Modulating the rate of a native ligation coupling between tripyrrole derivatives by using specific dsDNA sequences

Article abstract of DOI:10.1021/ol061551i

The intrinsic recognition code associated with dsDNA allows either accelerating or retarding of a native chemical ligation reaction between tripyrrole ligands. The rate changes most probably stem from the sequence-dependent characteristics of the dsDNA-ligand complexes.

Full text of DOI:10.1021/ol061551i

ORGANIC
LETTERS
2006
Vol. 8, No. 20
4433-4436
Modulating the Rate of a Native Ligation
Coupling between Tripyrrole Derivatives
by Using Specific dsDNA Sequences
Vero´nica I. Dodero,† Manuel Mosquera,‡ Juan B. Blanco,† Luis Castedo,† and
Jose´ L. Mascaren˜as*,†
Departamento de Qu´ımica Orga´nica, UniVersidad de Santiago de Compostela, 15782
Santiago de Compostela, Spain, and Departamento de Qu´ımica F´ısica, UniVersidad de
Santiago de Compostela, 15782 Santiago de Compostela, Spain
qojoselm@usc.es
Received June 23, 2006
ABSTRACT
The intrinsic recognition code associated with dsDNA allows either accelerating or retarding of a native chemical ligation reaction between
tripyrrole ligands. The rate changes most probably stem from the sequence-dependent characteristics of the dsDNA ligand complexes.
−
The biosynthesis of nucleic acids and proteins is a highly
regulated cellular process that requires an appropriate transla-
tion of the information stored in double-stranded DNA
(dsDNA).1 It would be of great interest to use the intrinsic
code provided by DNA to control the synthesis of chemical
entities other than those occurring in nature.2 In this context,
it has been recently shown that it is possible to use single-
stranded DNA sequences to template a variety of syntheti-
cally relevant transformations of small organic compounds.2a,3
The strategy relies on the covalent attachment of the reacting
units to appropriate complementary oligonucleotides that,
upon hybridization, increase the effective molarity of the
reagents and thereby promote their reaction.
The use of double-stranded instead of single-stranded DNA
to control chemical reactions is more challenging and much
less established. Indeed, to our knowledge, only a couple of
examples of dsDNA-promoted chemical transformations
have been reported,4 one based on a triplex recognition
strategy4a and the other relying on a dipolar cycloaddition
ligation between two six-ring hairpin polyamides capable of
binding to contiguous DNA sites.4b These strategies require
the use of stoichiometric amounts of the catalyst as a
† Departamento de Qu´ımica Orga´nica.
‡ Departamento de Qu´ımica F´ısica.
(1) Alberts, B.; Johnson, A.; Lewis, J.; Raff, M.; Roberts, K.; Walter, P.
Molecular Biology of the Cell, 4th ed.; Garland Science: New York, 2002;
Chapter 7, p 375.
(2) For general reviews on DNA-templated synthesis, see: (a) Li, X.;
Liu, D. Angew. Chem., Int. Ed. 2004, 43, 4848-4870. (b) Ja¨schke, A.;
Seelig, B. Curr. Opin. Chem. Biol. 2000, 4, 257-262. For reviews on DNA
enzymes, see: (c) Silverman, S. C. Org. Biomol. Chem. 2004, 2, 2701-
2706. (d) Breaker, R. R. Nat. Biotechnol. 1997, 15, 427-431. (e) Peracchi,
A. ChemBioChem 2005, 6, 1316-1322.
(3) (a) Summerer, D.; Marx, A. Angew. Chem., Int. Ed. 2002, 41, 89-
90. (b) Kanan, M. W.; Rozenman, M. M.; Sakurai, K.; Snyder, T. M.; Liu,
D. R. Nature 2004, 431, 545-549.
10.1021/ol061551i CCC: $33.50
© 2006 American Chemical Society
Published on Web 09/09/2006

