Synthesis and evaluation of a library of 2,5-bisdiamino-benzoquinone derivatives as probes to modulate protein-protein interactions in prions

Article abstract of DOI:10.1016/j.bmcl.2010.01.149

A small library combining two different benzoquinone cores with seven (l) amino acid methyl esters (alanine, Nω-nitro-arginine, Nε-BOC-lysine, isoleucine, methionine, phenylalanine and tryptophan) was prepared and tested for prion replication inhibition in ScGT1 cells. The most potent hit, 6a, displayed an EC₅₀ value of 0.87 μM, which is very close to that of quinacrine (0.4 μM).

Full text of DOI:10.1016/j.bmcl.2010.01.149

Bioorganic & Medicinal Chemistry Letters 20 (2010) 1866–1868
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and evaluation of a library of 2,5-bisdiamino-benzoquinone
derivatives as probes to modulate protein–protein interactions in prions
Hoang Ngoc Ai Tran a, Salvatore Bongarzone b,c, Paolo Carloni b,c, Giuseppe Legname a,c
,
Maria Laura Bolognesi d,
*
a Neurobiology Sector, International School for Advanced Studies (SISSA), 34151 Trieste, Italy
b Statistical and Biological Physics Sector, International School for Advanced Studies (SISSA), 34151 Trieste, Italy
c SISSA-Unit, Italian Institute of Technology, 34151 Trieste, Italy
d Department of Pharmaceutical Sciences—Alma Mater Studiorum-Bologna University, Via Belmeloro 6, I-40126, Bologna, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
A small library combining two different benzoquinone cores with seven (
L
) amino acid methyl esters (ala-
Received 15 December 2009
Revised 28 January 2010
Accepted 29 January 2010
Available online 4 February 2010
nine, N
pared and tested for prion replication inhibition in ScGT1 cells. The most potent hit, 6a, displayed an EC50
value of 0.87 M, which is very close to that of quinacrine (0.4 M).
x-nitro-arginine, Ne-BOC-lysine, isoleucine, methionine, phenylalanine and tryptophan) was pre-
l
l
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Prion diseases
Conformational diseases
Library design and synthesis
Protein–protein interactions (PPIs) are crucial elements in
mediating diverse cellular physiological and pathological events.1
They are involved in fibrillation processes and thus play a pivotal
role in the pathogenesis of several neurodegenerative diseases.2,3
Systematic analysis of PPI interfaces reveals great heterogene-
ity, from large and flat to narrow and structured interactions.4
However, the majority of PPIs deal with protein surfaces,5 where
a complex network of weak interactions takes place.
on the concept that the ‘hot spot’ regions in protein–protein inter-
faces are rich in aromatic residues.
Here, we focus on PPIs in prion diseases. These diseases are
characterized by the aggregation and accumulation of a misfolded
prion protein, PrPSc, which derives from a post-translational con-
formational change of the host-encoded cellular prion protein,
PrPC.12 The conformational transition process involved remains
enigmatic. However, regardless of the initiating event, PrPSc ap-
pears to act as a conformational template by which PrPC is con-
verted to a new molecule of PrPSc, through PPIs.12
So far, several peptides have been developed with the specific
aim of blocking PPIs and reversing the aberrant conformational
changes. A short synthetic peptide (iPrP13, DAPAAPAGPAVPV), de-
signed by Soto and co-workers on the basis of sequence homology
with PrPC, acted as a b-sheet breaker, inducing unfolding of b-
pleated sheet structure.13 More recently, Gilbert and co-workers14
Peptides may be good PPI blockers.6,7 However, they are not
optimal drug candidates, due to problems with bioavailability
and enzymatic degradation. To overcome this limitation, one could
use combinatorial chemical libraries based on small molecules.
However, the widely spaced interactions required for PPI blockers
are difficult to mimic with small molecules. Despite this challenge,
the strategy holds great potential for identifying novel lead com-
pounds against PPIs.8 The extreme attractiveness of PPIs as drug
targets has led to important progresses in this field in the last dec-
ade.1,9,10 In particular, Janda and co-workers have recently demon-
strated the ability of what they have named ‘‘credit card” libraries
to disrupt PPIs of biological relevance.11 The chemical structures of
these libraries are built upon flat, rigid scaffolds, decorated with
appended groups that span a wide range of size, aromaticity, polar-
ity, and hydrogen-bonding capability.11 Their rationale was based
reported on a series of small peptides active at levels of 100 lM in
two prion disease models and in an in vitro anti-aggregation poly-
merization assay. Prompted by the advantages of using small mol-
ecules as PPI inhibitors as opposed to peptides, here we propose
the planar 2,5-bisdiamino-benzoquinone scaffold as a privileged
motif in modulating PPIs. This is based on (i) Janda’s criteria for
credit card libraries11; (ii) the finding that a 2,5-bisdiamino-benzo-
quinone derivative binds to b-amyloid (Ab), and interferes with the
native ability of Ab to self-assemble, by disrupting PPIs.15 Due to a
resonance effect, a hydrophobic and planar system is generated in
2,5-bisdiamino-benzoquinones. This should, in principle, perturb
* Corresponding author. Tel.: +39 051 2099718.
