Identification of lead compounds as antagonists of protein Bcl-x L with a diversity-oriented multidisciplinary approach

Article abstract of DOI:10.1021/jm9010687

We report on the use of a diversity oriented synthesis (DOS) approach that resulted in the generation of a set of libraries of compounds presenting novel structural cores. These chemical cores have been employed to design potential antagonists of the anti

Full text of DOI:10.1021/jm9010687

7856 J. Med. Chem. 2009, 52, 7856–7867
DOI: 10.1021/jm9010687
Identification of Lead Compounds As Antagonists of Protein Bcl-xL with a Diversity-Oriented
Multidisciplinary Approach
Simone Di Micco,† Romina Vitale,§ Maurizio Pellecchia,‡ Michele F. Rega,‡ Renata Riva,§ Andrea Basso,*,§ and
Giuseppe Bifulco*,†
†Dipartimento di Scienze Farmaceutiche, Universitaꢀdegli studi di Salerno, Via Ponte don Melillo, 84084 Fisciano (SA), Italy, ‡Burnham Institute
for Medical Research, La Jolla, 92037 California, and §Dipartimento di Chimica e Chimica Industriale, Universitaꢀ degli Studi di Genova,
Via Dodecaneso 31, 16146 Genova, Italy
Received July 20, 2009
We report on the use of a diversity oriented synthesis (DOS) approach that resulted in the generation of a
set of libraries of compounds presenting novel structural cores. These chemical cores have been
employed to design potential antagonists of the antiapoptotic protein Bcl-xL through reiterated steps
of molecular docking calculations followed by experimental verification of binding. Our data suggest
that the DOS approach is suitable to generate novel scaffolds, which can be employed to target
protein-protein interactions.
Introduction
structurally different (therefore able to cover a higher area
of the chemical space) and characterized by high novelty
and complexity content (therefore potentially able to explore
new regions of the chemical space). To differentiate DOS from
total synthesis, the availability of chemical processes able to
generate high complexity in a limited number of synthetic steps
is very important. For these reasons, reactions able to create
new C-C bonds, rings, stereocenters, or functional groups,
possibly with fast kinetics, are highly favored;2 reactions pos-
sessing these requirements are, for example, cycloaddition,
multicomponent, tandem, and domino reactions, molecular
rearrangements.
We have recently started to apply this approach to a library
ofoxabicyclic derivatives. Asshownin Scheme1, aminoacid1
has been reacted with various aldehydes, isocyanides, and
alcohols according to a completely stereoselective intramole-
cular Ugi multicomponent reaction3 (U-5C-4CR) and a
combinatorial library of oxabicyclic peptidomimetics 2 has
been obtained, where decorative elements R1, R2, and R3 and
R4 can be varied at will. Compounds 2 are highly functiona-
lized substrates and can undergo a great number of transfor-
mations: for example, the oxabicyclic system can undergo a
retro-Diels-Alder reaction with loss of furan, leading to
sublibrary 3;4 if R1 and/or R4 possess an unsaturated double
or triple bond ring-opening/ring-closing metathesis (ROM/
RCM), processes can generate sublibraries 4, 5, and 6;5 if R2
has an additional amino group, the library of bicyclic lactams
7 can be obtained instead. All these processes usually just
require a single synthetic step, being therefore highly straight-
forward; in addition, they are independent and orthogonal
one to the other and can be performed in sequence, as
demonstrated by the generation of sublibrary 8 coupling a
lactam formation followed by a ROM/RCM step. In addi-
tion, the sublibraries obtained in this way can be further
elaborated, as demonstrated by the formation of library 9
from library 34 and library 10 from library 6.5 We have very
recently introduced an additional transformation of2, leading
to two classes of regioisomeric cyclohexenols 11 and 12,
Diversity oriented synthesis (DOSa)1 represents a new meth-
odological approach that in recent years is gaining more and
more importance in the field of organic synthesis and medicinal
chemistry. While traditional target oriented synthesis is direc-
ted toward obtaining specific compounds, the goal of diversity
oriented synthesis is the efficient preparation of collection of
substances characterized by high diversity content, in terms of
skeletons, appendages, and stereochemistry. One way to
achieve this is to employ pluripotent substrates (PSs), which
are molecules that can be synthetically elaborated, according to
their functional groups, in many different ways, independent
one to the other, each way leading to a different molecular
skeleton, in analogy with pluripotent stem cells.
Ideally, the PS is not a single entity but a collection of
molecules with a common skeleton, generated in a combina-
torial fashion, already displaying all or most of the final deco-
rative elements: this library of compounds is then further
elaborated to generate “n” different combinatorial libraries,
each with a different core structure. The ground-breaking
nature of this “star-burst” approach is that the final com-
pounds, although deriving from a common substrate, are
*To whom correspondence should be addressed. Phone: þ39-089-
969741 (G.B.); þ39-010-3536117 (A.B.). Fax: þ39-089-969602 (G.B.);
þ39-010-3536118 (A.B.). E-mail: bifulco@unisa.it (G.B.); andreab@
chimica.unige.it (A.B.).
a Abbreviations: DOS, diversity oriented synthesis; Bcl-2, B-cell
lymphoma 2; Bak, Bcl-2 antagonist/killer; Bcl-xL, B-cell lymphoma x
long; FPA, fluorescence polarization assay; PSs, pluripotent substrates;
ROM/RCM, ring-opening/ring-closing metathesis; BH, Bcl-2 homol-
ogy; NOEs, Nuclear Overhauser effects; rmsd, root-mean-square devia-
tion; MSMS, maximal speed molecular surface; OD, optical density;
HPLC, high-performance liquid chromatography; 1D 1H NMR, one-
dimensional proton nuclear magnetic resonance spectroscopy; IPTG,
Isopropyl β-D-1-tiogalattopiranoside; LB, lysogeny broth; TPPI, time
proportional phase incrementation; PFG, pulse field gradient; HEPES,
N-2-hydroxyethylpiperazine-N0-2-ethanesulfonic acid; LAH, lithium
aluminum hydride; MMFFs, Merck molecular force field; MCMM,
Monte Carlo multiple minimum; GB/SA, generalized Born/surface
area; PRCG, Polak-Ribier conjugate gradient; U-5C-4CR, Ugi five
centre four component reaction.
r
pubs.acs.org/jmc
Published on Web 10/23/2009
2009 American Chemical Society

