Toward encoding reactivity using double-stranded DNA. Sequence-dependent native chemical ligation of DNA binding polyamides

Article abstract of DOI:10.1016/j.tet.2013.05.126

The coupling reaction between distamycin-related polyamides equipped with cysteine and thioester groups can be accelerated in the presence of double-stranded (ds) oligonucleotides with selected sequences. While the coupling reaction of reactive partners containing three pyrrole units is accelerated by dsDNAs containing ATTTTA or ATGTTA sites, the heterodimeric coupling between a tripyrrole and a polyamide equipped with two pyrroles and one imidazole, is accelerated by the latter DNA, but not by the other containing the A/T rich tract. These differences can be exploited for selecting preferred coupling pairs from a mixture of reactive monomers; therefore the reaction outcome depends on the instructions provided by the dsDNA sequence.

Full text of DOI:10.1016/j.tet.2013.05.126

Tetrahedron 69 (2013) 7847e7853
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Toward encoding reactivity using double-stranded DNA. Sequence-
dependent native chemical ligation of DNA binding polyamides
*
ꢀ
ꢀ
ꢀ
~
Adrian Jimenez-Balsa, Veronica I. Dodero, Jose L. Mascarenas
ꢀ
ꢀ
ꢀ
Departamento de Química Organica and Centro Singular de Investigacion en Química Bioloxica e Materiales Moleculares (CIQUS).
Universidade de Santiago de Compostela, 15782, Santiago de Compostela, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The coupling reaction between distamycin-related polyamides equipped with cysteine and thioester
groups can be accelerated in the presence of double-stranded (ds) oligonucleotides with selected
sequences. While the coupling reaction of reactive partners containing three pyrrole units is accelerated
by dsDNAs containing ATTTTA or ATGTTA sites, the heterodimeric coupling between a tripyrrole and
a polyamide equipped with two pyrroles and one imidazole, is accelerated by the latter DNA, but not by
the other containing the A/T rich tract. These differences can be exploited for selecting preferred coupling
pairs from a mixture of reactive monomers; therefore the reaction outcome depends on the instructions
provided by the dsDNA sequence.
Received 7 March 2013
Received in revised form 21 May 2013
Accepted 28 May 2013
Available online 4 June 2013
Keywords:
DNA binding
Native chemical ligation
Molecular recognition
Polyamides
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
regulation of gene transcription, also depend on the correct reading
of specific DNA sequences.6 However, in this case, the information
The processes of gene transcription and replication rely on the
DNA-templated polymerization of activated nucleotide units. The
template effect increases the effective molarity of the reactants,
therefore favoring the coupling reactions, and ensures that the
biopolymeric product exhibits the correct sequence.1 Inspired by
these natural systems, chemists have long pursued the use of DNA
templates to promote and control synthetic transformations of
non-biological reactants. Although the first examples on this topic
were published in the 1990s,2 it was not until recently that DNA-
template synthesis (DTS) has emerged as a reliable and powerful
tool.3 The applications of this technology are not restricted to the
field of chemical synthesis, but also reach other areas, such as re-
action discovery, drug development, and biosensing.4 Despite all
the advances, the methodology still suffers from some limitations,
such as the requirement of tethering the reactants to linear DNA
chains, which are usually not desired parts of the final products. On
the other hand, achieving catalytic turnover is certainly difficult
owing to the usually higher affinity of the products for the DNA
template.5
is encoded in the surface of the major and minor grooves in the
double-stranded DNA (dsDNA). Although nature does not use
dsDNAs for templating reactivity, the implicit recognition code
provided by dsDNA, together with its unique structural features,
suggest that it might offer an interesting potential for promoting
chemical transformations in a programmed manner. Up to now, the
use of double-stranded DNA as an instructed reactivity template
has been scarcely explored,7 and is essentially restricted to the
examples reported by Dervan’s and co-workers on a triplex helix-
mediated formation of a phosphodiester bond7a and the click di-
merization of six-ring hairpin polyamides that bind to contiguous
DNA sites.7b More recently, in the context of generating new DNA
binders, Nagatsugi et al. have reported the use of dsDNA-templated
click chemistry to generate acridineepeptide conjugates,
which display slightly higher DNA affinity than the monomeric
precursors.7c
Based on the well-known ability of distamycin-related poly-
amides to selectively bind to dsDNA sequences through antiparallel
2:1 insertion into the dsDNA minor groove,8 and the recognition
diversity that can be engineered by combining different aromatic
amino acids like N-methylpyrrole (Py) or N-methylimidazole (Im),9
we envisioned the possibility of using specific double-stranded
oligonucleotides to control and program the coupling between
designed heterocyclic polyamides.
Natural DNA-encoded processes are not restricted to those
controlled by WatsoneCrick base pairings in single stranded DNA
(ssDNA) chains. Other extremely relevant cellular events, like the
We have previously shown that it is possible to accelerate
a native chemical ligation (NCL) reaction between tripyrroles 2a
* Corresponding
author.
E-mail
address:
joseluis.mascarenas@usc.es
~
(J.L. Mascarenas).
0040-4020/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2013.05.126

ꢀ
A. Jimenez-Balsa et al. / Tetrahedron 69 (2013) 7847e7853
7848
and 3a, which are, respectively, equipped with a thioester and
a cysteine unit, by using a ds-oligonucleotide featuring a sequence
that favors their homodimeric insertion in the minor groove
(ATTTTA).10 On the basis of these results we decided to study the
effect of replacing one N-methylpyrrole unit of the polyamide by an
N-methylimidazole. Hypothetically, this replacement should affect
the DNA binding preferences of the resulting polyamide, and thus
modify the sequence of the ds-oligonucleotides that best promote
the reaction. This would represent a step forward toward encoding
reactivity with specific dsDNAs.
