Synthesis of 1H-pyrrolo[3,2-h]quinoline-8-amine derivatives that target CTG trinucleotide repeats

Article abstract of DOI:10.1016/j.bmcl.2016.05.062

We describe a new molecular design, synthesis, and investigation of small molecules that bind to CTG trinucleotide repeats in DNA. 1H-Pyrrolo[3,2-h]quinoline-8-amine (PQA) has a tricyclic aromatic system with unique non-linear hydrogen-bonding surface complementary to thymine. We have synthesized a series of PQA derivatives with different alkylamino linkers. These PQAs showed binding to pyrimidine bulge DNAs and CNG (N = T and C) repeats depending on the linker structure, while quinoline derivatives lacking the pyrrole ring showed much lower binding affinity. PQA is a useful molecular unit for both CTG and CCG repeat binding.

Full text of DOI:10.1016/j.bmcl.2016.05.062

Accepted Manuscript
Synthesis of 1H-pyrrolo[3,2-h]quinoline-8-amine derivatives that target CTG
trinucleotide repeats
Jun Matsumoto, Jinxing Li, Chikara Dohno, Kazuhiko Nakatani
PII:
S0960-894X(16)30561-3
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.05.062
BMCL 23921
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
23 April 2016
16 May 2016
21 May 2016
Please cite this article as: Matsumoto, J., Li, J., Dohno, C., Nakatani, K., Synthesis of 1H-pyrrolo[3,2-h]quinoline-8-
amine derivatives that target CTG trinucleotide repeats, Bioorganic & Medicinal Chemistry Letters (2016), doi:
http://dx.doi.org/10.1016/j.bmcl.2016.05.062
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Synthesis of 1H-pyrrolo[3,2-h]quinoline-8-
amine derivatives that target CTG
trinucleotide repeats
Leave this area blank for abstract info.
Jun Matsumoto, Jinxing Li, Chikara Dohno, and Kazuhiko Nakatani

Bioorganic & Medicinal Chemistry Letters
Synthesis of 1H-pyrrolo[3,2-h]quinoline-8-amine derivatives that target CTG
trinucleotide repeats
a
a
a
a
Jun Matsumoto , Jinxing Li , Chikara Dohno , and Kazuhiko Nakatani *
a
The Institute of Scientific and Industrial Research, Osaka University, 8-1 Mihogaoka, Ibaraki 567-0047, Japan
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
We describe a new molecular design, synthesis, and investigation of small molecules that bind to CTG
trinucleotide repeats in DNA. 1H-pyrrolo[3,2-h]quinoline-8-amine (PQA) has a tricyclic aromatic system with
unique non-linear hydrogen-bonding surface complementary to thymine. We have synthesized a series of PQA
derivatives with different alkylamino linkers. These PQAs showed binding to pyrimidine bulge DNAs and
CNG (N = T and C) repeats depending on the linker structure, while quinoline derivatives lacking the pyrrole
ring showed much lower binding affinity. PQA is a useful molecular unit for both CTG and CCG repeat
binding.
Keywords:
Small molecule ligand
Mismatch base pair
CTG repeat
2
009 Elsevier Ltd. All rights reserved.
Myotonic dystrophy type 1 (DM1), the most common form of
muscular dystrophy presenting 1 in 8000 people, is one of more
than 30 inheritable disorders classified as trinucleotide repeats
(
TNR) disorders that are caused by aberrant expansion of triplet
1, 2
repeats of specific genes. DM1 is caused by the anomalous
expansion of CTG trinucleotide repeats in 3’-untranslated region
of dystrophia myotonica protein kinase (DMPK) gene. Normal
individuals have <37 CTG repeats, while DM1 patients carry
between 50 and many thousands of repeats sequence. The
expanded CTG is transcribed to CUG repeats, which cause toxic
3-8
gain of functions. Targeting the toxic RNA by oligonucleotides
and small molecular ligands has been a promising strategy to
8-14
alleviate the pathologic features.
Targeting CTG repeat DNA
could be another approach for the diseases, which could affect on
transcription step and repeat stability related to repeat expansion
15-17
and contraction.
Here, we report a series of new molecules
consisting of 1H-pyrrolo[3,2-h]quinoline-8-amine (PQA)
skeleton as a CTG repeat binding ligand. Binding assays
including thermal melting temperature, surface plasmon
resonance (SPR) and isothermal titration calorimetry (ITC)
measurement showed that the PQA derivatives bound to CTG
and CCG repeats but not to CAG and CGG repeats.
Figure 1. Molecular design of a new ligand targeting thymine. (a)
Hydrogen bonding between 2,6-diaminopyridine and thymine.
Hydrogen bond donor and acceptor are represented as D and A,
respectively. Secondary repulsive and attractive interactions are
represented by black and red arrows, respectively. (b) Structure of
PQA and possible hydrogen bonding motif with thymine. (c)
Simulated structure of the complex between PQA (green) and DNA
duplex containing T–T mismatch. The simulated structure shows that
PQA interacts with thymine and neighboring guanines. The hydrogen-
bondings and stacking interaction are represented by yellow and green
dashed line, respectively. (d) Energy-minimized structure of PQA–
thymine. The distances (Å) between two positively polarized donor
hydrogens are shown.
CTG repeats sequence was suggested to form hairpin
structure consisting of a tandem array of non-canonical T–T
18
mismatch base pair flanked by G–C base pairs. We have
developed a series of synthetic ligands for mismatched base pairs
19-22
in DNA.
The T–T mismatch could be a target in order to
develop small ligands for CTG repeats because T–T mismatch
specifically formed in the CTG repeats hairpin structure. To our
knowledge, there is few known small molecule that selectively
9, 23-25
and strongly binds to T–T mismatch.
The difficulty can be

