Design and modular parallel synthesis of a MCR derived α-helix mimetic protein-protein interaction inhibitor scaffold

Article abstract of DOI:10.1016/j.bmcl.2005.11.102

A terphenyl α-helix mimetic scaffold recognized to be capable of disrupting protein-protein interactions was structurally morphed into an easily amenable and versatile multicomponent reaction (MCR) backbone. The design, modular in-parallel library synthes

Full text of DOI:10.1016/j.bmcl.2005.11.102

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1740–1743
Design and modular parallel synthesis of a MCR derived
a-helix mimetic protein–protein interaction inhibitor scaffold
Walfrido Antuch,b Sanjay Menon,c Quin-Zene Chen,c Yingchun Lu,c Sukumar Sakamuri,c
b
b
c
ˇ
´
Barbara Beck, Vesna Schauer-Vukasinovic, Seema Agarwal,
Sibylle Hessb and Alexander Do¨mlinga,*
aABC Pharma, Franckensteinstr. 9a, 81243 Mu¨nchen, Germany
bMorphochem AG; Gmunderstrasse-37-37a, 81379 Mu¨nchen, Germany
cMorphochem Inc., 11 Deer Park Drive, Suite 116, Monmouth Junction, NJ, USA
Received 24 October 2005; revised 30 November 2005; accepted 30 November 2005
Available online 19 January 2006
Abstract—A terphenyl a-helix mimetic scaffold recognized to be capable of disrupting protein–protein interactions was structurally
morphed into an easily amenable and versatile multicomponent reaction (MCR) backbone. The design, modular in-parallel library
synthesis, initial cell based biological data, and preliminary in vitro screening for the disruption of the Bcl-w/Bak protein–protein
interaction by representatives of the MCR derived scaffold are presented.
ꢀ 2006 Published by Elsevier Ltd.
Protein–protein interactions (PPI) play an important
role in the regulation of a number of critical biological
functions. The formations of antibody–antigen, hor-
mone-receptor or certain enzyme-inhibitor complexes
represent a few examples of such PPI. The nature of
these protein–protein interface interactions is complex.
They are usually formed by discontinuous protein-
sequence and protein-surface elements and the contact
area is normally rather large. Attempts to classify the
different types of protein–protein interfaces have result-
ed in a rather larger number of families.1 Nevertheless,
PPI constitute an important target class for novel drug
discovery projects, and a growing number of small
molecule inhibitors of these interactions continue to be
identified.2
chemotherapeutic agents, utilizing a chemical genomics
type approach.4
In contrast to protein secondary structure elements such
as b-sheets or b-turns, small molecule a-helix mimetics
have rarely been described in the literature (Fig. 1).
1,1,6-Trisubstituted indanes suggested to operate as
mimetics of the i ꢀ 1, i, and i + 1 sidechain residues of
an a-helix have been shown to be NK1 binders with
micromolar affinity.5 2,6,30,50-Tetrasubstituted biphe-
nyls have been proposed to mimic the side chains of
the i, i + 1, i + 3, and i + 4 residues of an a-helix6 and
recently, Hamilton and co-workers have reported on
3,20,200-trisubstituted terphenyls as mimetics of the side
chains at the i, i + 3, and i + 7 positions of an a-helix,
capable of disrupting PPI.7
The major secondary structure elements in key apopto-
sis regulating PPI (e.g., Bcl-2 family proteins, p53/
HDM2) are a-helices.3 Since the disruption of these
PPI could constitute an effective and selective means
for the treatment of cancer, we have been interested in
the design and discovery of a-helix mimetic PPI inhibi-
tors as potentially novel, apoptosis promoting cancer
Our efforts into the design of novel a-helix mimetics that
disrupt PPI and induce cellular apoptosis were prompt-
ed by Hamilton’s terphenyl a-helix mimetic concept.
One major disadvantage inherent in the terphenyl scaf-
fold is the rather lengthy, multi-step synthetic approach
needed for its assembly.8 Hence, as a starting point, we
investigated alternate a-helix mimetic scaffolds that
would be accessible by a multicomponent reaction
(MCR), ideally in a single reaction step and in a modu-
lar fashion. Molecular modeling9 derived structural con-
siderations led us to the selection of an appropriately
Keywords: Protein interaction antagonist; Cancer; Bcl family;
Apoptosis; Multicomponent reaction; Isocyanide.
*
Corresponding author. E-mail: alex.doemling@abc-pharma.de
0960-894X/$ - see front matter ꢀ 2006 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2005.11.102