Herein we demonstrate the viability of this concept by showing
that the dsDNAs promoting the heterodimeric coupling are differ-
ent than those inducing the formation of tripyrrole homodimers.
Furthermore, we show that the reaction products of a mixture
containing three different reactants depend on the sequence of the
dsDNAs used as templates.
the introduction of this methyl substitution is convenient because
it increases the stability of the thioester toward hydrolysis, and
facilitates the subsequent coupling studies. The tripyrrole coun-
terpart, equipped with the nucleophilic cysteine (3a) was prepared
as shown in Scheme 1, by an initial alkylation of 1a with tert-butyl
(3-iodopropyl)carbamate, followed by deprotection, coupling of the
amine with a trityl cysteine amino acid, and removal of the trityl
group. If instead of coupling the cysteine to the intermediate amine,
this is treated with 5-oxo-5-(phenylthio)pentanoic acid, we obtain
the thioester 4a, which represents an interesting probe to test the
effect of the tether length in the coupling reaction.
In the case of the dipyrrole-imidazole system (ImPyPy), the
cysteine-equipped partner had to be synthesized using an alter-
native strategy (bottom part of Scheme 1), because in this case the
coupling of the iodo-propylamine to 1b was low yielding. The
compound was efficiently prepared by first introducing the side
chain in the dipyrrole, to give 5 (see Supplementary data), followed
by coupling of the imidazole, and introduction of the cysteine unit
in the pyrrole tether (Scheme 1, bottom).
2. Results and discussion
The NCL proceeds through an initial trans-thioesterification
between the electrophilic partner (thioester) and the nucleophile
(cysteine derivative) to give an intermediate (I), which evolves to
the final amide by means of an intramolecular acyl transfer
(Scheme 2).12
The design of our reactive units relies on the well-known in-
sertion mode of distamycin derivatives in the minor groove of
dsDNA, which places the methyl groups of the pyrroles facing the
solvent exposed surface of the complex.11 Therefore, considering
a 2:1 binding mode, these positions provide suitable linking points
to introduce the reactive groups required for the coupling reaction
(Fig. 1). The increased effective molarity provided by the template
should accelerate the coupling, provided that the tether exhibits
a suitable length and flexibility. As coupling reaction we chose
a native chemical ligation (NCL) as it is compatible with aqueous
media, and does not require additives or special pH or salt
conditions.12
In agreement with previous results,10 mixing the tripyrroles 2a
(5.0 mM) and 3a (5.0 mM) in a 200 mM NaCl, 20 mM phosphate
buffer containing 5 mM TCEP, at pH 7.0 and w22 ꢀC, does not
generate significant amounts of the coupling product after 20 min.
However, in the presence of 1 equiv of a hairpin ds-oligonucleotide
(A) featuring an ATTTTA sequence, there is a significant acceleration
in the reaction, as deduced from the HPLC traces of Fig. 2. Consid-
ering both the intermediate Iaa and the final product 7aa, we have
calculated an approximate conversion of 85e90% after 8 h.
To study the effect of different DNAs, we used more diluted
solutions (0.65
mM of the reactants, [phosphate]¼12 mM, [NaCl]¼
120 mM, [TCEP]¼3 mM, pH 7.0, and w22 ꢀC); and the results are
represented in Fig. 2b. In terms of initial rates, the reaction in the
presence of the DNA A is accelerated over 40 times with respect to
that in its absence. An oligonucleotide featuring a guanine in the
middle of the A/T tract (DNA B) is also capable of accelerating the
reaction, which is in consonance with the known ability of this type
of sequences to accept the antiparallel insertion of two polyamides
in its minor groove.8b Oligonucleotides that are known to favor 1:1
over 2:1 insertions, like DNA C are significantly poorer catalysts.
The reaction is also accelerated with sub-stoichiometric amounts of
the oligonucleotides while using 0.5 equiv of the oligo A (see
Supplementary data) generates almost similar results than with
1 equiv, further decrease in the amount of the catalyst leads to
slower reaction rate and yield.10
It was also interesting to observe that the thioester 4a, featuring
a longer linking tether, leads to inferior results (curve x, Fig. 2b),
suggesting a major entropic cost for the coupling reaction as
a consequence of the tether length.
Fig. 1. Model of the location of reactive groups X and Y introduced in the central
pyrrole of polyamides interacting with the DNA minor groove (built on a reported X-
ray structure: Pdb 378d).11
With these data at hand, we explored the heterodimeric
couplings between the tripyrroles (PyPyPy) and the polyamides
featuring an N-terminal N-methylimidazole instead of an
N-methylpyrrole (ImPyPy). As shown in the left part of Fig. 3, the
reaction between the tripyrrole-cysteine derivative 3a and the
imidazole-dipyrrole thioester 2b gives the corresponding in-
termediate Iab and reaction product 7ab, when carried out in the
presence of 1 equiv of the hairpin DNA B. The effects of the different
DNAs in the reaction rate are represented in Fig. 3b. It is important
to note that in this case, and in contrast to the coupling of two
tripyrroles, the reaction is accelerated only by the oligo B (ATGTTA,
over 23-fold rate enhancement) and not by the oligo A (ATTTTA),
which seems consistent with the known preferences of insertion of
The synthesis of the polyamide fragments was readily achieved
following the protocols indicated in Scheme 1. The required tri-
pyrrolic (PyPyPy) precursors 1a and 1b were assembled in four and
three steps (34% and 38% overall yields), respectively, from com-
mercially or readily available pyrrole and imidazole units (see
Supplementary data). Alkylation of these precursors with p-tolyl
2-bromoethanethioate gave the thioesters 2a and 2b, which would
be used as electrophilic partners in the NCL reaction. In our pre-
vious studies we had prepared a similar phenyl thioester that lacks
the methyl substitution at the phenyl ring; we have now found that

ꢀ
A. Jimenez-Balsa et al. / Tetrahedron 69 (2013) 7847e7853
7849
Scheme 1. Synthesis of the reactive partners. Reactions: (a) BrCH2COSPhMe, K2CO3, acetone/DMF, rt, 55% (for 2a), 35% (for 2b); (b) 1a, I(CH2)3NHBoc, K2CO3, acetone/DMF (6:1),
reflux, 38%; (c) TFA/CH2Cl2, 0 ꢀC to rt; (d) BoceCys(Trt)eOH, HATU, DIPEA, CH2Cl2, rt; (e) TFA/Et3SiH, CH2Cl2, 0 ꢀC to rt, 48% (three steps); (f) HOOC(CH2)3COSPh, HOBt, HBTU, DIPEA/
DMF, rt, 49%; (g) H2, Pd/C 10%; (h) 1-methyl-1H-imidazole-2-carboxilic acid, DECP, Et3N, DMAP, ꢁ10 ꢀC, 44% (two steps); (i) TFA/CH2Cl2, 0 ꢀC to rt; (j) BoceCys(Trt)eOH, HATU,
DIPEA, CH2Cl2, rt; (k) TFA/Et3SiH, CH2Cl2, 0 ꢀC to rt, 43% (three steps).