E-mail address: marialaura.bolognesi@unibo.it (M.L. Bolognesi).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.01.149

H. N. A. Tran et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1866–1868
1867
PPIs in the fibrillogenesis processes.16 Therefore, in our search for
novel anti-prion compounds, we decided to attach seven amino
acids methyl esters to two different benzoquinone cores, generat-
ing a small combinatorial library of fourteen 2,5-bisdiamino-ben-
zoquinones (1–7a and 1–7b), reported in Figure 1. The selected
to reduce PrPSc concentrations in ScGT1 cells was determined by
Western blot densitometry of the PK-resistant PrPSc. First, we
determined the effects of library compounds on cell viability (Ta-
ble 1). Compound toxicity is expressed as an average percentage
of viable cells when treated with a compound concentration of
amino acid esters (AlaOMe (1), N
x
-Nitro-ArgOMe (2), N
e
-BOC-Ly-
1 lM, versus control cells treated with no compound. For com-
sOMe (3), IleOMe (4), MetOMe (5), PheOMe (6), TrpOMe (7)) act as
capping groups, allowing us to enlarge the library’s chemical diver-
sity by exploiting differences in size, aromaticity, polarity, and
hydrogen-bonding capability. Analysis of natural amino acids in-
volved in PPIs revealed that Trp, Phe, Tyr, and Ile are the most
important in driving aggregation.17 Consequently, it is highly con-
ceivable that the novel derivatives bearing these motifs might
compete for binding and, therefore, efficiently disrupt the assem-
bly of prion protein.
pound 6a, the LC50 value (lethal concentration, 50%) for ScGT1 cells
was also identified. Anti-prion activity is expressed as the average
percentage of PK-resistant PrP remaining after incubation with
compound at the given concentration, versus control cells incu-
bated with no compound. For entries 6a, 6b, and 7a, we also calcu-
lated the EC50 values, which represent the effective concentrations
for half-maximal inhibition.
Analyzing the results in Table 1, we note some interesting
trends. The cytotoxic effects of 1–7a and 1–7b were first deter-
mined by a Calcein-AM assay in the ScGT1 cell line. As reported,
To develop an efficient parallel synthesis approach, we focused
on the displacement reaction of tetrahalo-substituted quinones
(chloranil, a and bromanil, b) with amino acid methyl esters (1–
treating ScGT1 with the compounds (1 lM) did not significantly
modify cell viability. However, treatment with the BOC-Lys deriv-
atives 3a and 3b decreased cell viability by percentages of 40% and
52%. Because of the toxicity profile shown 3a and 3b were not
screened for prion replication, whereas the other library members
7) to afford
a library of fourteen 2,5-diamino-3,6-dihalo-
[1,4]benzoquinone derivatives. We carried out a one-pot reaction
at room temperature that, in most cases, would achieve the quan-
titative conversion of the starting reactant within 3 h (see
Scheme 1). Moreover, we developed an operationally simple and
versatile workup protocol, which involved recovering high purity
final products by filtration upon addition of water to the reaction
mixture.