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7857
Scheme 1. Diversity Oriented Synthesis (DOS) Applied to a Library of Oxabicyclic Pluripotent Substrates
involving a ring-opening process mediated by metal catalysts
(Scheme 1).
To probe this hypothesis, we have chosen the antiapoptotic
protein Bcl-xL, a member of the Bcl-2 family of proteins that
contributes to the equilibrium between cell proliferation and
cell death (apoptosis), as possible target. Apoptosis, or pro-
grammed cell death, is a highly controlled biological mechan-
ism regulating the removal of aged, damaged, and unnece-
ssary cells.7-11 The antiapoptotic proteins bind the pro-
apoptotic counterparts and sequester them from the cellular
environment inhibiting the apoptosis process. The analysis of
experimental three-dimensional structures of some antiapop-
totic proteins showed the presence of a hydrophobic surface
groove, formed by BH1 (Bcl-2 homology domain 1), BH2,
and BH3 domains and called BH3 binding groove. This
hydrophobic crevice constitutes the binding cavity for the
pro-apoptotic macromolecules.12-14 The up regulation of
antiapoptotic members of this family (Bcl-2, Bcl-xL) is ob-
served in many cancers. This overexpression protects the can-
cer cells from the activation of apoptosis, favoring their proli-
feration and their survival to the anticancer compounds.12-14
The design of small molecules, binding the BH3 domain of
antiapoptotic proteins and able to inhibit the protein-protein
interactions, can be a new strategy for the cancer therapy.15
Hence, we sought to use the newly derived DOS core
structures to design new potential antagonists of Bcl-xL.
Our strategy consists of iterative molecular docking studies
followed by synthetic chemistry and experimental verification
of binding by using in vitro assays.
Ring-opening of heterobicyclic systems mediated by metal
catalysts has been extensively explored by the group of
Lautens6 mainly on symmetrical compounds; the substrates
employedin this study, onthe otherhand, are notsymmetrical
and therefore can give two distinct regioisomeric products
depending on which of the two C-O bonds is broken during
the Sn20 process. In this way, our synthetic approach is
convergent, because it allows the assemblage of complex
entities in a few synthetic steps from simple substrates, and
at the same time divergent, because it allows the formation of
two distinct molecules in one operation.
Although for the sake of clarity only one enantiomer is
shown in Scheme 1, both enantiomerically pure forms of each
scaffold are independently accessible by employing the correct
bicyclic amino acid 1.
It is worth noting that most of the libraries obtained possess
an original core structure, and therefore their potential bio-
logical activity cannot be assigned by analogies with com-
pounds reported in the relevant literature but has to be
determined by the aid of virtual screening or by high through-
put screening experiments.
Our approach consisted of designing a number of virtual
libraries originated from compounds of general formula 2, 4,
6, 8, and 10-12 (Scheme 1) and analyzing by virtual screening
their ability to interact with a given biological target. The
novel nature of the resulting scaffolds suggest that such
librariesmay beemployedinthe searchofcompoundscapable
of antagonizing protein-protein interactions, for which “tra-
ditional” drug-like libraries usually fail to provide suitable
leads or even hit compounds.
Results and Discussion
Synthesis of 14a,c and 15a-c. As already anticipated in the
introduction, ring-opening of oxabicyclic systems can be

7858 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Di Micco et al.
Scheme 2. The Synthetic Pathway to Polyfunctionalized Cyclohexenols 14 and 15
Scheme 3. Synthesis of Second Generation Cyclohexenols 19 and 20
performed with opportune nucleophiles in the presence
of metal catalysts. Our major concern was related to the fact
that this kind of reaction is often reported in the literature on
simple substrates with a limited number of additional func-
tional groups that could interfere with the reaction outcome.
On the contrary, our Ugi-derived substrates possessed var-
ious functional groups, like amino, alcohol, or amide, which
could interact with the catalyst or the nuclephile.
during the U-5C-4CR, and in the case of compound 13b,
only regioisomer 15b was found in the reaction mixture.17 All
the described reactions were separately conducted on both
enantiomers of 1, obtained according to a previously re-
ported procedure,4 and each couple of regioisomers 14-15
was separated by the aid of flash chromatography. Struc-
tures were univocally assigned with the aid of NMR 2D
experiments, NOEs, and J coupling constants.
After an optimization process, substrates 2a-c were pre-
pared by condensation of enantiomerically pure 1 with
opportune aldehydes and isocyanides in methanol according
to the conditions previously described by us.3 Compounds
2a-c where then treated with LAH, furnishing pure 13a-c in
nearly quantitative yield and leaving the other functionalities
of 2 untouched. Alcohols 13a-c were then reacted, employ-
ing conditions similar to those described by Lautens,16 with
phenylboronic acid in the presence of Pd(C6H5CN)2Cl2, a
bis-phosphine ligand and K2CO3aq in methanol (Scheme 2).
Interestingly the regioisomeric ratio for compounds 14 and
15 varied depending on the decorating elements introduced
Synthesis of Compounds 19a,b, 20a,b, ent-19a,b and ent-
20a,b. In a first attempt, these compounds were prepared by
reaction of the bicyclic amino acid 1 (R1 = H) with for-
maldehyde and an opportune isocyanide, however, this
reaction failed to give the desired U-5C-4C adducts. As an
alternative route, less efficient but more reliable, we reacted
bicyclic amino ester 16 with bromoacetamides 17 and 18,
previously prepared according to known procedures,18,19
and the resulting compounds were subjected to reduction
and ring-opening according to the previously described
methodologies (Scheme 3); it is worth noting that, although
the final regioisomers 19a,b and 20a,b were isolated with

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7859
Table 1. List of Substituents and Functionalities for All Members of the Small Virtual Libraries (A-G) Obtained from DOS Scaffolds of Scheme 1
acceptable yields, the bridgehead opening process was less
smooth in the presence of a secondary amine, as already
reported in the literature. The same synthetic pathway was
followed to prepare the corresponding enantiomers.
Molecular Docking and NMR Studies. The DOS derived
scaffolds 2, 4, 6, 8, and 10-12 have been used to design seven
small libraries A-G, inserting different substituent: alipha-
tic/aromatic groups and functionalities containing nitrogen,
oxygen, or sulfur (reported in Table 1).
gave the best docking result and it was taken into account
for the substituent choice. Unfortunately, while ring-open-
ing reaction of substrates 2 with C nucleophiles proceeded
smoothly, all synthetic efforts to perform the same reaction
mediated by nitrogen or oxygen nucleophiles failed to give
the desired amino-substituted products in acceptable yields.
Thus, taking into consideration this synthetic limitation,
compounds 59-61 and 62-64 (Scheme 4) were designed
using scaffolds 11 and 12, respectively. Moreover, a pyridine
was introduced in the molecules with the aim to have a bulky
hydrophobic group containing a hydrogen bond acceptor
able to interact with a macromolecular counterpart. To
fully explore the possibility to establish hydrogen bonds
by pyridine nitrogen, the substituent position on this
heteroaromatic ring was changed, obtaining the isomers
59-61. The same considerations applied for 62-64. Com-
pounds ent-59-64 (Scheme 4), enantiomers of 59-64, were
The initial libraries were tested in silico for their ability to
bind to the BH3 binding groove of Bcl-xL.
The analysis of docking calculations and the comparison
of all tested compounds (21-58, Table 1) revealed that the
scaffolds 11 and 12 could be used to design promising
molecules. In particular, based on visual inspection and on
scoring function value, the structure 56 (KD = 3.14 ꢀ 10-11
M-1), containing an additional benzylamino substituent,