Considering the above results, which confirmed that in-
troducing the N-terminal imidazole unit in the pyrrole-based
polyamides changes the sequence of the dsDNAs templates that
best promote the coupling process; we wondered whether given
a mixture of different monomeric reactants, we could alter the
reaction profile by using different dsDNAs. In consonance with the
reaction rates previously observed, if we add 1 equiv of the hairpin
Scheme 2. The native chemical ligation (NCL) involves the initial formation of thio-
A (ATTTTA) to an equimolecular mixture of tripyrrole 3a and thio-
ester intermediate I.
7aa
(b)
(c)
(a)
3a
100
80
60
40
20
0
2a
Iaa
t= 90 min (DNA A)
t= 20 min (DNA A)
t= 20 min (no DNA)
t= 0
0
100
200
time (min)
300
400
12
10
14
Fig. 2. (a) HPLC trace of the reaction between 2a and 3a, in the absence and in the presence of oligonucleotide A. (b) Graphic showing the formation of NCL products (Iaaþ7aa) as
function of time in the reaction of 2a and 3a (0.65
M each, 12 mM phosphate, 120 mM NaCl, pH 7.0, 3 mM TCEP, w22 ꢀC; average of three experiments) in the absence of DNA (C),
and in the presence of DNAs A (:), B (-) and C (A). We also show the reaction of 3a and 4a in the presence of DNA A (ꢂ). (c) Hairpin oligonucleotides used for the study.
m
imidazole-containing polyamides in minor grooves featuring
a guanine step. We have also found that the alternative hetero-
dimeric coupling, using the Im(Py)2 unit as nucleophile (6b) and
the tripyrrole as electrophile (2a), exhibits a similar acceleration
profile in the presence of the different DNAs (see Supplementary
data). Therefore, although in the case of the heterodimer coupling
the rate increase induced by the DNA is less pronounced than for
the formation of homodimers, the reaction is only accelerated by
the DNA B.
esters 2a and 2b, we observe almost exclusively the formation of
the homodimeric product (Iaa plus 7aa, brown peaks in Scheme 3),
and only traces of the intermediate Iab (green peak). On the other
hand, if we add the DNA B (ATGTTA), in addition to the homodimer
we also observe the formation of significant amounts of the het-
erodimer (Iab and 7ab, green peaks). Although, of course, still far
from the objective of getting a complete change in selectivity in
function of the DNA inputs, these results demonstrate the viability
of the concept, and set the basis for future work toward this goal.

ꢀ
7850
A. Jimenez-Balsa et al. / Tetrahedron 69 (2013) 7847e7853
3a
7ab
(b)
(a)
100
2b
80
60
Iab
40
20
0
t= 90 min
t= 20 min
t= 0
0
80
160
240
320
400
time (min)
10
12
14
Fig. 3. (a) HPLC trace of the reaction between cysteine derivative 3a and thioester 2b in the presence of oligo B (conditions: 5.0
m
M of each reactant, 200 mM NaCl, 20 mM
M each reactant, 12 mM phosphate,
phosphate buffer at pH 7.0, 5 mM TCEP, w22 ꢀC). (b) Graphic showing the formation of NCL products (Iabþ7ab) as function of time (0.65
m
120 mM NaCl, pH 7.0, 3 mM TCEP, w22 ꢀC) in the absence of DNA (C), and in the presence of DNAs A (:), B (-), and C (A).
Scheme 3. Different outcome of the reaction of 3a with the thioester partners 2a and 2b, depending on the DNA promoter. Conditions: 0.65
mM of reactants, 12 mM phosphate
buffer, 120 mM NaCl, pH 7.0, 3 mM TCEP, 30 min of reaction, w22 ꢀC. Marked in brown the intermediate Iaa and product 7aa, and in green the corresponding products Iab and 7ab.
3. Conclusion
sequence accelerates both coupling reactions. Therefore, given
a mixture of the three coupling partners, the reaction outcome
is different depending on the sequence of the template DNA.