W. Antuch et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1740–1743
1741
Phenyl-substituted tosylmethyl isocyanides (TosMICs)
constitute one of the versatile starting materials in our
synthesis of a MCR derived imidazole based a-helix
mimetic. They are easy to synthesize or are commercial-
ly available.11 During the course of this work, we pre-
pared several substituted phenyl TosMICs by reacting
the corresponding substituted benzaldehyde, formam-
ide, and para-methyl phenyl sulfinic acid in a MCR (a-
aminoalkylation) followed by dehydration (Fig. 3).12
The chosen substituents were selected to effectively mim-
ic one of the side-chain residues of the amino acids con-
stituting a part of the triad i, i + 3, and i + 7.
Ri
Ri+3
Ri+7
For the synthesis of the desired trisubstituted imidaz-
oles, we initially investigated the use of anilines
(R3 = substituted phenyl) as the amine component in
the van Leusen synthesis (Fig. 4).
R i
Under a variety of experimental conditions and for a
small set of aldehydes investigated, meta-substituted
anilines failed to react with ortho-substituted phenyl
TosMICs to provide the desired imidazole products A
(Fig. 2). Alternatively, when meta-substituted phenyl
TosMICs were employed in the same reaction, the cor-
responding imidazoles were obtained, albeit in moderate
yields only. Significantly, key inversely substituted imi-
dazoles B could be generated starting from ortho-substi-
tuted anilines and meta-substituted phenyl TosMICs,
although in low yields.
R i+1
R i+3
R i+4
Due to the difficulty encountered with suitably reacting
ortho-substituted phenyl TosMICs with anilines, we next
turned our attention to the use of appropriately substi-
tuted benzylamines (R3 = substituted phenylmethyl) as
the amine component in the van Leusen imidazole
synthesis (Fig. 4). Gratifyingly, we observed that benzyl-
amines typically reacted much more robustly than the
corresponding anilines in the van Leusen reaction. A
Ri
Ri+1
O
NH
SO2H
+
O
O
O
O
Figure 1. Rationally designed a-helix mimetic scaffolds, superimposed
on an ideal a-helix. The relative location in the sequence of the a-helix
residues addressed by the small molecules has been indexed.
S
NHCHO
S
NC
POCl3
TEA
+
CHO
R
R
R
NH2CHO
trisubstituted imidazole backbone (Fig. 2), which could
be easily obtained utilizing the synthetic approach of
van Leusen et al.10
O
S
O
O
NO2
O
Cl
Cl
Ri
Ri
O
O
Ri
O
Cl
O
N
N
Ri+3
n
Figure 3. Preparation of phenyl-substituted TosMICs.
N
Ri+3
Ri+3
n
Ri+7
N
Ri+7
R2
R1
Ri+7
base
R1
R3
CHO
NH2
R2
N
R3
+
+
A
B
-TosSO2H
S
NC
N
O
O
Figure 2. Structural morphing of the terphenyl scaffold into a van
Leusen imidazole backbone (n = 0–2).
Figure 4. The van Leusen MCR synthesis of trisubstituted imidazoles.