7860 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Scheme 4. Molecular Structures of 59-64 and ent-59-64
Di Micco et al.
pyridine and p-methoxyphenyl ring along the protein
groove. In particular, in our model, the pyridine ring seems
to be able to establish cation-π interactions with the Arg143
side chain, and the CH2OH group seems to form hydrogen
bonds with NH(CO) of Gly142 and with Arg143 side chain.
The docked conformations of 62-64 resulted in being super-
imposable (Figure 2a-c). In our poses, the p-methoxyphenyl
group and the benzyl ring faced the positively charged
side chain of Arg104. The 62 established a hydrogen
bond with NH of Gly142 by its CH2OH functionality,
whereas the 63 by the same group interacted with carboxylic
oxygen of Trp141. Their enantiomers (ent-62-64)
showed docking poses similar to 62-64. The ent-62-64
rotated about 180° the positions of the amide bond, benzyl
group, and the OH functionalities with respect to their
enantiomers (Figure 2d-f). Moreover, ent-62 and ent-63
established a hydrogen bond with CO in the side chain of
Asn140 by the CH2OH group. As found for 62-64, the
remaining aromatic ring pointed toward the solvent exposed
surface.
The above-described theoretical results traced the struc-
tural features responsible for potential binding affinity and
brought to the design and synthesis of compounds 14a,c and
15a-c (Scheme 2) and their enantiomers (ent-14a,c and ent-
15a-c).
The ability of 14a,c, 15a-c, ent-14a,c, and ent-15a-c to
bind Bcl-xL was tested by NMR-based binding assays. In
detail, 1D 1H NMR spectra of unlabeled protein were
recorded in the absence and presence of the putative ligands
and the potential binding processes were detected by obser-
ving the shifts of Bcl-xL resonances in the aliphatic region of
spectra: active site methyl groups of Ile, Leu, Thr, Val, or Ala
(region between -0.8 and 0.3 ppm). It was possible to work
with the unlabeled protein because the tested compounds did
not display chemical shift signals in this upfield region of the
spectrum and the protein signals were well resolved and
critical residues previously assigned.20
Upon titration of test ligands, unfortunately, the analysis
of 1D spectra did not reveal any significant shift of protein
resonances indicating absence or at least a very low binding
affinity. However, these compounds showed low solubility in
the solvent medium used for NMR experiments, and so
affecting the quality of spectra and the interpretation of
the experimental data. Taking into account the NMR
spectroscopy results and the solubility problems that could
affect their eventual employment in pharmacological studies
and eventually in therapy, 14a, 14c, 15a-c, ent-14a, ent-14c,
and ent-15a-c were modified by obtaining a new clus-
ter of the more hydrophilic analogues 19a,b and 20a,b
(Scheme 3).
The design of the new compounds resulted from the
analysis of the docking poses of parent compounds 59-64
and ent-59-64 (Figure 1 and 2). However, it also appeared
that the benzyl group and/or phenyl ring in the com-
pound series were not directly involved in interactions
with the protein, suggesting that those molecular groups
could be removed without affecting activity but presumably
resulting in compounds with increased solubility. The
modified analogues underwent docking calculations in
order to predict and rationalize their binding mode to the
protein.
also considered in the docking studies to verify the influ-
ence of the absolute configuration on the binding to the
protein.
The docking studies indicated that 59-64 and ent-59-64
presented a comparable predicted binding affinity for Bcl-xL
(see Supporting Information Table S1). They showed a
similar binding mode (overlaid docked structures are re-
ported in Supporting Information Figures S35-S38), occu-
pying equivalent spaces of the active site (Figures 1 and 2). In
particular, the small molecules spanned their structural
elements along the hydrophobic groove of the protein,
establishing van der Waals interactions with the macromo-
lecular counterparts, giving a significant contribution to the
complex stability. Moreover, for 59-64 and ent-59-64, the
analysis of the docked structures reveals, for all the pyridine
regioisomers, that the pyridine nitrogen is not involved in
any H-bond and consequently in the small collection of
derivatives (Scheme 2) the pyridine was substituted by a
phenyl ring.
The 3D spatial arrangements of 59-61 docked conforma-
tions were equivalent, giving similar interactions with the
protein (Figure 1a-c). In detail, the benzyl groups interacted
with the positively charged side chain of Arg104, and small
differences among the compounds regard the ability of
forming hydrogen bonds by means of the CH2OH substitu-
ent of cyclohexene. Their enantiomers (ent-59-61) presented
overlapped docking poses between them (Figure 1d-f).
They differed from 59-61 for the inverted position of
The theoretical studies (Figure 3) highlighted that all
compounds occupied equivalent active site portions, estab-
lishing hydrophobic contacts along the groove of Bcl-xL,

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7861
Figure 1. Three-dimensional models of the interactions formed by 59-61 (respectively, (a), (b), and (c)) and ent-59-61 (respectively, (d), (e),
and (f)) with Bcl-xL. The protein is represented by molecular surface and colored according to the hydrophobicity (dodger blue = hydrophilic,
orange red = hydrophobic, see Experimental Section). All ligands are depicted by sticks (black for 59 and ent-59, dodger blue for 60 and ent-60,
magenta for 61 and ent-61) and balls (colored by atom type: C, gray; polar H, white; N, dark-blue; O, red). The green lines indicate the hydrogen
bonds between ligand and protein. The figure highlights essential interactions: the compounds establish hydrophobic contacts along the groove.
especially by their aromatic rings, which may possibly
result in compounds with higher affinity for the protein.
In particular, the analysis of docking studies revealed the
potential ability of compounds 19a, ent-19b, 20a,b, and ent-
20a,b to interact with side chains of Asn140 and Arg143 by
their CH2OH and CO groups, respectively.
came from the overlapped methyl groups frequencies of
Leu94 and Leu134 (-0.12 ppm). The same effect, but lower,
was observed with the compounds ent-20a (Figure 4B). The
ent-19b (Figure 4C), followed by 20a (Figure 4A) and 19a
(Figure 4A), induced an appreciable shift of resonance of
Val139 (-0.79 ppm). This residue is very close to Arg143,
so then the shift of Val135 frequency is an indirect proof
of the interaction with Arg143, as also assessed by pre-
vious studies.21
These results could be in agreement with docking studies
that showed the possibility to form a hydrogen bond with
Arg143 and the proximity to the above aliphatic protein
residues (Figure 3). To confirm the data from 1D 1H NMR
spectra, 2D-NOESY experiments of the compounds were
performed in the absence and presence of protein. In the latter
Using the same 1D 1H NMR strategy, the binding proper-
ties of 19a,b-20a,b, and ent-19a,b-20a,b for Bcl-xL were
tested (Figure 4). The analysis of the experimental NMR
data showed a weak binding with Bcl-xL, monitoring the
chemical shift variation of aliphatic resonances of the biolo-
gical target. In particular, there was the appearance of a
signal at -0.06 ppm in the presence of 20b (Figure 4A), 19a
(Figure 4A), and ent-19a (Figure 4C), together with a slight
shift of the resonance at -0.12 ppm. This appeared signal