Although the strategy is yet to be optimized, our results suggest the
viability of using dsDNAs to encode reactivity. Potential advantages
of using dsDNA over ssDNA as reactivity templates stem from the
possibility of inducing asymmetric transformations because of the
intrinsic chirality of the dsDNA scaffold, the potential applications
for encoding synthetic processes in natural settings without the
In summary, we have demonstrated that the rate of an NCL
between appropriately functionalized DNA binding heterocyclic
polyamides can be considerably increased by using specific double-
stranded DNAs. While a dsDNA featuring an ATTTTA site promotes
homodimeric coupling of tripyrroles (PyPyPy), but not the forma-
tion of heterodimers between tripyrroles and a polyamide with two
pyrroles and one imidazole unit (ImPyPy), a dsDNA with an ATGTTA

ꢀ
A. Jimenez-Balsa et al. / Tetrahedron 69 (2013) 7847e7853
7851
requirement of separating the DNA strands, and the presumably
easier implementation of catalytic chemistry.
104.6 (CH); 68.7 (CH2); 65.2 (CH2); 53.3 (CH3); 38.5 (CH3); 25.2
(CH2); 22.9 (CH3); 20.5 (CH3). MS (ESI): [MH]þ calcd for
C33H41N8O5S¼661.2, found: m/z 661.2. HRMS ESI Q-TOF: [MH]þ
calcd for C33H41N8O5S¼661.2915, found: m/z 661.2919.
4. Experimental section
4.1. General
4.2.2. N2-[3-(Dimethylamine)propyl]-1-methyl-4-[(1-[2-oxo-2-(p-
tolylthio)ethyl]-4-{1-methyl-1H-imidazole-2-carboxamide}-1H-2-
pyrrolyl)carboxamide]-1H-2-pyrrolcarboxamide (2b). Polyamide 1b
(14.9 mg, 0.034 mmol) was added to a suspension of dry K2CO3
(70.1 mg, 0.508 mmol) in DMF (0.25 mL). A solution of S-p-tolyl 2-
bromoethanethioate (49.8 mg, 0.203 mmol) in 0.8 mL of dry ace-
tone was slowly added to the resulting solution and the mixture
was stirred for 15 min at rt. The solid was removed by filtration
through Celite, the filtrate was concentrated, and the resulting
residue was purified by RP-HPLC (H2O/CH3CN) to give 2b as an
amorphous brown solid (7.2 mg, 35%).
Dimethylformamide (DMF) and trifluoroacetic acid (TFA) were
from Scharlau, CH3CN from Merck, CH2Cl2 from Panreac, the
BoceCys(Trt)eOH from NovaBioChem, and the coupling agents
HOBt (1-hydroxybenzotriazole), HBTU (O-benzotriazole-N,N,N0,N0-
tetramethyluroniumhexafluorophosphate) and HATU (2-(1H-7-
azabenzotriazol-1-yl)-1,1,3,3-
tetramethyluroniumhexafluorophosphate methanaminium) from
GL Biochem (Shanghai). All other chemicals were purchased from
SigmaeAldrich or Fluka. All solvents were dry and synthesis grade,
unless specifically noted. Oligonucleotides were acquired from
Thermo-Fischer Sci.
1H NMR (250 MHz, CD3OD)
d (ppm): 7.26e6.75 (m, 10H); 4.82
(s, 6H); 4.65 (s, 2H); 3.98 (s, 3H); 3.86 (s, 3H); 3.36 (m, 2H); 3.21
Reactions were performed under Ar, unless specifically stated.
External bath thermometers were used to record all reaction
temperatures. Thin-layer chromatography (TLC) was performed on
(s, 2H); 3.15 (s, 3H); 1.98 (s, 2H). 13C NMR (62.9 MHz, CD3OD)
d
(ppm): 170.4 (C); 164.3 (C); 163.5 (C); 162.3 (C); 160.8 (C); 143.1
(C); 136.4 (CH); 132.0 (CH); 126.3 (C); 125.7 (C); 125.6 (C); 125.2
(CH); 123.3 (C); 123.2 (C); 121.5 (CH); 116.7 (CH); 106.1 (CH); 105.9
(CH); 104.6 (CH); 66.7 (CH2); 64.2 (CH2); 53.3 (CH3); 37.8 (CH3);
37.5 (CH3); 36.8 (CH2); 25.2 (CH2); 23.5 (CH3). MS (ESI): [MH]þ
calcd for C30H37N8O4S¼605.2, found: m/z 605.2. HRMS ESI Q-TOF:
[MH]þ calcd for C30H37N8O4S¼605.2653, found: m/z 605.2642.
F254 250-mm silica gel plates or 79e230 mesh aluminum oxide,
and components were visualized by observation under UV light, or
by treating the plates with ninhydrine or cerium nitrate followed by
heating.
Some reactions were followed by analytical reverse-phase HPLC
(High-Performance Liquid Chromatography) with an Agilent 1100
series LC/MSD (G1956A Quad VL model; ESI-quadrupole, positive
4.2.3. N2-[3-(Dimethylamine)propyl]-1-methyl-4-[(1-{3-(2-amine-
scan) using a Zorbax Eclipse XDB-C8 (5
m
m, 4.6ꢂ150 mm) analytical
3-mercaptopropanamide)
propyl}-4-{[1-methyl-4-(methylcarbox-
column from Agilent. Standard conditions for analytical RP-HPLC
consisted on a linear gradient from 5% to 95% of solvent B for
30 min at a flow rate of 1 mL/min (solvent A: water with 0.1% TFA,
solvent B: acetonitrile with 0.1% TFA). Compounds were detected by
UV at 220 and 304 nm.
NMR spectra were registered with Brucker WM-250, or with
Varian Inova 400 and Inova 500 spectrometers, at rt.