Scheme 1. Synthesis of PQA
attributed to the alternating hydrogen-bonding motif of acceptor-
donor-acceptor (A-D-A) in thymine, which is known to form
weak base pair, because of the repulsive secondary interactions
between the positively polarized donor hydrogens and between
2
6-28
the negatively polarized atoms.
constant (K ) between thymine and 2,6-diaminopyridine having
A-D-A/D-A-D motif was reported to be 100 M , which is 100 to
000-fold lower than that of G–C base pair having A-D-D/D-A-
For example, association
A
–
1
1
A motif (Figure 1a). This can be rationally explained by the
presence of secondary repulsive force in the hydrogen-bonded
pair: 2,6-diaminopyridine–thymine have four repulsive
interactions, while G–C has two repulsive and two attractive
interactions..The limitation in A-D-A/D-A-D motif underscores
the need to develop a new motif for thymine recognition. We
here designed 1H-pyrrolo[3,2-h]quinoline-8-amine (PQA) as a
thymine recognition unit (Figure 1b). PQA is a tricyclic aromatic
heterocycle having a complementary hydrogen-bonding surface
to thymine. Non-linear arrangement of hydrogen bond donor and
acceptor in PQA will reduce the secondary repulsive interaction
by the longer distance, and the extended aromatic system will
increase the stacking interaction with the neighboring bases. We
carried out the molecular modeling simulations using
MacroModel in Maestro (Schrödinger). In this simulation, we
used DNA duplex containing a T–T mismatch as a host structure
in which one of thymine base is flipped out. The simulated
complex structure and the PQA–thymine pair in the complex are
shown in Figure 1c and d, respectively. The distance between
hydrogens of pyrrole N-H and thymine N3-H was 2.69 Å (Figure
Scheme 2. Synthesis of PQA derivatives with alkylamino linker.
3
’/3’-d(T CCA G_G CAA C)-5’). The effects of ligands were
calculated by the difference of the T (ΔT ) in the absence and
presence of each ligand (Table 1). The increase of T was
m
m
m
observed for all PQA derivatives 14–19 and indicated the
stabilization of DNA duplex by the binding of the ligands.
Especially, ligand 15 and 19 provided the highest ΔT of +4.7 ºC
m
and +5.2 ºC among the tested ligands (Table 1). The simulation
studies indicated that the linker in the ligand 14 is too short to
interact with the neighboring bases and phosphate groups. On the
other hand, terminal amino group in ligands 15–17 can reach to
the neighboring bases and phosphate groups, indicating the
linkers could contribute to the stabilization of the ligand binding.
1
(
d) and is longer than that in 2,6-diaminopyridine–thymine pair
2.34 Å, Figure 1a). Tricyclic system of PQA was well stacked
with neighboring G–C base pair in this binding pocket (Figure
c).
1
We adopted Leimgruber–Batcho indole synthesis for
construction of a key 1H-pyrrolo[3,2-h]quinoline structure
m
The high ΔT for ligand 15 may be explained by the shortest
29
(Scheme 1).
Starting from commercially available 7-
linker among the ligand 15–17 that provides the minimum loss of
conformational freedom upon the binding. The carbamide
linkage in the ligand 19 is favorable for the binding, which was
further verified by the SPR experiments described later. In
contrast, control quinoline derivatives lacking the pyrrole ring
methylquinoline 1, corresponding N-oxide 2 was obtained by
oxidation of 1 with 3-chloroperbenzoic acid. Chlorination of 2
with phosphoryl chloride provided chloroquinoline 3. 3 was
nitrated with nitric acid at 8 position to give 4. Buchwald–
Hartwig cross coupling of 4 with methylcarbamate followed by
hydrolysis of the carbamate with lithium hydroxide gave
aminoquinoline 6. 6 was reacted with N,N-dimethylformamide
dimethyl acetal to give a precursor of Leimgruber–Batcho indole
synthesis 7. The cyclization of 7 followed by deprotection with
hydrogen chloride to give 9 (PQA). In order to enhance
interaction of the PQA ligands with anionic DNA and solubility
in water, cationic alkylamino linker was attached on amino group
of PQA at position 8 (Scheme 2). PQA derivatives 14-17 have
an alkylamino liker via amide linkage with different number of
intervening methylenes, while an alkylamino linker was
introduced via carbamide linkage in 19. In order to investigate
the effect of pyrrole ring in PQA, control compounds 22 and 24
having a quinoline instead of PQA were synthesized.
m
(22 and 24) gave little T increase, indicating that the third
pyrrole ring in PQA has significant effect on the binding to the
T-bulge DNA.
To know the bulge base selectivity, we have conducted T
measurement of DNA duplex containing the other bulged base in
the presence of 15 or 19 (Table 2). C-bulge DNA exhibited ΔT
of +3.5 ºC and +5.8 ºC for 15 and 19, respectively, which are
comparable to T-bulge DNA. Since little ΔT was observed in A-
m
m
m
or G-bulge DNA and full-complementary DNA (Table 2), PQA
ligands apparently showed a selectively toward pyrimidine base
bulge.
Binding ability of synthetic ligands to thymine bulge was
investigated by measuring a melting temperature (T
m
) of DNA
duplex containing a thymine bulge (5’-d(A GGT CTC GTT G)-