were assayed at a concentration of 1 lM. Notably, despite this low
concentration range, three couples of library hits were active
against PrPSc accumulation. IleOMe (4a–b), MetOMe (5a–b) and
PheOMe (6a–b) derivatives of both series at 1
lM displayed activ-
ities ranging from 4.8% to 73%. Conversely, AlaOMe (1a–b), N
x
-Ni-
A cell-screening assay was used to test anti-prion activity across
the library of synthesized compounds. The ability of 1–7a and 1–7b
tro-ArgOMe (2a–b), TrpOMe (7a–b) derivatives had no effect at
that concentration. Due to its low cytotoxicity, 7a could also be
tested at higher concentrations, revealing a fair EC50 value of
7.7
(0.87 0.1
(0.4 0.1
l
M. For 6a, we found a remarkable submicromolar EC50 value
M; see Fig. 2), comparable to that of quinacrine
M), a reference anti-prion compound.18,19 The high
O
O
O
O
l
l
H
N
H
Cl
H
Br
H
N
R1
O
O
R1
O
O
R1
H
Cl
R1
H
Br
activity of 6a and 6b was not unexpected, as it is in line with the
well-known central role of pi-stacking interactions in self-assem-
bly processes in the fields of chemistry and biochemistry.20
To better rationalize the obtained results, we applied a system-
atic procedure for identifying key fragments responsible for a given
activity.21 We used an algorithm which breaks down a structure
into fragments.22 Subsequently, all the obtained substructures
were related to biological activities to identify hot fragments. The
analysis provides a score [100–0] for each fragment, which gives
an indication of how often a fragment occurs in the active and inac-
tive structures (Fig. 3). From this preliminary computational study,
we have identified that the 2,5-bisdiaminobenzoquinone linked
with two phenyl rings by a spacer is a good anti-prion motif. In
addition, our analysis suggests the relevance of the atomic size of
the substituents in position 3 and 6 at benzoquinone ring (Cl being
better than Br), with an inverse relationship to van der Waals ra-
dius. Both items of information can be exploited to design further
series of anti-prion small molecules.
O
O
N
H
N
H
O
1-7a
O
1-7b
R1
S
1
2
5
6
NH
O
N
N
H
N
H
O
H
N
O
3
4
7
O
N
H
Despite the small number of compounds synthesized, the re-
sults suggest that some are active against prion replication.
Although the existing derivatives were overall quite cytotoxic to-
ward ScGT1 cells, we identified entries 6a and 6b as hit compounds
for further lead optimization studies. At a time when it remains
Figure 1. Chemical structures of 1–7a and 1–7b.
O
O
O
O
H
N
R1
H
X
X
X
X
X
R1
H
O
a
O
H
R1
NH3
Cl-
O
N
H
X
O
O
O
1-7
a: X = Cl
b: X = Br
1-7a
1-7b
Scheme 1. Reagents and conditions: (a), EtOH, NEt3, 3 h, 30–45% yield.

1868
H. N. A. Tran et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1866–1868
Table 1
Cell viability and anti-prion activity on ScGT1 cells of library compounds
Mb
% PrPSc inhibition at 1
l
Mc
EC50
(
lM)
LC50 (lM)
c
b
Cmpd
% of Viable cells at 1
l
1a
1b
2a
2b
3a
3b
4a
4b
5a
5b
6a
6b
7a
7b
71.6 5.4a
91.2 6.8
81.4 8.7
74.1 4.3
40.7 5.5
52.3 5.9
88.2 6.1
91.3 8.8
95.5 6.2
96.1 7.5
68.4 7.3
80.2 5.8
96.0 7.6
65.9 3.4
0.36 0.034a
0.29 0.05
0.23 0.01
0.31 0.03
ND
ND
6.6 0.4
28.1 1.5
4.8 0.7
11.4 0.5
73.2 3.3
18.1 0.5
0.25 0.04
0.43 0.01
0.87 0.1
3.6 0.5
7.7 1.2
2.4 0.2
a
Values are the mean of three experiments, standard deviations are given.
ScGT1 cells were cultured in DMEM with 10% FBS, plated 25000 cells in each well of 96-well plates. The compounds were dissolved in DMSO (100%) and diluted in PBS 1X
b
before adding various concentrations (1 nM–10 lM) and incubated for 5 days at 37 °C, 5% CO2. The results were developed by Calcein-AM fluorescence dye and read by
microplate reader.
c
The effect of library compounds on inhibition of scrapie prion replication. ScGT1 cells were cultured in DMEM with 10% FBS, split 1:10 into Petri dishes and incubated for
M), being non-toxic for the cells, were added to the plates. After a 5-day incubation, proteins
2 days at 37 °C and 5% CO2. Then, various compound concentrations (0.1 nM–1
l
of cells were extracted, quantified, digested with proteinase K (PK), and Western-blotted.
challenging to design chemical entities able to target PPIs, these
studies might shed light on the underlying principles governing
molecular recognition and the chemical basis for the inhibition of
quinone derivatives in prions.
Acknowledgement
This work was supported by the University of Bologna (to MLB)
and by Friuli-Venezia-Giulia Region through the project ‘Search for
new drugs’ (SeND) (to GL). The authors wish to thank Gabriella
Furlan for editing and proofreading the manuscript. Damiano Roc-
chi is acknowledged for excellent technical assistance.