1742
W. Antuch et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1740–1743
parallel solution phase synthesis of a focused library of
about 60 trisubstituted imidazoles was conducted. Ra-
pid parallel purification of the crude reaction products
was accomplished via Ôchromfiltration’13 to afford the
desired trisubstituted imidazole products in >80%
purity.
Based on our primary interest in initially identifying
novel compounds that inhibit cell proliferation and pro-
mote cellular apoptosis, the synthesized imidazole deriv-
atives were initially profiled in phenotypic cell based
assays and tested for apoptosis-inducing ability.
Figure 5. Apoptosis induced by representative trisubstituted imidaz-
oles in HL-60 cells in a DNA fragmentation assay. Lanes 1, 14: 100 bp
DNA markers; 2, 13: DMSO control; 3, 12: 1.25 lM camptothecin; 4,
5, 6, and 7: 100, 10, 1, and 0.1 lM 3, respectively; 8, 9, 10, and 11: 0.1,
1, 10, and 100 lM 2, respectively.
It is observed from Table 1 that representative trisubsti-
tuted imidazoles exhibited micromolar potency for cell
growth inhibition14 in a HL-60 cell line. The induction
of apoptosis assessed in a DNA fragmentation assay
in HL-60 cells15 for two compounds listed in Table 1,
2 and 3 is shown in Figure 5. Both compounds were able
to induce apoptosis in a dose-dependent manner consis-
tent with their IC50 values in the cell proliferation assay.
groove. Terphenyl based a-helix mimetics of a BH3 pep-
tide epitope that are capable of inhibiting this PPI have
been reported.7
In our initial screen, three representative compounds
(Table 1) were found to disrupt the interaction of
Bcl-w with the Bak-BH3 peptide. Surprisingly, all three
compounds were inactive against other Bcl family mem-
bers such as Bcl-2 or Bcl-XL (data not shown). Modeling
studies on imidazole 2 demonstrated the ability of these
compounds to mimic the critical i, i + 3, and i + 7 resi-
dues of an a-helix epitope of a proapoptotic Bcl family
peptide (Fig. 6).
To preliminarily delineate the mode of action of the syn-
thesized trisubstituted imidazoles as apoptosis inducers,
we screened them in a FP assay17 established in-house
for inhibition of Bcl family PPI. Significantly, the major
secondary structure elements in the Bcl-2 subfamily of
antiapoptotic proteins are a-helices, arranged in such a
way that a large hydrophobic groove is formed on the
protein surface. The BH3 domains of the proapoptotic
proteins then bind as an amphipathic a-helix within this
In summary, we have introduced the design of novel
imidazole based a-helix mimetics that are conceptually
derived from the Hamilton terphenyl compounds, but
are synthetically amenable in one step utilizing a
MCR. The described trisubstituted imidazoles comprise
a novel class of lead molecules suitable for further opti-
mization. Further studies are needed to fully delineate
the biochemical mode of action of representatives of this
lead series. The selection of a wider set of starting mate-
rials should lead to compounds with improved biologi-
cal and pharmacokinetic activities.
Table 1. Cell proliferation data (HL-60) and in vitro FP data (Bcl-w)
Compound
Structure
GI50 (lM)
HL-60
IC50 (lM)
FP, Bcl-w
MeO
N
1
12.95 0.11
26.85 1.24
N
O
N
2
12.51 2.68
13.99 4.11
N
MeO
O
N
Figure 6. A designed van Leusen MCR derived trisubstituted
imidazole a-helix mimetic modeled into the structure of the Bcl-XL/
Bad peptide complex (pdb-code 1G5J). Compound 2 (gold) is
3
9.43 1.3
8.09 2.62
N
superimposed onto the a-helical Bad peptide epitope i, i + 3, and
i + 7 residues. Thus, the phenyloxy substituent on the imidazole
superimposes with the side chain of Tyr (i + 7), the n-propyl moiety
with the side chain of Leu (i + 3), and the methoxyphenyl moiety
with the side chain of Met (i).
MeO