m
Table 1. ΔT (ºC) values for DNA duplex containing a T-bulge in the
a
presence of ligands.
a
m
T values of duplexes (5’-d(A GGT CTC GTT G)-3’/3’-d(T CCA G_G CAA
C)-5’) (5.0 M)ꢀ in buffer (pH 7.0) containing 100 mM NaCl. All
measurements were taken three times, and standard deviations are shown in
b
c
parentheses. T
M).
m
values of duplex.ꢀ
m
T values in the presence of ligand (50

m
Table 2. ΔT (ºC) values for DNA duplexes containing a A, G, C-bulge or
a
full-complementary duplex.
Figure 2. SPR single cycle kinetic analysis of ligand binding to the d(CTG)
a) 19. (b) 15. (c) 17. (d) 24. Ligand was added stepwise at concentrations of
.63 M, 1.3 M, 2.5 M, 5.0 M, and 10 M.
9
.
(
0
a
m
T values of duplexes (5’-d(A GGT CNC GTT G)-3’/3’-d(T CCA G_G
CAA C)-5’) (5.0 M)ꢀ in buffer (pH 7.0) containing 100 mM NaCl. N is
either A, G, or C. Full-matched duplex is a 10-mer lacking X. All
measurements were taken three times, and standard deviations are shown in
b
c
parentheses. T
m
values of duplexes.ꢀ
m
T values in the presence of ligand (50

M).
Figure 3. (a) Isothermal titration calorimetry (ITC) for the binding of 19 to
d(CTG) at 25 ºC. Titration was conducted by adding 2  of 19 (300 Μ)
every 2 min into a buffer solution (10 mM sodium cacodylate pH7.0, 100
mM NaCl) containing d(CTG) (5 Μ). (b) The best fit curve for a single set
of identical binding sites model.
Investigations of the ligand-binding to CTG repeat sequence
were carried out by surface plasmon resonance (SPR) assay
Figure 2 and S2). 5’-Biotinylated DNA consisting of 9 CTG
repeats (5’-biotin-d(CTG) -3’) was immobilized to streptavidin-
coated (SA) sensor chip. Binding of the ligands to d(CTG)
repeat was analyzed by single cycle kinetics assay, where ligands
of different concentrations were injected stepwise to the sensor
chip without regeneration step of the surface. Sensorgrams for
binding of 19 and 15 are shown in Figure 2a and b. Ligand 19
9
(
9
9
9
The effects of linker and additional pyrrole ring on the binding
to d(CTG) repeat were studied by SPR assay. Figure 2c showed
9
SPR sensorgrams for ligand 17. Higher binding ability of 19 than
that of 17 can be attributable to the carbamide linkage, because
the alkylamino linker in 17 consists of the same number of atoms
to the linker in 19 but is connected by an amide linkage instead
of carbamide linkage. The rigid linker enhance the binding likely
due to the extended hydrogen bonding surface and possible
increase of stacking interaction. Comparison between 19 and 24
clearly showed the importance of the PQA skeleton for the
binding (Figure 2d). In order to examine the stacking interaction
between the PQA chromophore and neighboring bases, UV
binds to the d(CTG)
constant (K ) of 20 M. Among the tested ligands, ligand 19
showed the highest response units and the lowest apparent K
value. We also conducted isothermal titration calorimetry (ITC)
measurement for the binding of 19 to d(CTG) repeat (Figure 3).
The binding was analyzed with estimation of 4:1 stoichiometry
of ligand to d(CTG) repeat, since possible hairpin structure of
d(CTG) consists of four CTG/CTG sites. Fitting the isotherm
with a single set of identical binding sites model gave K of 17
M, which is in good agreement with the SPR analysis.
9
repeat with an apparent dissociation
D
D
9
9
9
9
titration experiments of the ligand 19 with d(CTG) repeat were
performed (Figure S1). The ligand 19 exhibits an absorption band
at 349 nm, which was red-shifted by 8 nm upon titrating with the
D