7862 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Di Micco et al.
Figure 2. Three-dimensional models of the interactions formed by 62-64 (respectively, (a), (b), and (c)) and ent-62-64 (respectively, (d), (e),
and (f)) with Bcl-xL. The protein is represented by molecular surface and colored according to the hydrophobicity (dodger blue = hydrophilic,
orange red = hydrophobic, see Experimental Section). All ligands are depicted by sticks (yellow for 62 and ent-60, green for 63 and ent-63,
turquoise for 64 and ent-64) and balls (colored by atom type: C, gray; polar H, white; N, dark blue; O, red). The green lines indicate the
hydrogen bonds between ligand and protein. The figure highlights essential interactions: the compounds establish hydrophobic contacts along
the groove.
case, intense cross-peaks of ligand were observable, indi-
cating that transiently the small molecules bound the protein;
due to the acquired macromolecular τc, positive cross peaks
appeared (see Supporting Information, Figures S19-S34). In
the absence of the macromolecule, the positive cross-
peaks disappeared due to the fast tumbling of the small
molecules.
BH3-peptide ligand (120 nM) and the relative small size of
our molecules. On the other hand, the results obtained
by means of combined NMR techniques and molecular
modeling trace a route in the elaboration of new compounds
endowed with increased activity.
Conclusions
With these results in hand, we also tested compounds 19a,
b and 20a,b in a fluorescence polarization assay in order to
check if the ability of these molecules to bind the nude
protein was accompanied also by their ability to displace
the natural ligand (a fluoresceinated Bak peptide). However,
these experiments did not show appreciable displacement at
200 μM concentration. This result was not completely
unexpected due to the strong affinity of the competitive
The strategy of diversity oriented synthesis (DOS) has
been used to obtain small libraries of compounds present-
ing original structural cores. A combined molecular dock-
ing-NMR spectroscopy iterative approach has led to the
identification of novel potential inhibitory scaffolds
against Bcl-xL. The synthetic schemes and the molecular
docking studies, supported by experimental verification by

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7863
Figure 3. Three-dimensional models of the interactions formed by 19a,b-20a,b (respectively, (a), (b), (c), and (d)) and ent-19a,b-20a,b
(respectively, (e), (f), (g), and (h)) with Bcl-xL. The protein is represented by molecular surface and colored according to the hydrophobicity
(dodger blue = hydrophilic, orange red = hydrophobic, see Experimental Section). All ligands are depicted by sticks (dark-slate-gray for 19a
and ent-19a, turquoise for 19b and ent-19b, yellow for 20a and ent-20a, green for 20b and ent-20b) and balls (colored by atom type: C, gray; polar
H, white; N, dark blue; O, red). The green lines indicate the hydrogen bonds between ligand and protein. The figure highlights essential
interactions: the compounds establish hydrophobic contacts along the groove.
NMR spectroscopy, provide a robust platform onto which
further optimized compounds can be obtained. The data
also provide some validation of our central hypothesis that
the DOS generated libraries could provide novel and
suitable starting scaffolds in targeting protein-protein
interactions.