MS spectra were acquired using a Micromass Autospec (for CI
and FAB), an HP-5972-MSD (for EI), and both Bruker Biotof II and
micrOTOF (for ESI).
amide)-1H-2-pyrrolyl]carboxamide}-1H-2-pyrrolyl)carboxamide]-
1H-2-pyrrolcarboxamide (3a). Tripyrrole 1a (50.4 mg, 0.102 mmol)
was added to a suspension of dry K2CO3 (210 mg, 1.52 mmol) in
3 mL of dry acetone and DMF (0.5 mL). tert-Butyl (3-iodopropyl)
carbamate (173.8 mg, 0.610 mmol) was added, and the resulting
mixture was refluxed for 8 h. The solid was removed by filtration
through Celite, the filtrate was concentrated, and the resulting
residue was purified by RP-HPLC (H2O/CH3CN) to give an amor-
phous pale-brown solid (25 mg, 38%).
1H NMR (400 MHz, CD3OD)
d (ppm): 7.31e6.85 (m, 6H); 4.29 (m,
When necessary, HPLC purifications were performed in an
Agilent 1100 diode array HPLC system and using as column packing
Nucleosil 120-10 C-18 (250ꢂ8 mm) for semipreparative purifica-
tions. Mixtures of CH3CN (B) and H2O (A) with 0.1% TFA were used
as mobile phase with linear gradients. Detection was made at 220
and 304 nm simultaneously. Water was purified using a Millipore
MilliQ water purification system.
2H); 3.91 (s, 3H); 3.69 (s, 3H); 3.35 (m, 2H); 2.78 (m, 2H); 2.54 (m,
2H); 2.40 (m, 2H); 2.34 (s, 6H); 1.80 (s, 3H); 1.42 (m, 2H); 1.04 (s, 9H).
13C NMR (100.6 MHz, CD3OD)
d (ppm): 173.3 (C); 167.2 (C); 164.4 (C);
163.7 (C); 150.8 (CH); 136.1 (C); 135.6 (C); 126.2 (C); 126.1(C); 120.4
(CH); 109.6 (CH); 109.1 (CH); 107.9 (CH); 107.6 (CH); 79.8 (C); 61.2
(CH2); 48.2 (CH2); 47.6 (CH3); 41.0 (CH3); 39.8 (CH2); 39.7 (CH3); 37.2
(CH3); 31.6 (CH2); 31.1 (CH3); 28.2 (CH3); 25.8 (CH2). MS (ESI): [MH]þ
calcd for C32H48N9O6¼654.7, found: m/z 654.3.
4.2. Synthesis
The resulting product (25 mg, 0.043 mmol) was dissolved in
1 mL of 20% TFA/CH2Cl2 at 0 ꢀC and then stirred for 75 min and
allowed to reach rt. The solvents were removed under reduced
4.2.1. N2-[3-(Dimethylamine)propyl]-1-methyl-4-[(1-[2-oxo-2-(p-
tolylthio)ethyl]-4-{[1-methyl-4-(methylcarboxamide)-1H-2-pyrrolyl]
carboxamide}-1H-2-pyrrolyl)carboxamide]-1H-2-pyrrolcarboxamide
(2a). Tripyrrole 1a (38.0 mg, 0.077 mmol) was added to a suspen-
sion of dry K2CO3 (159 mg, 1.150 mmol) in DMF (0.65 mL). A solu-
tion of S-p-tolyl 2-bromoethanethioate (112.6 mg, 0.460 mmol) in
2 mL of dry acetone was added and the mixture stirred for 30 min at
rt. The solid was removed by filtration through Celite, the filtrate
was concentrated, and the resulting residue was purified by RP-
HPLC (H2O/CH3CN) to give 2a as an amorphous yellow-brown
solid (28 mg, 55%).
pressure and the resulting crude was re-dissolved in DIPEA (10.6
0.057 mmol) and 1 mL CH2Cl2 and cooled to 0 ꢀC. A solution of
HATU (18.2 mg, 0.048 mmol), DIPEA (8.8 L, 0.051 mmol), and
mL,
m
BoceCys(Trt)eOH (22.3 mg, 0.048 mmol) in 3 mL of CH2Cl2 at 0 ꢀC
was then added. The mixture was stirred for 30 min at 0 ꢀC and
then purified by RP-HPLC (H2O/CH3CN). Fractions containing the
desired product were lyophilized and re-dissolved in a solution of
Et3SiH (1 m
L) in 1 mL of 20% TFA/CH2Cl2 at 0 ꢀC. The mixture was
stirred for 45 min at rt, the solvents were removed under reduced
pressure, and the resulting residue was purified by RP-HPLC (H2O/
CH3CN) to give 3a as an amorphous brown solid (13.4 mg, 48%).
1H NMR (250 MHz, CD3OD)
d (ppm): 7.26e6.80 (m,10H); 4.63 (s,
6H); 4.41 (s, 2H); 3.80 (s, 3H); 3.60 (s, 2H); 3.31 (s, 2H); 3.10 (s, 6H);
1H NMR (500 MHz, CD3OD)
d (ppm): 7.00e6.82 (m, 6H); 3.89
2.27 (s, 3H); 2.02 (s, 3H); 0.93 (m, 2H). 13C NMR (62.9 MHz, CD3OD)
(m, 2H); 3.69 (s, 3H); 3.67 (s, 3H); 3.31 (br s, 5H); 2.92 (s, 6H); 1.94
d
(ppm): 192.5 (C); 171.3 (C); 164.2 (C); 161.5 (C); 160.8 (C); 136.1
(s, 3H); 1.82e1.75 (br s, 4H); 1.15e1.07 (m, 4H). 13C NMR
(C); 132.4 (CH); 131.9 (C); 131.0 (CH); 125.3 (C); 124.7 (C); 124.6 (C);
123.3 (C); 123.2 (C); 120.5 (CH); 114.7 (CH); 106.1 (CH); 105.9 (CH);
(125.8 MHz, CD3OD)
(C); 161.0 (C); 130.9 (CH); 129.7 (CH); 127.7 (CH); 125.5 (CH); 125.0
d (ppm): 171.3 (C); 169.2 (C); 164.7 (C); 164.5

ꢀ
A. Jimenez-Balsa et al. / Tetrahedron 69 (2013) 7847e7853
7852
(C); 124.4 (C); 123.3 (C); 122.9 (C); 63.9 (CH2); 63.6 (CH2); 56.6
(CH3); 52.1 (CH); 37.9 (CH2); 37.5 (CH2); 37.3 (CH3); 26.7 (CH2); 24.7
(CH2); 24.3 (CH2); 22.9 (CH3). MS (ESI): [MH]þ calcd for
C30H45N10O5S¼657.3, found: m/z 657.3. HRMS ESI Q-TOF: [MH]þ
calcd for C30H45N10O5S¼657.3290, found: m/z 657.3283.