9
d(CTG) repeat, indicating stacking interaction of the PQA with
neighboring base pairs (Figure S1a). In contrast, 2,6-
diaminopyridine did not show any apparent spectral changes
under the same condition (Figure S1c). 2,6-Diaminopyridine is
capable of hydrogen-bonding to thymine, but is unfavorable for
binding to the repeat DNA. The extended tricyclic -system
and/or additional hydrogen bond donor in the pyrrole ring
contributed to the enhanced binding ability of 19.

References and notes
1
2
3
.
.
.
McMurray, C. T. Nat. Rev. Genet. 2010, 11, 786–799.
Mirkin, S. M. Nature 2007, 447, 932–940.
Tian, B.; White, R. J.; Xia, T.; Welle, S.; Turner, D. H.; Mathews,
M. B.; Thornton, C. A. RNA 2000, 6, 79–87.
4
5
.
.
Teplova, M.; Patel, D. J. Nat. Struct. Mol. Biol. 2008, 15, 1343–
1
351.
de Haro, M.; Al-Ramahi, I.; De Gouyon, B.; Ukani, L.; Rosa, A.;
Faustino, N. A.; Ashizawa, T.; Cooper, T. A.; Botas, J. Hum. Mol.
Genet. 2006, 15, 2138–2145.
6
7
.
.
Fernandez-Costa, J. M.; Garcia-Lopez, A.; Zuñiga, S.; Fernandez-
Pedrosa, V.; Felipo-Benavent, A.; Mata, M.; Jaka, O.; Aiastui, A.;
Hernandez-Torres, F.; Aguado, B.; Perez-Alonso, M.; Vilchez, J.
J.; Lopez de Munain, A.; Artero, R. D. Hum. Mol. Genet. 2013,
2
2, 704–716.
Pettersson, O. J.; Aagaard, L.; Jensen, T. G.; Damgaard, C. K.
Nucleic Acids Res. 2015, 43, 2433–2441.
8
9
.
.
Chau, A.; Kalsotra, A. Dev. Dyn. 2015, 244, 377–390.
Arambula, J. F.; Ramisetty, S. R.; Baranger, A. M.; Zimmerman,
S. C. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 16068–16073.
1
1
0. Jahromi, A. H.; Nguyen, L.; Fu, Y.; Miller, K. A.; Baranger, A.
M.; Zimmerman, S. C. ACS Chem. Biol. 2013, 8, 1037–1043.
1. Wong, C. H.; Nguyen, L.; Peh, J.; Luu, L. M.; Sanchez, J. S.;
Richardson, S. L.; Tuccinardi, T.; Tsoi, H.; Chan, W. Y.; Chan, H.
Y. E.; Baranger, A. M.; Hergenrother, P. J.; Zimmerman, S. C. J.
Am. Chem. Soc. 2014, 136, 6355–6361.
Figure 4. SPR single cycle kinetic analysis of 19 binding to the d(CNG)
repeats. (a) d(CCG) , (b) d(CAG) , and (c) d(CGG) . Ligand 19 was added
9
9
9
9
stepwise at concentrations of 0.63 M, 1.3 M, 2.5 M, 5.0 M, and 10 M.
12. Parkesh, R.; Childs-Disney, J. L.; Nakamori, M.; Kumar, A.;
Wang, E.; Wang, T.; Hoskins, J.; Tran, T.; Housman, D.;
Thornton, C. A.; Disney, M. D. J. Am. Chem. Soc. 2012, 134,
4
731–4742.
We carried out SPR single cycle kinetics assay for the other
1
1
1
3. Warf, M. B.; Nakamori, M.; Matthys, C. M.; Thornton, C. A;
Berglund, J. A. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 18551–
18556.
4. Siboni, R. B.; Bodner, M. J.; Khalifa, M. M.; Docter, A. G.; Choi,
J. Y.; Nakamori, M.; Haley, M. M.; Berglund, J. A. J. Med. Chem.
CNG repeats (Figure 4a–c). Ligand 19 binds to the d(CCG)
repeat with an apparent K of 33 M which is comparable to that
for d(CTG) (K = 20 M). In contrast, the binding to d(CAG)
and d(CGG) repeats are undetectable under the same conditions.
9
D
9
D
9
9
2
015, 58, 5770–5780.