Cmpd 6a
µM
0
0.2
0.4
0.6
0.8
1.0
kDa
50
37
PK -
References and notes
25
1. Arkin, M. R.; Wells, J. A. Nat. Rev. Drug Disc. 2004, 3, 301.
2. Forman, M. S.; Trojanowski, J. Q.; Lee, V. M. Nat. Med. 2004, 10, 1055.
3. Trojanowski, J. Q.; Lee, V. M. Ann. NY Acad. Sci. 2000, 924, 62.
4. Nooren, I. M.; Thornton, J. M. Embo J. 2003, 22, 3486.
5. Kranjc, A.; Bongarzone, S.; Rossetti, G.; Biarnes, X.; Cavalli, A.; Bolognesi, M. L.;
Roberti, M.; Legname, G.; Carloni, P. J. Chem. Theory Comput. 2009, 5, 2565.
6. Soto, C. FEBS Lett. 2001, 498, 204.
50
37
PK +
7. Soto, C. Nat. Rev. Neurosci. 2003.
8. Gestwicki, J. E.; Marinec, P. S. Comb. Chem. High Throughput Screening 2007, 10,
667.
25
9. Haydar, S. N.; Yun, H. E. D.; Staal, R. G. W.; Hirst, W. D. Small-Molecule Protein–
Protein Interaction Inhibitors as Therapeutic Agents for Neurodegenerative
Diseases: Recent Progress and Future Directions. In Annual Reports in Medicinal
Chemistry; Elsevier Academic: San Diego, 2009; Vol. 44, p 51.
10. Blazer, L. L.; Neubig, R. R. Neuropsychopharmacology 2009, 34, 126.
11. Xu, Y.; Shi, J.; Yamamoto, N.; Moss, J. A.; Vogt, P. K.; Janda, K. D. Bioorg. Med.
Chem. 2006, 14, 2660.
Figure 2. Western blot of protease-digested ScGT1 cell lysates depicting the
presence or absence of prions (PrPSc) after treatment with 6a before (up) or after
(bottom) proteinase K (PK).
12. Caughey, B.; Baron, G. S. Nature 2006, 443, 803.
13. Soto, C.; Kascsak, R. J.; Saborio, G. P.; Aucouturier, P.; Wisniewski, T.; Prelli, F.;
Kascsak, R.; Mendez, E.; Harris, D. A.; Ironside, J.; Tagliavini, F.; Carp, R. I.;
Frangione, B. Lancet 2000, 355, 192.
14. Sellarajah, S.; Boussard, C.; Lekishvili, T.; Brown, D. R.; Gilbert, I. H. Eur. J. Med.
Chem. 2008, 43, 2418.
O
O
O
O
H
N
H
N
A
A
A
N
H
A
N
H
15. Bartolini, M.; Bertucci, C.; Bolognesi, M. L.; Cavalli, A.; Melchiorre, C.;
Andrisano, V. ChemBioChem 2007, 8, 2152.
16. Bolognesi, M. L.; Banzi, R.; Bartolini, M.; Cavalli, A.; Tarozzi, A.; Andrisano, V.;
Minarini, A.; Rosini, M.; Tumiatti, V.; Bergamini, C.; Fato, R.; Lenaz, G.; Hrelia,
P.; Cattaneo, A.; Recanatini, M.; Melchiorre, C. J. Med. Chem. 2007, 50, 4882.
17. Pawar, A. P.; Dubay, K. F.; Zurdo, J.; Chiti, F.; Vendruscolo, M.; Dobson, C. M. J.
Mol. Biol. 2005, 350, 379.
18. Yung, L.; Huang, Y.; Lessard, P.; Legname, G.; Lin, E. T.; Baldwin, M.; Prusiner, S.
B.; Ryou, C.; Guglielmo, B. J. BMC Infect. Dis. 2004, 4, 53.
19. Ghaemmaghami, S.; Ahn, M.; Lessard, P.; Giles, K.; Legname, G.; DeArmond, S.
J.; Prusiner, S. B. PLoS Pathog. 2009, 5, e1000673.
20. Gazit, E. Faseb J. 2002, 16, 77.
21. Lewell, X. Q.; Judd, D. B.; Watson, S. P.; Hann, M. M. J. Chem. Inf. Comput. Sci.
1998, 38, 511.
22. Fragmenter was used for molecular decomposition to fragments and R-groups,
JChem 5.0.0, 2008, ChemAxon (http://www.chemaxon.com).
A= Cl > Br
A= Cl > Br
Score=100
Score=88
Amine1
O
A
H
N
Amine1
H
N
A
N
H
O
Ester
O
Amine1
A= Cl > Br
Score=0
Score=90
Figure 3. Substructures identified from the synthesized library.