Krajewski, K.;
Jiang, S.; Roller, P. P.; Abaan, H. O.; Tomita, Y.; Wang, S. J. Med. Chem. 2006, 49,
6139.
23. Manion, M. K.; Fry, J.; Schwartz, P. S.; Hockenbery, D. M. Curr. Opin. Investig.
Drugs 2006, 7, 1077.
24. Czabotar, P. E.; Lee, E. F.; van Delft, M. F.; Day, C. L.; Smith, B. J.; Huang, D. C.;
Fairlie, W. D.; Hinds, M. G.; Colman, P. M. Proc. Natl. Acad. Sci. U.S.A. 2007, 104,
6217.
Mouse Bcl-XL containing deletion in the C-terminus (
D197–
233) and the internal loop ( 45–84) and mouse Mcl-1 (152–308
D
30
amino acids) were prepared as described earlier for Bcl-XL and
Mcl-1.27 The pET-29b+ plasmid for Bcl-XL and pGEX-6P-1 for
Mcl-1 were used, and proteins were expressed in Escherichia coli
BL21 cells. For NMR studies, cultures were grown in M9 media sup-
plemented with 15N ammonium chloride to produce uniformly
15N-labeled proteins. Soluble Bcl-XL protein was purified by Ni2+
-
affinity chromatography, and GST-Mcl-1 protein was purified by
affinity chromatography using glutathione–Sepharose 4B and
cleaved with PreScission Protease. NMR samples contained
0.1 mM protein in 80% H2O/10% DMSO-d6/10% D2O, 20 mM sodium
phosphate (pH 6.8), 5 mM EDTA, and 3 mM DTT.
NMR spectra were recorded on Bruker DRX 600 MHz spectrom-
eter equipped with triple-resonance cryoprobe. 15N–1H HSQC spec-
tra were recorded at 1:2, 1:1, 3:1, and 10:1 drug to protein ratios
and temperature 35 °C for Bcl-XL or 25 °C for Mcl-1. Values of the
amide chemical shift changes were calculated as [(D
1H
25. van Delft, M. F.; Wei, A. H.; Mason, K. D.; Vandenberg, C. J.; Chen, L.; Czabotar,
P. E.; Willis, S. N.; Scott, C. L.; Day, C. L.; Cory, S.; Adams, J. M.; Roberts, A. W.;
Huang, D. C. Cancer Cell 2006, 10, 389.
26. Naim, M.; Bhat, S.; Rankin, K. N.; Dennis, S.; Chowdhury, S. F.; Siddiqi, I.; Drabik,
P.; Sulea, T.; Bayly, C. I.; Jakalian, A.; Purisima, E. O. J. Chem. Inf. Model. 2007, 47,
122.
shift)2 + (
D
15N shift ꢀ 0.2)2]1/2 in ppm. Dissociation constants
(KD) were determined from the changes of chemical shifts versus
drug concentration with precision ꢂ50% of KD value.
27. Day, C. L.; Chen, L.; Richardson, S. J.; Harrison, P. J.; Huang, D. C. S.; Hinds, M. G.
J. Biol. Chem. 2005, 280, 4738.
Acknowledgments
28. Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D. A. J. Comput. Chem.
2004, 25, 1157.
29. Jakalian, A.; Jack, D. B.; Bayly, C. I. J. Comput. Chem. 2002, 23, 1623.
30. Denisov, A. Y.; Madiraju, M. S.; Chen, G.; Khadir, A.; Beauparlant, P.; Attardo, G.;
Shore, G. C.; Gehring, K. J. Biol. Chem. 2003, 278, 21124.
The DOS group at the Broad Institute is thanked for providing
the alkylsilyl linker-based macrobeads. This work was supported
by the NRC Genomics and Health Initiative, Canadian Cancer Soci-
ety (CCS), National Cancer Institute of Canada (NCIC), and Canadian
Institutes of Health Research (CIHR) Grant MOP-82177. We thank