5. Hagihara, M.; He, H.; Nakatani, K. ChemBioChem 2011, 12,
686–1689.
16. Nakamori, M.; Gourdon, G.; Thornton, C. A. Mol. Ther. 2011, 19,
222–2227.
7. Pearson, C. E.; Nichol Edamura, K.; Cleary, J. D. Nat. Rev. Genet.
005, 6, 729–742.
8. Petruska, J.; Arnheim, N.; Goodman, M. F. Nucleic Acids Res.
996, 24, 1992–1998.
These results indicated that ligand 19 bound selectively to CNG
repeats containing pyrimidine-pyrimidine mismatch base pairs.
These data are consistent with the T data showing preferable
m
1
2
binding to T and C-bulges (Table 2). PQA derivatives bound to
the pyrimidine bulges and mismatches most likely because the
tricyclic system fit to the space opposite the pyrimidine base in
terms of size and shape. On the other hand, hydrogen-bonding
recognition by PQA is not high enough to distinguish between
thymine and cytosine bases. PQA was designed to form three
hydrogen bonds to thymine, but two or three hydrogen bonds can
be involved in the binding to the unpaired cytosine (Figure S3).
To append higher selectivity between CTG and CCG repeats,
further optimization in hydrogen bonding recognition is required.
1
1
2
1
19. Nakatani, K.; Hagihara, S.; Goto, Y.; Kobori, A.; Hagihara, M.;
Hayashi, G.; Kyo, M.; Nomura, M.; Mishima, M.; Kojima, C. Nat.
Chem. Biol. 2005, 1, 39–43.
2
2
2
0. Peng, T.; Dohno, C.; Nakatani, K. Angew. Chem. Int. Ed. 2006,
5, 5623–5626.
4
1. Kobori, A.; Horie, S.; Suda, H.; Saito, I.; Nakatani, K. J. Am.
Chem. Soc. 2004, 126, 557–562.
2. Kobori, A.; Nakatani, K. Bioorg. Med. Chem. 2008, 16, 10338–
In conclusion, we have designed 1H-pyrrolo[3,2-h]quinoline
PQA) as a recognition unit for thymine and synthesized a series
1
0344.
23. Lian, C.; Robinson, H.; Wang, A. H.-J. J. Am. Chem. Soc. 1996,
18, 8791–8801.
(
1
of PQA derivatives targeting CTG repeat. Among the tested
ligands, PQA ligand containing carbamide linkage showed the
highest binding ability to T-bulge DNA and CTG repeat
sequence. Tricyclic system in PQA is a useful molecular unit for
both CTG and CCG repeat binding, but further optimization is
required to distinguish between CTG and CCG repeat.
2
4. David, A.; Bleimling, N.; Beuck, C.; Lehn, J. M.; Weinhold, E.;
Teulade-Fichou, M. P. ChemBioChem 2003, 4, 1326–1331.
5. Zhao, C.; Rajendran, A.; Dai, Q.; Nishizawa, S.; Teramae, N.
Anal. Sci. 2008, 24, 693–695.
2
26. Murray, T. J.; Zimmerman, S. C. J. Am. Chem. Soc. 1992, 114,
010–4011.
4
2
7. Quinn, J. R.; Zimmerman, S. C.; Del Bene, J. E.; Shavitt, I. J. Am.
Chem. Soc. 2007, 129, 934–941.
8. Ong, H. C.; Arambula, J. F.; Rao Ramisetty, S.; Baranger, A. M.;
Zimmerman, S. C. Chem. Commun. 2009, 668–670.
Acknowledgments
2
This work was supported by JSPS KAKENHI Grant-in-Aid
for Specially Promoted Research (26000007) for KN.
29. Siu, J.; Baxendale, I. R.; Ley, S. V. Org. Biomol. Chem. 2004, 2,
160–167.
Supplementary Material
Supplementary data associated with this article can be found, in
the online version, at XXX.