7864 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Di Micco et al.
following procedure: the substrate (103 μmol), [Pd(C6H5CN)2]-
Cl2 (10.3 μmol) 1,3 bis-diphenylphosphinopropane (11.3 μmol)
were dissolved in MeOH (1.5 mL) together with water (25 μL)
and K2CO3 (103 μmol) under argon. The reaction is heated at
50 °C for 2-5 h, and then the solvents were evaporated under
vacuum and the crude taken up in DCM and washed with brine.
Chromatographic purification is performed using EtOAc/PE
mixtures.
(2R)-N-Benzyl-2-{N-benzyl-N-[(1R,4R,5S,6R)-5-hydroxy-6-
(hydroxymethyl)-4-phenyl-2-cyclohexene-1-yl]amino}-3-phenyl-
propanamide 14a. 1H NMR (300 MHz): δ 1.62 (1H, tt, J = 10.6,
3.2), 1.97 (1H, d, J = 7.4), 3.12 (1H, d, J = 3.1), 3.19-3.38 (3H,
m), 3.65-3.74 (3H,m), 3.94 (1H, d, J = 14.0), 4.12 (1H, d, J =
14.0), 4.07-4.15 (2H, m), 4.17 (1H, dd, J = 14.8, 5.6), 4.27 (1H,
dd, J = 14.8, 6.0), 5.42 (1H, broad s), 5.88 (1H, ddd, J = 10.1,
5.0, 2.4), 6.18 (1H, d, J = 10.2), 6.88-7.27 (20H, m). 13C NMR
(75 MHz): δ 33.27 (CH2), 42.20 (CH), 43.19 (CH2), 46.71 (CH),
52.23 (CH2), 55.43 (CH), 60.01 (CH2), 65.29 (CH), 69.12 (CH),
126-130 (12 ꢀ CH, ArH), 129.57 (CH, olefine), 130.85 (CH,
olefine), 137.71 (C, Ar), 138.43 (2 ꢀ C, Ar), 140.36 (C, Ar),
172.76 (C). HPLC: elution time 14.6 min (100% at 220 nm). For
all compounds, HPLC analyses were carried out on a HP 1100
instrument, using a gradient from H20/AcCN 9:1 to pure AcCN
in 15 min (flow 0.4 mL/min) with a Gemini C6Phenyl
column (150 mm ꢀ 3 mm, 3um), confirming a purity g95%.
HR-MS: expected for C36H38N2O3 547.2961; found 547.2958,
-0.5 ppm.
(2R)-N-Benzyl-2-{N-benzyl-N-[(1S,2R,5S,6R)-6-hydroxy-2-
(hydroxymethyl)-5-phenyl-3-cyclohexene-1-yl]amino}-3-phenyl-
propanamide 15a. 1H NMR: δ 2.77 (1H, broad s), 3.18 and 3.24
(2H, part AB of an ABX system Jax=3.4, Jbx=11.8, Jab = 11.8),
3.24-3.51 (3H, m), 3.66-3.75 (3H, m), 3.90 (1H, d, J=13.5),
4.18 (1H, dd, J = 14.8, 5.4), 4.31 (1H, dd, J=14.8, 4.3), 4.38
(1H, s), 4.57 (1H, d, J=13.5), 5.41 (1H, t, J=4.8), 5.67 (1H, d,
J = 10.5), 5.80 (1H, dt, J=10.5, 1.9), 6.90-7.42 (20H, m). 13
C
NMR: δ 34.82 (CH2), 37.45 (CH), 43.22 (CH2), 49.69 (CH),
52.25 (CH2), 59.10 (CH), 63.84 (CH), 64.62 (CH2), 71.35 (CH),
126-129. (12 ꢀ CH, ArH), 127.28 (CH, olefine), 131.45 (CH,
olefine), 137.88 (C, Ar), 138.71 (C, Ar), 140.45 (C, Ar), 140.75
(C, Ar), 172.90 (C). HPLC: elution time 14.4 min (98% at
220 nm). HR-MS: expected for C36H38N2O3 547.2961; found
547.3001; 7.3 ppm.
(2R)-N-Benzyl-2-{N-benzyl-N-[(1S,2R,5S,6R)-6-hydroxy-2-(hy-
droxymethyl)-5-phenyl-3-cyclohexene-1-yl]amino}-2-phenylace-
tamide 15b. 1H NMR: δ 2.81 (2H, broad s), 3.33 (1H, broad s),
3.70-3.90 (3H, m), 4.29-4.48 (5H, m), 4.83 (1H, t, J=6.3), 5.63
(1H, d, J=9.9), 5.81 (1H, t, J=5.4), 5.94 (1H, dt, J=9.9, 2.1),
6.84-7.47 (20H, m). 13C NMR: δ 37.35 (CH), 43.40 (CH2),
48.99 (CH), 52.82 (CH2), 58.50 (CH), 63.02 (CH2), 66.02 (CH),
71.74 (CH), 126-129 (12 ꢀ CH, ArH), 127.70 (CH, olefine),
133.03 (CH, olefine), 137.63 (C, Ar), 138.07 (C, Ar), 140.58
(C, Ar), 140.93 (C, Ar), 172.94 (C). HPLC: elution time 14.3 min
(100% at 220 nm). HR-MS: expected for C35H36N2O3
533.2804; found 533.2800; -0.8 ppm.
(2R)-N-(4-Methoxyphenyl)-2-{N-benzyl-N-[(1R,4R,5S,6R)-
5-hydroxy-6-(hydroxymethyl)-4-phenyl-2-cyclohexene-1-yl]
amino}-3-phenylpropanamide 14c. 1H NMR: δ 1.74 (2H, m),
2.88 (1H, broad s), 3.13 (1H, dd, J=11.9, 1.8), 3.43-3.56 (2H,
m), 3.66-3.85 (3H, m), 3.74 (3H, s), 3.90 (1H, d, J=8.5), 3.98
(1H, d, J=13.8), 4.02-4.12 (1H, m), 4.14 (1H, d, J=13.8), 5.89
(1H, ddd, J=10.3, 4.9, 2.4), 6.22 (1H, d, J=10.3), 6.77 (1H, d, J=
9.0), 7.18-7.38 (19H, m, ArH). 13C NMR: δ 33.37 (CH2), 42.34
(CH), 46.59 (CH), 52.38 (CH2), 55.41 (CH3), 56.22 (CH), 61.13
(CH2), 65.44 (CH), 69.72 (CH), 114.10 (CH, ArH), 121.58 (CH,
ArH), 126.37 (CH, ArH), 127.32 (CH, ArH), 127.39 (CH, ArH),
128-130 (6 ꢀ CH, ArH), 129.63 (CH, olefine), 130.30 (CH,
olefine), 130.50 (C, Ar), 137.94 (C, Ar), 138.77 (C, Ar), 139.96
(C, Ar), 156.41 (C, Ar), 171.08 (C). HPLC: 13.8 min (95% at 220
nm). HR-MS: expected for C36H38N2O4 563.2910; found
563.2899, -1.9 ppm.
Figure 4. Aliphatic region of the 1H 1D NMR spectrum of Bcl-xL
reported in absence (black spectrum in (A), (B), and (C)) and
presence of 19a (blue, A), 19b (red, B), ent-19a (red, C), ent-19b
(green, C), 20a (red, A), 20b (green, A), ent-20a (green, B), and ent-
20b (blue, B). The gray row indicates the appeared signal coming
from the overlapped methyl groups frequencies of L90 and L130
(-0.12 ppm, in the black spectrum). The gray dashed lines show the
shifts of labeled diagnostic residues of the alone protein (black
spectrum) in the presence of the ligands (red, green, and blue
spectra). The residues number refers to the structure 2O2M.
Experimental Section
Synthetic Procedures. Compounds 14a,c, 15a-c, ent-14a,c,
ent-15a-c, 19a,b, 20a,b, ent-19a,b, and ent-20a,b were prepared
from the corresponding bicyclic derivatives according to the