1H NMR (500 MHz, CD3OD)
d (ppm): 7.49e6.89 (m, 6H); 4.48 (m,
2H); 4.13 (s, 3H); 4.05 (d, 1H, J¼6.8 Hz); 3.92 (s, 3H); 3.46 (t, 2H,
J¼6.8 Hz); 3.36 (m, 2H); 3.23 (t, 4H, J¼7.3 Hz); 2.95 (d, 6H,
J¼7.3 Hz); 2.06 (m, 2H); 1.40e1.30 (m, 2H). 13C NMR (125.8 MHz,
CD3OD)
d (ppm): 168.4 (C); 164.9 (C); 161.1 (C); 140.0 (C); 127.5
(CH); 127.1 (CH); 124.3 (C); 124.0 (C); 123.3 (C); 123.1 (C); 122.8 (C);
121.1 (CH); 120.0 (CH); 106.6 (CH); 106.5 (CH); 56.6 (CH2); 56.4
(CH); 49.1 (CH3); 47.2 (CH2); 43.5 (CH3); 37.9 (CH2); 37.0 (CH3); 36.7
(CH2); 36.2 (CH3); 32.4 (CH2); 26.5 (CH2); 26.3 (CH2). MS (ESI):
[MH]þ calcd for C27H41N10O4S¼601.3, found: 601.3; [M2H]þ calcd
for C54H82N20O8S2¼1202.6, found: m/z 1201.8. HRMS ESI Q-TOF:
[MH]þ calcd for C27H41N10O4S¼601.3027, found: 601.3029.
4.2.4. N2-[3-(Dimethylamine)propyl]-1-methyl-4-[(1-{3-(5-oxo-5-
[phenylthio]pentanamide)propyl}-4-{[1-methyl-4-(methylcarbox-
amide)-1H-2-pyrrolyl]carboxamide}-1H-2-pyrrolyl)carboxamide]-
1H-2-pyrrolcarboxamide (4a). The crude amine (21 mg,
0.034 mmol) resulting after removal of the Boc protecting group, as
described in the previous experiment, was re-dissolved in DMF
(0.7 mL) and 0.5 M DIPEA/DMF (68
added to a previously prepared solution of 5-oxo-5-((phenylthio)
oxy)pentanoic acid (11.33 mg, 0.0505 mmol) in 285 L of 0.5 M
mL). The resulting solution was
4.3. DNA-templated reactions
m
DIPEA/DMF containing HBTU (19.1 mg, 0.0505 mmol) and HOBt
(9.1 mg, 0.0674 mmol), and the mixture was stirred for 3 h at rt. The
crude residue after concentration was purified by RP-HPLC (H2O/
CH3CN) to give 4a as an amorphous brown solid (12.45 mg, 49%).
The concentration of the reactants was calculated by UV mea-
surements in water at 20 ꢀC in a Jasco V-630 spectrophotometer
coupled to an ETC-717 Peltier, using a standard UV Hellma micro
cuvette 108.002-QS (10 mm light path). Extinction coefficients ( )
3
1H NMR (500 MHz, CD3OD)
d
(ppm): 7.33e6.75 (m, 11H); 4.48 (br
were obtained for each compound in water:
3
¼34.000 Mꢁ1 cmꢁ1
304
s, 4H); 3.79 (m, 2H); 3.39 (m, 2H); 3.16 (s, 3H); 2.96 (s, 6H); 2.61 (m,
for tripyrrole substrates,
dipyrrole substrates, and
3
¼26.000 Mꢁ1 cmꢁ1 for imidazole-
304
2H); 2.17 (m, 2H); 2.11 (s, 3H); 1.97 (s, 3H); 1.93 (m, 2H); 1.87e1.83
3
304
¼60.000 Mꢁ1 cmꢁ1 for the hex-
(m, 2H); 1.10e1.06 (m, 2H). 13C NMR (125.8 MHz, CD3OD)
d
(ppm):
apyrrolic products or
products.
3
304
¼50.000 Mꢁ1 cmꢁ1 for the heterodimeric
135.7 (CH); 130.6 (CH); 130.3 (CH); 106.0 (CH); 63.3 (CH2); 51.6
(CH3); 43.5 (CH2); 37.0 (CH2); 36.9 (CH3); 35.7 (CH2); 30.7 (CH2); 30.0
(CH3); 22.5 (CH3); 22.4 (CH2); 22.1 (CH2). MS (ESI): [MH]þ calcd for
C38H50N9O6S¼760.3, found: m/z 760.3. HRMS ESI Q-TOF: [MH]þ
calcd for C38H50N9O6S¼760.3599, found: m/z 760.3593.
