Rational hopping of a peptidic scaffold into non-peptidic scaffolds: Structurally novel potent proteasome inhibitors derived from a natural product, belactosin A

Article abstract of DOI:10.1039/c3cc48818g

Rational scaffold hopping of a natural product belactosin A derivative was successfully achieved based on the pharmacophore model constructed. The peptidic scaffold was replaced by significantly simplified non-peptidic scaffolds, by which weak belactosin A (IC₅₀ = 1440 nM) was converted into highly potent non-peptidic inhibitors (IC₅₀ = 26-393 nM).

Full text of DOI:10.1039/c3cc48818g

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Rational hopping of a peptidic scaffold into
non-peptidic scaffolds: structurally novel potent
proteasome inhibitors derived from a natural
product, belactosin A†
Cite this: Chem. Commun., 2014,
50, 2445
Received 19th November 2013,
Accepted 8th January 2014
Shuhei Kawamura,a Yuka Unno,b Takatsugu Hirokawa,c Akira Asai,b
DOI: 10.1039/c3cc48818g
Mitsuhiro Arisawad and Satoshi Shuto*ae
www.rsc.org/chemcomm
Rational scaffold hopping of a natural product belactosin A derivative significantly different molecular backbones’’.11 Recently, Sun and
was successfully achieved based on the pharmacophore model con- co-workers classified scaffold hopping approaches into four major
structed. The peptidic scaffold was replaced by significantly simplified categories based on the degree of structural change: (a) 11 hop:
non-peptidic scaffolds, by which weak belactosin A (IC50 = 1440 nM) was heterocycle replacement, (b) 21 hop: ring opening or closure, (c) 31
converted into highly potent non-peptidic inhibitors (IC50 = 26–393 nM). hop: peptidomimetics, and (d) 41 hop: topology-based hopping, in
which a scaffold is converted into a completely new chemical back-
In the drug discovery process, not only the desired biological effect on bone retaining the required topology of the key functional groups.7 In
target biomolecule, but also various other properties such as meta- general, the higher the hop level becomes, the more drastic the
bolic stability, membrane permeability, solubility, toxicity profile, and change in molecular properties, while the success rate inversely
synthetic accessibility of leads, have to be optimized.1,2 The introduc- correlates with the magnitude of the structural change. Thus, success
tion and/or modification of substituents of a lead are commonly examples of topology-based hopping are significantly rare,7 despite its
investigated to optimize these properties, but a more drastic struc- potential to improve the molecular properties drastically.
tural change, i.e., scaffold hopping, is occasionally needed to identify
Belactosin A is a tripeptide natural product produced by Strepto-
drug candidates with the desired properties.3–7 Although natural myces sp. that comprises L-alanine, 3-(trans-2-aminocyclopropyl)-
products are useful sources of drug candidates, their complex L-alanine, and a chiral carboxy-b-lactone moiety,12 and inhibits the
structures sometimes limit their use as clinical drugs.8 Peptides are proteasome ChT-L activity13 by acylating the active-site Thr residue via
also potentially useful for the treatment of diseases, but their low its strained b-lactone opening, as confirmed by X-ray crystallographic
membrane permeability and biological stability often limit their analysis.14–16 Due to its proteasome inhibitory activity, belactosin A
application as clinical drugs.9 To create promising drug candidates prevents cell cycle progression at the G2/M stage in tumor cells and is
from these leads, including natural products and peptides, the therefore a novel lead for developing potent anticancer agents.12,13
development of a rational methodology for scaffold hopping to
In recent years, we have performed systematic structure–activity
address these limitations is required. Here we report the development relationship (SAR) studies of belactosin A16–19 and identified a highly
of potent non-peptidic proteasome inhibitors derived from a peptidic potent proteasome inhibitor 4 (IC50 = 5.7 nM) (Fig. 1). However, the
natural product, belactosin A, based on rational scaffold hopping.10 synthetic accessibility of 4 is too low (total synthetic steps 26) due to
The term ‘‘scaffold hopping’’ was defined by Schneider et al. in its multichiral peptidic scaffold. Furthermore, peptidic scaffolds
1999 as the ‘‘identification of isofunctional molecular structures with generally impair membrane permeability and bioavailability.9 There-
fore, to identify novel leads with superior drug-like properties, we
a Faculty of Pharmaceutical Sciences, Hokkaido University, Kita-12, Nishi-6, Kita-ku,
Sapporo 060-0812, Japan. E-mail: shu@pharm.hokudai.ac.jp;
Fax: +81-11-706-3769
planned to convert the multichiral peptidic scaffold of 4 into an
achiral non-peptidic scaffold by topology-based hopping.