Article
(2R)-N-(4-Methoxyphenyl)-2-{N-benzyl-N-[(1S,2R,5S,6R)-6-
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7865
Molecular Docking. All ligands structures were built and their
geometries optimized through MacroModel 8.5 software22
package and using the MMFFs force field.23 MonteCarlo
Multiple Minimum (MCMM) method (10000 steps) of the
MacroModel package22 was used in order to allow a full
exploration of the conformational space. The so obtained
geometries were optimized using the Polak-Ribier conjugate
gradient algorithm (PRCG, 9 ꢀ 107 steps, maximum derivative
less than 0.001 kcal/mol). A GB/SA (generalized Born/surface
area)24 solvent treatment was used, mimicking the presence of
H2O, in the geometry optimization and in the conformational
search steps. The proteins for the docking calculations were
prepared using MacroModel software: all hydrogen were
added, bond order and missing atoms were checked by visual
inspection, and the charges of side chains were assigned con-
sidering their pKa.
hydroxy-2-(hydroxymethyl)-5-phenyl-3-cyclohexene-1-yl]ami-
no}-3-phenylpropanamide 15c. 1H NMR: δ 1.51 (1H, d, J = 2.3),
2.87 (1H, broad s), 3.21-3.50 (5H, m), 3.70 (1H, broad s), 3.74
(3H, s), 3.77-3.80 (1H, m), 3.94 (1H, d, J=13.7), 3.92-3.98
(1H, m), 4.42 (1H, broad s), 4.50 (1H, d, J=13.7), 5.66 (1H, d,
J = 10.2), 5.78 (1H, dt, J = 10.2, 1.9), 6.77 (2H, d, J = 9.0),
7.12-7.43 (17H, m, ArH). 13C NMR: δ 34.78 (CH2), 37.39
(CH), 49.68 (CH), 52.30 (CH2), 55.42 (CH3), 60.00 (CH), 64.05
(CH), 65.29 (CH2), 70.51 (CH), 114.02 (CH, ArH), 121.79 (CH,
ArH), 126.37 (CH, ArH), 126.93 (CH, olefine), 127.36 (CH,
ArH), 127.50 (CH, ArH), 128-129 (6 ꢀ CH, ArH), 130.54 (CH,
olefine), 131.14 (C, Ar), 139.00 (C, Ar), 139.98 (C, Ar), 140.41
(C, Ar), 156.37 (C, Ar), 171.27 (C). HPLC: 13.6 min (96% at 220
nm). HR-MS: expected for C36H38N2O4 563.2910; found
563.2892; -3.2 ppm.
Autodock 3.0.5 software25 was used for all docking calcula-
tions belonging. Before performing the virtual screening of the
libraries (Table 1), the validation of docking methodology was
N-Benzyl-2-{N-[(1R,4R,5S,6R)-5-hydroxy-6-(hydroxymethyl)-
1
4-phenyl-2-cyclohexene-1-yl]amino}acetamide 19a. H NMR: δ
1.83 (1H, m), 2.16 (3H, broad s), 3.16 (1H, d, J = 5.2), 3.39 (1H,
done by using as reference structure the NMR (2O2M.pdb)26
A
d, J=17.0), 3.46 (1H, d, J=17.0), 3.59 (1H, broad s), 3.72 and
3.76 (2H, part AB of an ABX system Jax=4.0, Jbx=4.0, Jab =
3D model of the complex formed by Bcl-xL and N-acylsulfona-
mide ligand was deposited in the Protein Data Bank (PDB).27,28
In detail, a blind docking was performed to verify the ability
of search method to detect the well-known binding cavity. In a
subsequent step, the conformational search parameters were
validated docking the N-acylsulfonamide ligand in the binding
site and evaluating the skill of the search method to reproduce
the experimental bioactive conformation of the ligand from the
NMR solution structure. For the blind docking calculations, a
9.7), 3.97 (1H, dd, J = 8.8, 5.4), 4.45 (2H, d, J=5.1), 5.78 (1H,
ddd, J=10.0, 4.0, 1.4), 5.93 (1H, d, J=10.0), 7.27-7.37 (10H,
m), 7.57 (1H, t). 13C NMR: δ 43.09 (CH), 43.16 (CH2), 45.96
(CH), 49.43 (CH2), 55.68 (CH), 62.55 (CH2), 70.10 (CH),
127-130 (6 ꢀ CH, ArH), 128.39 (CH, olefine), 128.99 (CH,
olefine), 138.14 (C, Ar), 138.21 (C, Ar), 171.59 (C). HPLC:
elution time 10.3 min (100% at 220 nm). HR-MS: expected for
C22H26N2O3 366.1943; found 366.1943; 0.0 ppm.
˚
N-Benzyl-2-{N-[(1S,2R,5S,6R)-6-hydroxy-2-(hydroxymethyl)-
grid box size of 256 ꢀ 256 ꢀ 256 with spacing of 0.375 A between
1
5-phenyl-3-cyclohexene-1-yl]amino}acetamide 20a. H NMR: δ
the grid points, centered on the macromolecule, covering all
protein surface, was used. Ten calculations consisting of 256
runs were performed, obtaining 2560 structures (256 ꢀ 10).
The protein, derived from the NMR solution structure 2O2M
(pdb archive code), was used as Bcl-xL model for the docking
calculations. For all the docked structures, all bonds were
treated as active torsional bonds except the amide bonds. A
2.43 (4H, broad s), 2.80 (1H, dd, J = 9.9, 1.7), 3.40 (2H, s), 3.56
(1H, broad s), 3.63 (1H, dd, J=10.6, 7.4), 3.87 (1H, dd, J = 10.6,