Reverse-phase HPLC analysis was performed using a Merck
LiChrospher WP 300 RP-18 (5
m
m, 250ꢂ4 mm) column. Binary
gradients of solvents A (H2O, 0.1% TFA) and B (CH3CN, 0.1% TFA)
were employed at a flow rate of 1.0 mL/min. The injection volume
was 200e100
mL. The HPLC peaks were detected by monitoring the
4.2.5. N2-[3-(Dimethylamine)propyl]-1-methyl-4-[(1-{3-(2-amine-
3-mercaptopropanamide) propyl}-4-{1-methyl-1H-imidazole-2-
carboxamide}-1H-2-pyrrolyl)carboxamide]-1H-2-pyrrolcarboxamide
(6b). A solution of 5 (50 mg, 0.096 mmol) in MeOH (2 mL) was
hydrogenated for 1.5 h over 10% palladium on charcoal (35 mg) at
rt. The catalyst was removed by filtration through Celite and the
filtrate concentrated. The residue was immediately dissolved in
DMF (1.2 mL) and cooled to 0 ꢀC. It was then slowly added (20 min)
to a solution of 1-methyl-1H-imidazole-2-carboxylic acid (14.0 mg,
0.110 mmol), DECP (0.07 mL, 0.482 mmol), Et3N (0.02 mL,
0.144 mmol), and DMAP (catalytic) in THF (1.0 mL) at ꢁ10 ꢀC. The
mixture was stirred for 13 h at ꢁ10 ꢀC. The solvents were removed
under reduced pressure and the resulting residue was purified by
RP-HPLC (H2O/CH3CN) to give the expected amide (25.4 mg, 44%).
UV absorbance at
by coupled MS spectrometry. Graphics were represented using the
program Kaleidagraph (v 3.5 by Synergy Software).
l
¼304 nm and their assignments were confirmed
The following methodology was applied on DNA-templated NCL
studies: All the reactions were repeated twice using conditions a to
identify final product and intermediate, and three times using
conditions b, with and without DNA (1 equiv), in both cases at rt
(w22 ꢀC) in phosphate buffer (pH 7.0) with NaCl. A typical exper-
iment consists on the preparation of a solution of the cysteine
containing compound (in 50 mM TCEP) and TCEP in the corre-
sponding buffer/water mixture. Then this mixture was stirred for
5 min at rt with or without the corresponding ds-oligonucleotide
and NCL was initiated by adding the thioester compound. Ali-
quots at different times were quenched on 0.1% TFA, frozen, and
kept at ꢁ20 ꢀC for analysis.
1H NMR (400 MHz, ketone)
d (ppm): 7.51e6.90 (m, 6H); 4.19 (m,
2H); 3.93 (s, 3H); 3.00 (s, 3H); 2.01 (d, 2H, J¼6.7 Hz); 1.40 (s, 6H);
1.36 (m, 2H); 1.34 (m, 2H); 1.23 (t, 4H, J¼6.7 Hz); 0.56 (s, 9H). 13C
ꢃ Conditions a: Final concentration of the species [thioester]¼
NMR (100.6 MHz, ketone)
d
(ppm): 210.5 (C); 164.4 (C); 131.5 (CH);
[cysteine derivative]¼[ADN]¼5.0 M, [TCEP]¼5.0 mM, [phos-
m
129.7 (C); 129.0 (C); 127.1 (C); 124.0 (C); 123.2 (C); 109.7 (CH); 109.5
(CH); 83.1 (C); 70.0 (CH2); 67.6 (CH2); 60.2 (CH2); 51.6 (CH2); 47.5
(CH3); 37.3 (CH2); 33.6 (CH3); 30.2 (CH2); 20.6 (CH3); 5.7 (CH3). MS
(ESI): [MH]þ calcd for C29H43N9O5¼598.3, found: m/z 598.3.
A solution of this amide (14.4 mg, 0.024 mmol) in 1 mL of 20%
TFA/CH2Cl2 at 0 ꢀC was stirred for 1.5 h and that temperature and
overnight at rt. The solvents were removed under reduced pressure
phate buffer]¼20 mM, [NaCl]¼200 mM in 1000
100 L aliquots were directly injected on HPLCeMS.
ꢃ Conditions b: slower kinetics. Final concentration of the species
[thioester]¼[cysteine derivative]¼[ADN]¼0.65 M, [TCEP]¼
3.0 mM, [phosphate buffer]¼12 mM, [NaCl]¼120 mM in
3000 L, pH 7.0. 200 L aliquots were taken at different times.
ꢃ Condition b was used to evaluate ds-oligonucleotide template
processes, yields were calculated using AbsUV
mL, pH 7.0.
m
m
m
m
and the resulting crude was re-dissolved in DIPEA (8.0
mL,
(lmax¼304 nm)
0.045 mmol) and 1 mL CH2Cl2 and cooled to 0 ꢀC. This solution was
added to a previously prepared solution of HATU (10.6 mg,
for coupling products (final product and intermediate) apply-
ing previously described extinction coefficients, and consider-
ing the sum of both peak areas, the thioester intermediate I and
the coupled products 7. The approximate rate of the NCL re-
action was calculated from the slope for initial times.