Previously, we analyzed binding mode of 4 around the transition
state (Fig. 2b, green tube).20 In addition, two crystal structures of
homobelactosin C derivatives in complex with proteasomes (Fig. 2b,
yellow and orange wires) have been analyzed by Meijere et al.14,15
These analyses importantly demonstrate that no hydrogen bond is
formed between peptidic scaffolds of these belactosin derivatives and
proteasomes, suggesting that these peptidic scaffolds can be replaced
with non-peptidic scaffolds. Furthermore, in these binding struc-
b Graduate School of Pharmaceutical Sciences, University of Shizuoka, Yada,
Shizuoka 422-8526, Japan
c Molecular Profiling Research Center for Drug Discovery (MOLPROF),
National Institute of Advanced Industrial Science and Technology (AIST), 2-4-7,
Aomi, Koutou-ku, Tokyo 135-0064, Japan
d Graduate School of Pharmaceutical Sciences, Osaka University, 1-6 Yamada-oka,
Suita, Osaka 565-0871, Japan
e Center for Research and Education on Drug Discovery, Hokkaido University,
Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812, Japan
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c3cc48818g tures, the peptidic moieties are bent to place aromatic rings onto
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 2445--2447 | 2445

View Article Online
Communication
ChemComm
Fig. 1 Potent proteasome inhibitor 4 developed by us.
hydrophobic proteasome surfaces, which implies that these peptidic
scaffolds can be shortened.
This structural information and the previous SAR studies15–18
suggest that the only b-lactone moiety and two aromatic groups are
essential for the strong binding so that we constructed a pharmaco-
phore model composed of these three moieties, as shown in Fig. 2c.
Based on the pharmacophore model, we designed the non-peptide
derivatives 5–12, in which two aromatic groups were placed sym-
metrically to eliminate chiral centers. The designed compounds were
flexibly superimposed onto the bioactive conformation of 4 and
homobelactosin C derivatives in Fig. 2b. As shown in Fig. 2d, the
b-lactone moiety and two hydrophobic groups were well super-
imposed in all these compounds without steric repulsion with
proteasomes, suggesting that these newly designed compounds can
retain the strong inhibitory activity of 4.
The inhibitory effects of the compounds on the ChT-L activity of
human 20S proteasomes were measured using a chromophoric
substrate Suc-LLVY-AMC, and the results are summarized in
Table 1. All of the compounds with amine-type scaffolds (5–8) showed
potent proteasome inhibitory activity (IC50 = 175–393 nM). Com-
pounds 9–12 with ether-type scaffolds showed further strong inhibi-
tory activity (IC50 = 26–56 nM), in spite of their achiral non-peptidic
scaffolds, which were markedly simplified in comparison with the
multichiral peptidic scaffold of 4. In these compounds, the linker
length only weakly impacts upon their proteasome inhibitory activ-
ities. The flexible nature of these scaffolds would allow the two
aromatic rings to be placed into the proteasome binding sites
regardless of their linker length as expected (Fig. 2d).
We also evaluated the cell growth inhibitory effect of these
compounds on HCT116 (Table 1). Most of these compounds
retained the potent cell growth inhibitory effects of their parent
compound 4, and, in particular, the cell growth inhibitory effect of
11 was as potent as 4.21
To clarify the impact of the scaffold hopping on the molecular
properties as leads, we calculated the binding efficiency index (BEI)
and the surface-binding efficiency index (SEI) of these compounds
(Table 1): BEI and SEI indicate binding efficiency per molecular
weight and polar surface area (PSA), respectively.22 Because molecular
Fig. 2 Rational design of the non-peptidic belactosin
A derivatives by
topology-based scaffold hopping. (a) Structure of the compound 4, homo-
belactosin C derivatives14,15 and newly designed non-peptide derivatives 5–12.