3.8), 3.99 (1H, broad s), 4.42 and 4.47 (2H, part AB of an ABX
system Jax = 5.0, Jbx = 5.0, Jab = 13.7), 5.60 (1H,dt, J = 10.2,
2.3), 5.70 (1H, d, J=10.2), 7.16-7.32 (11H, m). 13C NMR: δ
39.24 (CH), 43.37 (CH2), 47.24 (CH), 49.42 (CH2), 61.54 (CH),
66.30 (CH2), 67.71 (CH), 127-129 (6 ꢀ CH, ArH), 128.37 (CH,
olefine), 128.57 (CH, olefine), 138.10 (C, Ar), 140.68 (C, Ar),
171.63 (C). HPLC: elution time 9.6 min (100% at 220 nm).
HR-MS: expected for C22H26N2O3 366.1943; found 366.1911,
-8.7 ppm.
N-(4-Methoxyphenyl)-2-{N-[(1R,4R,5S,6R)-5-hydroxy-6-(hy-
droxymethyl)-4-phenyl-2-cyclohexene-1-yl]amino}acetamide 19b.
1H NMR: δ 1.96 (1H, m), 2.05 (3H, broad s), 3.23 (1H, d, J=5.0),
3.48 (1H, d, J = 16.9), 3.55 (1H, d, J = 16.9), 3.61 (1H, t,
J = 3.6), 3.78 (3H, s), 3.87 (2H, d, J = 5.7), 4.04 (1H, dd, J = 8.5,
5.4), 5.83 (1H, dd, J = 10.1, 3.9), 6.00 (1H, dt, J = 10.1, 1.8), 6.84
(2H, d, J= 9.0), 7.24-7.38 (5H, m), 7.47 (2H, d, J=9.0), 9.21(1H,
s). 13C NMR: δ 43.44 (CH), 45.85 (CH), 49.91 (CH2), 55.45 (CH
and CH3), 62.62 (CH2), 70.23 (CH), 114.13 (CH, ArH), 121.26
(CH, ArH), 127.49 (CH, olefine), 128.57 (CH, ArH), 128.62 (CH,
olefine), 128.72 (CH, ArH), 129.92 (CH, ArH), 130.78 (C, Ar),
138.14 (C, Ar), 156.30 (C, Ar), 169.45 (C). HPLC: elution time
10.2 min (100% at 220 nm). HR-MS: expected for C22H26N2O4
382.1893; found 382.1890; -0.7 ppm.
N-(4-Methoxyphenyl)-2-{N-[(1S,2R,5S,6R)-6-hydroxy-2-(hy-
droxymethyl)-5-phenyl-3-cyclohexene-1-yl]amino}acetamide 20b.
1H NMR (CD3OD): δ 2.30 (1H, broad s), 2.90 (1H, dd, J = 9.8,
1.0), 3.45 (1H, d, J = 17.3), 3.51 (1H, d, J = 17.3), 3.60 (1H,
broad s), 3.77 (3H, s), 3.82 (2H, dd, J = 10.8, 7.1), 3.95 (2H, dd,
J=10.8, 3.8), 4.10 (1H, broad s), 5.69 (1H, d, J=11.5), 5.74 (1H,
dt, J=11.5, 2.4), 6.84 (2H, d, J=9.0), 7.22-7.38 (5H, m), 7.52
(2H, d, J = 9.0). 13C NMR: δ 39.71 (CH), 47.57 (CH), 49.91
(CH2), 55.52 (CH3), 61.03 (CH), 66.15 (CH2), 67.80 (CH), 114.25
(CH, ArH), 121.31 (CH, ArH), 127.31 (CH, olefine), 128.39 (CH,
olefine), 128.55 (CH, ArH), 128.60 (CH, ArH), 128.90 (2 ꢀ CH,
ArH), 131.11 (C, Ar), 140.47 (C, Ar), 156.45 (C, Ar), 169.55 (C).
HPLC: elution time 9.5 min (100% at 220 nm). HR-MS: expected
for C22H26N2O4 382.1893; found 382.1887; -1.5 ppm.
˚
grid box size of 64 ꢀ 64 ꢀ 64 with spacing of 0.375 A between the
grid points was used, presenting a grid center with the following
x, y, and z coordinates respectively: 3.675, 5.538, and 2.0. In the
theoretical studies, we considered the amine functionality of the
ligands as protonated at physiological pH and, consequently,
positively charged in the calculations.
To achieve a representative conformational space during the
docking studies and taking into account the variable number of
active torsions, from 8 to 10 calculations consisting of 256 runs
were performed, obtaining 2048/2560 structures. The Lamar-
kian genetic algorithm was employed for dockings. An initial
population of 600 randomly placed individuals. The maximum
number of energy evaluations and of generations was set up
on the number of active torsions presented by the different
compounds. The first parameter ranged from 4 ꢀ 106 to 6 ꢀ 106
and the maximum number of generations from 5 ꢀ 106 to 7 ꢀ
107. A mutation rate of 0.02 and a crossover rate of 0.8 were
used, and the local search frequency was set up at 0.26. Results
˚
differing by less than 2 A in positional root-mean-square devia-
tion (rmsd) were clustered together and represented by the result
with the most favorable free energy of binding.
All the 3D models were depicted using the UCSF Chimera
package:29 molecular surfaces were rendered using Maximal
Speed Molecular Surface (MSMS).30 The 3D models of the
protein are represented by hydrophobicity surface showing the
amino acid property in the Kyte-Doolittle scale.31 The hydro-
phobicity surface colors range from dodger blue for the most
hydrophilic amino acid to orange-red for the most hydropho-
bic one, with the white representing the 0.0 value of the
Kyte-Doolittle scale.31
NMR Experiments. NMR experiments were performed on a
Bruker Avance 700 spectrometer fitted with a cryoprobe at T =
1
300 K. 1D H experiments were run with 500 μL of solution