0.028 mmol), DIPEA (5.2
mL, 0.03 mmol), and BoceCys(Trt)eOH
(13 mg, 0.028 mmol) in 3 mL of CH2Cl2 at 0 ꢀC. The mixture was
stirred for 30 min at 0 ꢀC and then purified by RP-HPLC (H2O/
CH3CN). Fractions containing the desired product were lyophilized
and then dissolved in a solution of Et3SiH (5
m
L) in 1 mL of 20% TFA/
The coupled intermediates (I) and products (7) were identified by
LCeMS: Compound 7aa: ESI [MH]þ calcd for C56H77N18O10S¼1193.6,
CH2Cl2 at 0 ꢀC. The resulting mixture was stirred for 30 min
allowing to reach rt, the solvents were concentrated under reduced
pressure, and the crude was purified by RP-HPLC (H2O/CH3CN) to
give 6b as an amorphous yellow-brown solid (6.3 mg, 44%).
found¼1193.5. UV (H2O) lmax, nm
(
): 304 (60,000 Mꢁ1 cmꢁ1). Com-
3
pound 7aa0 (resulting by the coupling of 4a and 3a): ESI [MH]þ calcd
for C62H88N19O11S¼1306.7, found¼1307.0. UV (H2O) lmax, nm ): 304
(
3

ꢀ
A. Jimenez-Balsa et al. / Tetrahedron 69 (2013) 7847e7853
7853
(60,000
M
ꢁ1 cmꢁ1). HRMS ESI Q-TOF: [MH]þ calcd for
References and notes
C62H88N19O11S¼1306.6613, found¼1306.6592. Compound 7ab: ESI
[MH]þ calcd for C53H73N18O9S¼1137.6, found¼1136.6. UV (H2O) lmax,
1. Alberts, B.; Johnson, A.; Lewis, J.; Raff, M.; Roberts, K.; Walter, P. Molecular Bi-
ology of the Cell, 5th ed.; Garland Science: 2007, Chapters 5e7.
2. (a) Orgel, L. E. Acc. Chem. Res. 1995, 28, 109e118; (b) Luther, A.; Brandsch, R.; von
Kiedrowski, G. Nature 1998, 396, 245e248.
3. (a) Erben, A.; Grossmann, T. N.; Seitz, O. Bioorg. Med. Chem. Lett. 2011, 21,
4993e4997; (b) Li, X.; Liu, D. R. Angew. Chem., Int. Ed. 2004, 43, 4848e4870; (c)
He, Y.; Liu, D. J. Am. Chem. Soc. 2011, 133, 9972e9975.
4. (a) Chen, X.-H.; Roloff, A.; Seitz, O. Angew. Chem., Int. Ed. 2012, 51, 4479e4483;
(b) Grossmann, T. N.; Roglin, L.; Seitz, O. Angew. Chem., Int. Ed. 2008, 47,
7119e7122; (c) Grossmann, T. N.; Seitz, O. Chem.dEur. J. 2009, 15, 6723e6730.
5. Grossmann, T. N.; Strohbach, A.; Seitz, O. ChemBioChem 2008, 9, 2185e2192.
6. (a) Latchman, D. Gene Control; Garland Science, Taylor and Francis Group: New
York, London, 2010; (b) Latchman, D. Eukaryotic Transcription Factors, 5th ed.;
Elsevier/Academic Press: New York, 2008.
(
): 304 (50,000 Mꢁ1 cmꢁ1). HRMS ESI Q-TOF: [MH]þ calcd for
3
nm
C53H73N18O9S¼1137.5511, found¼1137.5440. Compound 7ba
(resulting by the coupling of 6b and 2a): ESI [MH]þ calcd for
C53H73N18O9S¼1137.6, found¼1138.8. UV (H2O) lmax,
(
3
): 304
nm
(50,000
M
ꢁ1 cmꢁ1). HRMS ESI Q-TOF: [MH]þ calcd for
C53H73N18O9S¼1137.5511, found¼1137.5522.
Acknowledgements
7. (a) Luebke, K. J.; Dervan, P. B. J. Am. Chem. Soc. 1989, 111, 8733e8735; (b) Poulin-
Kerstien, A. T.; Dervan, P. B. J. Am. Chem. Soc. 2003, 125, 15811e15821; (c) Imoto,
S.; Hirohama, T.; Nagatsugi, F. Bioorg. Med. Chem. Lett. 2008, 18, 5660e5663.
8. (a) Pelton, J. G.; Wemmer, D. E. Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 5723e5727;
We thank the financial support provided by the Spanish grants
SAF2010-20822-C02, CTQ2012-31341, CSD2007-00006, Consolider
Ingenio 2010 and the Xunta de Galicia GRC2010/12. A.J-B. thanks
ꢀ
(b) Dervan, P. B. Bioorg. Med. Chem. 2001, 9, 2215e2235; (c) Vazquez, M. E.;
ꢀ
the Spanish Ministerio de Educacion for his PhD fellowship.
~
~
Caamano, A. C.; Mascarenas, J. L. Chem. Soc. Rev. 2003, 32, 338e349; (d) Pazos,
ꢀ
~
E.; Mosquera, J.; Vazquez, M. E.; Mascarenas, J. L. ChemBioChem 2011, 12,
1958e1973.
Supplementary data
9. (a) Dervan, P. B.; Edelson, B. S. Curr. Opin. Struct. Biol. 2003, 13, 284e299; (b)
White, S.; Szewczyk, J. W.; Turner, J. M.; Baird, E. E.; Dervan, P. B. Nature 1998,
391, 468e471.
We present information on the synthesis of the precursors 1a, 1b,
and 5. We also include the graphics for the reaction between 2a and
6b and for the coupling between 2a and 3a using 0.5 equiv of DNA A.
Additional characterization data, such as MS spectra, are also in-
cluded. Supplementary data associated with this article can be found
in the online version, at http://dx.doi.org/10.1016/j.tet.2013.05.126.
~
10. Dodero, V. I.; Mosquera, M.; Blanco, J. B.; Castedo, L.; Mascarenas, J. L. Org. Lett.
2006, 8, 4433e4436.
11. Mitra, S. N.; Wahl, M. C.; Sundaralingam, M. Acta Crystallogr., Sect. D 1999, 55,
602e609.
12. (a) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H. Science 1994, 266,
776e779; (b) Hackeng, T. M.; Griffin, J. H.; Dawson, P. E. Proc. Natl. Acad. Sci. U.S.
A. 1999, 96, 10068e10073.