(b) Previously analyzed non-covalent binding mode of 4 (green tube) around
the transition state20 and X-ray analyzed binding mode of the homobelactosin
C derivatives (yellow and orange wires).14,15 (c) Plausible pharmacophore model
constructed. (d) Results of flexible alignment of the designed compounds (5 and
9, aquamarine; 6 and 10, turquoise; 7 and 11, yellow green; 8 and 12, pink) onto
the binding conformation of the belactosin derivatives shown in (b).
weight and PSA are well known physicochemical properties well- SEI can be effective indicators of drug-likeness. The desirable leads in
related to membrane permeability and bioavailability,23–27 BEI and the drug discovery process generally have high BEI and SEI values.22
2446 | Chem. Commun., 2014, 50, 2445--2447
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
Table 1 Proteasome and cell growth inhibitory effects and molecular
properties of novel non-peptide proteasome inhibitors 5–12
We successfully achieved the topology-based scaffold hopping
of 4. The multichiral peptidic scaffold of 4 was rationally replaced
with achiral non-peptidic scaffolds to identify the superior leads
suitable for further optimization. The present study demonstrates
that topology-based hopping is a highly effective strategy to change
the undesired molecular properties of leads, including natural
products and peptides. This study is also of vital importance in terms
of the use of a natural product as a lead. Thus, a weak peptidic
natural product inhibitor belactosin A (IC50 = 1440 nM) was converted
into a highly potent non-peptide inhibitor 11 (IC50 = 26 nM).
This work was supported by Grant-in-Aids for Scientific
Research 24390023 (S.S.) from the Japan Society for the Promo-
tion of Science and for Platform for Drug Discovery, Infor-
matics, and Structural Life Science (T.H.) from the Ministry of
Education, Culture, Sports, Science and Technology, Japan.
ChT-L activity
Properties compared with 4
Cell growth
IC50 [mM]
Cpd IC50 [nM] BEIb SEIc Mw
tPSAd
a
a
5
6
7
8
175 Æ 32 37
393 Æ 24 33
215 Æ 40 35
352 Æ 25 32
26 À141
25 À113
26 À127
À64
À64
À64
À64
8.90
>10
2.20
25
À99
3.80
9
28 Æ 2.2 41
29 Æ 13 38
26 Æ 6.1 40
56 Æ 16 36
23 À138
23 À110
24 À124
À49
À49
À49
À49
4.10
>10
1.80
10
11
12
22
15
À96
4.20
4
5.7 Æ 1.2 34
—
—
1.82
a
b
Based on three experiments. pIC50 per molecular weight (kDa).
c
d
pIC50 per polar surface area (PSA) normalized to 100 Å2. Calculated
by ChemBioDraw Ultra 12.
Notes and references
Fig. 3 shows the plot of BEI vs. SEI of the newly identified non-
peptide inhibitors 5–12 and the previously developed inhibitors with
the peptide scaffold identical to that of 4.16 The BEI values of the
peptidic inhibitor 4 and its congeners vary considerably (BEI = 25–43),
but their SEI values are analogous (SEI = 13–16), indicating that the
substituent modifications effectively improved the affinity for the
proteasome but not the molecular properties arising from the scaf-
fold. On the other hand, the newly identified non-peptidic inhibitors
5–12 show remarkably higher SEI values compared with their parent
4. Furthermore, their BEI are comparable with or superior to 4 due to
their significantly lower molecular weight (Table 1). These improved
BEI and SEI values suggest that these novel non-peptide inhibitors are
clearly superior to 4 as lead compounds for optimization.
1 R. Borchardt, E. Kerns, M. Hageman, D. Thakker and J. Stevens,
Optimizing the ‘‘drug-like’’ Properties of Leads in Drug Discovery,
Springer, 2006.
2 E. Kerns and L. Di, Drug-like properties: concepts, structure design and
methods: from ADME to toxicity optimization, Academic Press, 2010.
¨
3 H.-J. Bohm, A. Flohr and M. Stahl, Drug Discovery Today: Technol.,
2004, 1, 217.