7866 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Di Micco et al.
(deuterated buffer at pH = 7.5) of Bcl-xL at 20 μM, in the
absence and presence of added compounds, everyone at 250 μM
concentration. In the final concentrations of all NMR solutions,
a 2% of DMSO-d6 99.95% was used. The same conditions were
employed for 2D-NOESY experiments but with concentrations
of Bcl-xL and ligands of 10 μM and 1 mM, respectively.
The NOESY32 spectra in D2O were acquired with a mixing
time of 500 ms, a number of 16 scans/t1, and a t1max value of
15.24 ms. All 2D spectra were acquired in the phase-sensitive
mode, and the TPPI method33 was used for quadrature detec-
tion in the ω1 dimension. In all experiments, a watergate PFG
sequence was used for water suppression.
(6) Lautens, M.; Fagnou, K.; Hiebert, S. Transition metal-catalyzed
enantioselective ring-opening reactions of oxabicyclic alkenes. Acc.
Chem. Res. 2003, 36, 48–58.
(7) Adams, J. M.; Cory, S. The Bcl-2 protein family: arbiters of cell
survival. Science 1998, 281, 1322–1326.
(8) Reed, J. C. Bcl-2 family proteins: strategies for overcoming che-
moresistance in cancer. Adv. Pharmacol. 1997, 41, 501–553.
(9) Chao, D. T.; Korsmeyer, S. J. Bcl-2 family: regulators of cell death.
Annu. Rev. Immunol. 1998, 16, 395–419.
(10) Minn, A. J.; Swain, R. E.; Ma, A.; Thompson, C. B. Recent
progress on the regulation of apoptosis by bcl-2 family members.
Adv. Immunol. 1998, 70, 245–279.
(11) Storey, S. Targeting apoptosis: selected anticancer strategies. Nat.
Rev. Drug Discovery 2008, 7, 971–972.
(12) Sattler, M.; Liang, H.; Nettesheim, D.; Meadows, R. P.; Harlan, J.
E.; Eberstadt, M.; Yoon, H. S.; Shuker, S. B.; Chang, B. S.; Minn,
A. J.; Thompson, C. B.; Fesik, S. W. Structure of Bcl-xL-Bak
peptide complex: recognition between regulators of apoptosis.
Science 1997, 275, 983–986.
(13) Petros, A. M.; Nettesheim, D. G.; Wang, Y.; Olejniczak, E. T.;
Meadows, R. P.; Mack, J.; Swift, K.; Matayoshi, E. D.; Zhang, H.;
Thompson, C. B.; Fesik, S. W. Rationale for Bcl-xL/Bad peptide
complex formation from structure, mutagenesis, and biophysical
studies. Protein Sci. 2000, 9, 2528–2534.
(14) Day, C. L.; Chen, L.; Richardson, S. J.; Harrison, P. J.; Huang, D.
C.; Hinds, M. G. Solution structure of prosurvival Mcl-1 and
characterization of its binding by proapoptotic BH3-only ligands.
J. Biol. Chem. 2005, 280, 4738–4744.
(15) Becattini, B.; Kitada, S.; Leone, M.; Monosov, E.; Chandler,
S.; Zhai, D.; Kipps, T. J.; Reed, J. C.; Pellecchia, M. Rational
design and real time, in-cell detection of the proapoptotic activity
of a novel compound targeting Bcl-xL. Chem. Biol. 2004, 11, 389–
395.
(16) Lautens, M.; Dockendorff, C. Palladium(II) catalyst systems for
the addition of boronic acids to bicyclic alkenes: new scope and
reactivity. Org. Lett. 2003, 5, 3695–3698.
(17) Detailed studies on the ring-opening reaction of oxabicyclic pepti-
domimetics will be the subject matter of a forthcoming paper.
(18) Ronsisvalle, G.; Marrazzo, A.; Prezzavento, O.; Pasquinucci, L.;
Vittorio, F.; Pittala, V.; Pappalardo, M. S.; Cacciaguerra, S.;
Spampinato, S. (þ)-cis-N-Ethyleneamino-N-normetazocine deri-
vatives. Novel and selective sigma ligands with antagonist proper-
ties. J. Med. Chem. 1998, 41, 1574–1580.
The unlabeled Bcl-xL samples were prepared and purified as
reported in literature.12 Briefly, Escherichia coli strain BL21 was
transformed with the pET-21b plasmid (Novagen) carrying the
gene coding for Bcl-xL ΔTM (Bcl-xL deletion mutant lacking the
transmembrane domain). Bacteria were grown on LB supported
media. Induction of protein expression was carried out at OD600
= 0.8 with 1 mM IPTG for 4 h at 37 °C. Following cell harvest
and lysis by sonication, the protein was purified using a Ni-
affinity column (Amersham). The eluate was extensively dia-
lyzed against 40 mM phosphate buffer (pH = 7.5).
Fluorescence Anisotropy. The formation of heterodimers was
measured in terms of increased anisotropy (Polarion, Tecan
Co.) using fluoresceinated Bak peptide as natural ligand. All the
experiments were performed in polar buffer (HEPES pH 7, 100
mM; KCl, 250 mM; MgCl2, 15 mM; DTT, 5 mM; EDTA,
5 MM); the experiment volume was always 200 μL per well, and
the plates used were 96-well black nonbonding plates (Corning).
Protein and ligands were mixed at room temperature and
maintained at target temperature (30 °C) in the dark for the
whole duration of the experiments. Anisotropy measurements
were done at different times of incubation to assess the stability
of the reactions. Two sets of experiments were carried out with
the following concentrations: experiment-1 Bcl-xL 200 nM,
Fluo-Bak 20 nM, inhibitor 2 μM; experiment-2 Bcl-xL 100
nM, Fluo-Bak 10 nM, inhibitor 20 μM.
(19) Harte, A. J.; Gunnlaugsson, T. Synthesis of alpha-chloroamides in
water. Tetrahedron Lett. 2006, 47, 6321–6324.
(20) Petros, A. M; Fesik, S. W; Olejniczak, E. T. 1H, 13C and 15N
resonance assignments of a Bcl-xL/Bad peptide complex. J. Bio-
mol. NMR 2005, 32, 260.
Acknowledgment. We thank MIUR (PRIN contract num-
ber 2006031151_005), and the EMBO for the financial sup-
port of SDM.
(21) Rega, M. F.; Leone, M.; Jung, D.; Cotton, N. J.; Stebbins, J. L.;
Pellecchia, M. Structure-based discovery of a new class of Bcl-xL
antagonists. Bioorg. Chem. 2007, 35, 344–353.
Supporting Information Available: 1H- and 13C NMR spectra
for compounds 14a,c, 15a-c, 19a,b, and 20a,b. 2D-NOESY
spectra of 19a,b, 20a,b, ent-19a,b, and ent-20a,b in presence and
absence of Bcl-xL acquired at 300 K (pH 7.5) in D2O solution;
calculated inhibition constant of the complex formed by 19a,
b-20a,b, ent-19a,b-20a,b, 59-64, and ent-59-64 with Bcl-xL;
calculated fitness scores of the complex formed by 19a,b-20a,b,
and ent-19a,b-20a,b with Bcl-xL; superimposition of docked
poses of 59-64 and ent-59-64. This material is available free of
charge via the Internet at http://pubs.acs.org.
€
(22) MacroModel, version 8.5; Schrodinger LLC: New York, NY, 2003.
(23) Halgren, T. A. J. Merck molecular force field. V. Exten-
sion of MMFF94 using experimental data, additional
computational data, and empirical rules. J. Comput. Chem. 1996,
17, 616–641.
(24) Still, W. C.; Tempczyk, A.; Hawley, R. C; Hendrickson, T. The
GB/SA continuum model for solvation. A fast analytical method
for the calculation of approximate born radii. J. Am. Chem. Soc.
1990, 112, 6127–6129.
(25) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W.
E.; Belew, R. K.; Olson, A. J. Automated docking using a
Lamarckiangenetic algorithm and an empirical binding free energy
function. J. Comput. Chem. 1998, 19, 1639–1662.
References
(26) Brukno, 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. Studies leading to
potent, dual inhibitors of Bcl-2 and Bcl-xL. J. Med. Chem. 2007, 50,
641–662.
(1) Schreiber, S. L. Target-oriented and diversity-oriented organic
synthesis in drug discovery. Science 2000, 287, 1964–1969.
(2) Burke, M. D.; Berger, E. M.; Schreiber, S. L. Generating diverse
skeletons of small molecules combinatorially. Science 2003, 302,
613–618.
(3) Basso, A.; Banfi, L.; Riva, R.; Guanti, G. U-4C-3CR versus U-5C-
4CR and stereochemical outcomes using suitable bicyclic beta-
amino acid derivatives as bifunctional components in the Ugi
reaction. Tetrahedron Lett. 2004, 45, 587–590.
(4) Basso, A.; Banfi, L.; Riva, R.; Guanti, G. A novel highly selective
chiral auxiliary for the asymmetric synthesis of L- and D-alpha-
amino acid derivatives via a multicomponent Ugi reaction. J. Org.
Chem. 2005, 70, 575–579.
(5) Basso, A.; Banfi, L.; Riva, R.; Guanti, G. Preparation of opti-
cally pure fused polycyclic scaffolds by Ugi reaction followed
by olefin and enyne metathesis. Tetrahedron 2006, 62, 8830–
8837.
(27) Protein Data Base www.pdb.org.
(28) Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T. N.;
Weissig, H.; Shindyalov, I. N.; Bourne, P. E. The Protein Data
Bank. Nucleic Acids Res. 2000, 28, 235–242.
(29) Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.;
Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. UCSF Chimera;a
visualization system for exploratory research and analysis. J.
Comput. Chem. 2004, 25, 1605–1612.
(30) Sanner, M. F.; Olson, A. J.; Spehner, J. C. Reduced surface: an
efficient way to compute molecular surfaces. Biopolymers 1996, 38,
305–320.

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7867
€
(31) Kyte, J.; Doolittle, R. F. A simple method for displaying the
hydropathic character of a protein. J. Mol. Biol. 1982, 157, 105–132.
(32) Macura, S.; Ernst, R. R. Elucidation of cross relaxation in liquids by
two-dimensional NMR spectroscopy. Mol. Phys. 1980, 41, 95–117.
(33) Marion, D.; Wuthrich, K. Application of phase sensitive two-
dimensional correlated spectroscopy (COSY) for measurements
of 1H-1H spin-spin coupling constants in proteins. Biochem.
Biophys. Res. Commun. 1983, 113, 967–974.