W. Antuch et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1740–1743
1743
from a water aspirator. The column was first eluted with
0.1% MeOH in CH2Cl2 (60 mL) to selectively remove
References and notes
reaction impurities.
A subsequent wash with CH2Cl2
1. Tsai, C.-J.; Lin, S. L.; Wolfson, H. J.; Nussinov, R. J.
Mol. Biol. 1996, 260, 604.
2. (a) Toogood, P. L. J. Med. Chem. 2002, 45, 1543; (b)
Pagliaro, L.; Felding, J.; Audouze, K.; Nielsen, S. J.;
Terry, R. B.; Krog-Jensen, C.; Butcher, S. Curr. Opin.
Chem. Biol. 2004, 8, 442.
3. (a) 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. Science 1997, 275, 963; (b) Bo¨ttger, A.; Bo¨ttger, V.;
(20 mL) containing MeOH (1%) eluted the desired imidazole
product (18 mg, 43%). MS (electrospray, + ions): m/z = 413
(M+H).
14. Cells were plated in 96-well plates in 50 ll of media at a
density of 70,000 cells/well. Compounds were serially
diluted into cell media as 2· stocks containing 1%
DMSO and 50 ll of each dilution was added to the wells
in triplicate. After 48 h of incubation with the com-
pounds, cell survival was assessed by adding 10 ll WST-1
reagent (Roche Applied Science) in each well. After
incubation for an additional 3 h, the plates were read in a
plate reader (Spectramax 250 from Molecular Devices,
Sunnyvale, CA) at 450 nm. The time 0-h for cells with
only DMSO (final concentration 0.5%) was also assessed
for each plate and was subtracted from the 48-h time
points for both treated and untreated data points. The
XLfit program (ID Business Solutions Inc., Parsippany,
NJ) was used to obtain and calculate the GI50 values for
each compound. Each experiment was done a minimum
of three times.
`
Garcia-Echeverria, C.; Chene, P.; Hochkeppel, H.-K.;
Sampson, W.; Ang, K.; Howard, S. F.; Picksley, S. M.;
Lane, D. P. J. Mol. Biol. 1997, 259, 744.
4. Haggarty, S. J.; Mayer, T. U.; Miyamoto, D. T.; Fathi, R.;
King, R. W.; Mitchison, T. J.; Schreiber, S. L. Chem. Biol.
2000, 7, 275.
5. (a) Nolan, W. P.; Ratcliffe, G. S.; Rees, D. C. Tetrahedron
Lett. 1992, 33, 6879; (b) Horwell, D. C.; Howson, W.;
Ratcliffe, G. S.; Willems, H. M. G. Bioorg. Med. Chem.
1996, 4, 33.
6. Jacoby, E. Bioorg. Med. Chem. Let. 2002, 12, 891.
7. Kutzki, O.; Park, H. S.; Ernst, J. T.; Orner, B. P.; Yin, H.;
Hamilton, A. D. J. Am. Chem. Soc. 2002, 124, 11838.
8. Orner, B. P.; Ernst, J. T.; Hamilton, A. D. J. Am. Chem.
Soc. 2001, 123, 5382.
15. Briefly 106 cells per well were plated into a 24-well plate.
The cells were incubated with or without the compound in
0.5% DMSO for 24 h. After incubation, the cells were
collected, the DNA was extracted and subjected to gel
electrophoresis as described.16 Camptothecin was used as
a positive control in this assay.
9. Moloc; Paul Gerber. Gerber Molecular Design, http://
www.moloc.ch.
16. Wang, J. L.; Zhang, J. G.; Choksi, S.; Shan, S.; Lu, Z.;
Croce, C. M.; Alnemri, E. S.; Korngold, R.; Huang, Z.
Cancer Res. 2000, 60, 1498.
10. Recent reviews on van Leusen chemistry (a) van Leusen, A.
M.; van Leusen, D. Org. React. 2003, 57, 419; (b) http://
www.organic-chemistry.org/Highlights/2005/05Mai.shtm.
11. A variety of substituted TosMICs are commercially
available at: www.synthatex.com.
12. Sisko, J.; Kassick, A. J.; Mellinger, M.; Filan, J.; Allen,
A.; Olsen, M. A. J. Org. Chem. 2000, 65, 1516.
13. General procedure for the parallel, solution-phase multi-
component synthesis of 1,4,5-trisubstituted imidazoles; 5-
butyl-1-(2-methoxybenzyl)-4-(2-phenoxy-phenyl)-1H-imidaz-
17. A primary screen for Bcl project is based on fluorescence
polarization (FP) technology and is well described in the
literature.18 We have established assays for Bcl-2 (Santa
Cruz Biotech), Bcl-XL (R&D Systems), and Bcl-w (R&D
Systems) using as the binding partner the 5-carboxyfluo-
rescein-labeled 16-mer peptide tracer Flu-Bak-BH3
(sequence GQVGRQLAIIGDDINR is derived from the
Bak BH3 domain). The assays were performed in a 384-
well format, in 20 mM potassium phosphate buffer, pH
7.4, containing 1 mM EDTA, 50 mM NaCl, and 0.05%
Pluronic F-68. The final concentration of DMSO in all
assays was 10%. The reaction was carried out in a 50 lL
volume and the resulting polarization signal was measured
at kex = 485 nm/kem = 535 nm using an UltraReader (Tec-
an) after 2 h incubation of the reaction mixture at room
temperature. Validation of the assays was performed using
non-labeled Bak-BH3 peptide as a control inhibitor.
18. Zhang, H.; Nimmer, P.; Rosenberg, S. H.; Ng, S. C.;
Joseph, M. Anal. Biochem. 2002, 307, 70.
ole:
A
solution of 2-methoxy-benzylamine (13 lL,
0.10 mmol) and valeraldehyde (11 lL, 0.10 mmol) in MeOH
(0.5 mL) was stirred at room temperature for 1 h. (2-
Phenoxyphenyl) tosylmethyl isonitrile (36 mg, 0.10 mmol)
and triethylamine (28 lL, 0.20 mmol) were now added and
the resulting mixture was stirred and heated at 65 ꢁC for 4 h.
Solvent was evaporated under reduced pressure and the
residue was redissolved in CH2Cl2 (0.5 mL). The crude
product was loaded onto a short silica gel column (ꢁ1 g),
packed in a disposable syringe (10 mL), that had been
previously pre-eluted with CH2Cl2 (10 mL) containing
MeOH (0.1%) and then sucked dry by applying suction