4 G. Schneider, P. Schneider and S. Renner, QSAR Comb. Sci., 2006, 25, 1162.
5 H. Zhao, Drug Discovery Today, 2007, 12, 149.
6 S. R. Langdon, P. Ertl and N. Brown, Mol. Inf., 2010, 29, 366.
7 H. Sun, G. Tawa and A. Wallqvist, Drug Discovery Today, 2012, 17, 310.
8 A. L. Harvey, Drug Discovery Today, 2008, 13, 894.
9 C. Adessi and C. Soto, Curr. Med. Chem., 2002, 9, 963.
10 During our study, simplified belactosin C analogues were reported
by A. de Meijere, et al., Org. Biomol. Chem., 2012, 10, 6363.
11 G. Schneider, W. Neidhart, T. Giller and G. Schmid, Angew. Chem.,
Int. Ed., 1999, 38, 2894.
12 A. Asai, A. Hasegawa, K. Ochiai, Y. Yamashita and T. Mizukami,
J. Antibiot., 2000, 53, 81.
13 A. Asai, T. Tsujita, S. V. Sharma, Y. Yamashita, S. Akinaga,
M. Funakoshi, H. Kobayashi and T. Mizukami, Biochem. Pharmacol.,
2004, 67, 227.
Furthermore, synthetic accessibility of these novel non-peptide
derivatives is significantly superior to that of 4 due to their signifi-
cantly simplified scaffolds with no chiral center.
14 M. Groll, O. V. Larionov, R. Huber and A. De Meijere, Proc. Natl.
Acad. Sci. U. S. A., 2006, 103, 4576.
15 V. S. Korotkov, A. Ludwig, O. V. Larionov, A. V. Lygin, M. Groll and
A. de Meijere, Org. Biomol. Chem., 2011, 9, 7791.
16 S. Kawamura, Y. Unno, A. List, A. Mizuno, M. Tanaka, T. Sasaki,
M. Arisawa, A. Asai, M. Groll and S. Shuto, J. Med. Chem., 2013, 56, 3689.
17 K. Yoshida, K. Yamaguchi, T. Sone, Y. Unno, A. Asai, H. Yokosawa,
A. Matsuda, M. Arisawa and S. Shuto, Org. Lett., 2008, 10, 3571.
18 K. Yoshida, K. Yamaguchi, A. Mizuno, Y. Unno, A. Asai, T. Sone,
H. Yokosawa, A. Matsuda, M. Arisawa and S. Shuto, Org. Biomol.
Chem., 2009, 7, 1868.
19 S. Kawamura, Y. Unno, A. Asai, M. Arisawa and S. Shuto, Org.
Biomol. Chem., 2013, 11, 6615.
20 S. Kawamura, Y. Unno, M. Tanaka, T. Sasaki, A. Yamano, T. Hirokawa,
T. Kameda, A. Asai, M. Arisawa and S. Shuto, J. Med. Chem., 2013,
56, 5829.
21 The cell growth inhibitory activity of the belactosin derivatives may be
moderate than that expected from their inhibitory potency on ChT-L
activity probably because of the instability of the b-lactone moiety of
these compounds under the assay conditions with cell system.
22 C. Abad-Zapatero and J. T. Metz, Drug Discovery Today, 2005, 10, 464.
23 M. Vieth, M. G. Siegel, R. E. Higgs, I. A. Watson, D. H. Robertson,
K. A. Savin, G. L. Durst and P. A. Hipskind, J. Med. Chem., 2003, 47, 224.
24 M. C. Wenlock, R. P. Austin, P. Barton, A. M. Davis and P. D. Leeson,
J. Med. Chem., 2003, 46, 1250.
25 W. J. Egan, K. M. Merz and J. J. Baldwin, J. Med. Chem., 2000, 43, 3867.
26 P. Ertl, B. Rohde and P. Selzer, J. Med. Chem., 2000, 43, 3714.
27 K. Palm, P. Stenberg, K. Luthman and P. Artursson, Pharm. Res., 1997,
14, 568.
Fig. 3 BEI and SEI values of belactosin derivatives synthesized by our group.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 2445--2447 | 2447