consequence of the high affinity of the product for the DNA
template. This limitation might be overcome if the reaction
system generates products that bind to the DNA template
with moderate affinity and not more strongly than the
reactants. We envisaged that the well-known ability of
distamycin-related analogues to recognize the minor groove
of specific DNA sequences with different affinities and
stoichiometries (1:1 or 2:1)5,6 could provide new, unique
opportunities to advance toward such a goal.
Scheme 1
Here, we report preliminary results in this area which
demonstrate the feasibility of modulating the rate and
efficiency of a “native chemical ligation” (NCL)7 between
tripyrrole peptide derivatives, by using specific dsDNA
sequences.
The available structural data on the interaction of dista-
mycin derivatives with DNA suggested that the N-methyl
groups of the pyrroles, which protrude out of the minor
groove, could be appropriate sites to introduce the required
reactive moieties.5,6 Therefore, we focused our efforts on the
coupling between tripyrrole derivatives 2 and 3, compounds
in which short alkyl chains containing the reactive thioester
and cysteine units are incorporated into the middle pyrrole
of the tripeptides. At the outset of the work, we were aware
that the similar nature of the DNA recognition unit of these
compounds might hamper a rigorous analysis and interpreta-
tion of the results, mainly because of the presumable
formation of competing homodimeric and heterodimeric
complexes in the presence of dsDNA. However, the synthetic
accessibility of these compounds, which can be straightfor-
wardly prepared from the same precursor 1 (Scheme 1),8
advised their use as initial probes to check the feasibility of
the concept.
The NCL coupling should proceed through an initial trans
thioesterification between the electrophilic partner (thioester
2) and the nucleophile (cysteine derivative 3) to give the
corresponding intermediate 4, which would then rearrange
to the final amide 5 (Scheme 2).7
In theory, a dsDNA sequence capable of bringing 2 and 3
into proximity in a 2:1 interaction mode might facilitate the
coupling process, whereas perhaps other sequences that could
favor the formation of homodimers or other types of
unproductive complexes might retard the reaction. Unfor-
tunately, it is difficult to predict which sequences would favor
homo- or heterodimeric assemblies.
The ligation reaction between 2 and 3 in the absence of
DNA was studied using several types of buffers and in the
presence of different concentrations of the reacting partners
Scheme 2
(4) (a) Luebke, K. J.; Dervan, P. B. J. Am. Chem. Soc. 1989, 111, 8733-
8735. (b) Poulin-Kerstein, A. T.; Dervan, P. B. J. Am. Chem. Soc. 2003,
125, 15811-15821. (c) An example of the use of dsDNA as a chiral
auxiliary in a Cu-catalyzed reaction has recently been reported: Roelfes,
G.; Feringa, B. L. Angew. Chem., Int. Ed. 2005, 44, 3230-3232.
(5) For reviews on pyrrole polyamide DNA recognition, see: (a) Dervan,
P. B. Bioorg. Med. Chem. 2001, 9, 2215-2235. (b) Dervan, P. B.; Edelson,
B. S. Curr. Opin. Struct. Biol. 2003, 13, 284-299. (c) Wemmer, D. E.
Biopolymers 2001, 52, 197-211.
(6) (a) Coll, M.; Fredrick, C. A.; Wang, A. H. J.; Rich, A. Proc. Natl.
Acad. Sci. U.S.A. 1987, 84, 8385-8389. (b) Pelton, J. G.; Wemmer, D. E.
Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 5723-5727. (c) Pelton, J. G.;
Wemmer, D. E. J. Am. Chem. Soc. 1990, 112, 1393-1399. (d) Dwyer, T.
J.; Geierstanger, B. H.; Bathini, Y.; Lown, J. W.; Wemmer, D. E J. Am.
Chem. Soc. 1992, 114, 5911-5919. (e) Chen, F.; Sha, F. Biochemistry 1998,
37, 11143-11151.
(7) (a) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H.
Science 1994, 266, 776-779. (b) Tam, P.; Lu, Y.-A.; Liu, C.-F.; Shao, J.
Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 12485-12489. (c) Yeo, D. S.;
Srinivasan, R.; Chen, G. Y. J.; Yao, S. Q. Chem.-Eur. J. 2004, 10, 4664-
4672.
(8) For the synthesis of tripyrrole polyamides containing alkyl side chains
other than methyl, see, for instance: Satz, A. L.; Bruice, T. C. Acc. Chem.
Res. 2002, 35, 86-95 and references cited therein.
4434
Org. Lett., Vol. 8, No. 20, 2006

(see the Supporting Information). The results of these studies
showed that the reaction can be accomplished in an aqueous
phosphate buffer at pH 7.5, with its efficiency being quite
sensitive to the ionic strength of the medium, so that small
increases in the [Na+] lead to higher reaction rates. We also
observed that hydrolysis of the thioester is a significant
competitive process, although this can be minimized by
carrying out the reaction at pH 7.0.
To study the effect of DNA on the coupling, it was
convenient to use conditions that induce a slow reaction: 12
mM phosphate buffer containing 120 mM NaCl and 5 mM
triscarboxyethylphosphine (TCEP) at pH 7.0 and equimolar
amounts of 2 and 3 (0.65 µM each). These conditions induce
only a 5% conversion to 5 after 4 h. Remarkably, the addition
of hairpin dsDNA A (1 equiv) to the reaction medium
promoted a substantial acceleration of the coupling process,
as can be deduced from the HPLC traces in Figure 1 (see
Figure 2. (a) Graphic showing the product (4 + 5) formation as
a function of time in the reaction between 2 and 3 (0.65 µM each,
12 mM phosphate buffer, 120 mM NaCl, pH 7.0, 5 mM TCEP, 25
°C) in the absence of DNA (x), in the presence of 1 equiv of dsDNA
A (O), dsDNA B (0), dsDNA A with 1.5 equiv of 5 (+), or a
single chain of DNA B (]) (both strands gave similar results), or
in the presence of 0.5 equiv of dsDNA A (2). (b) Graphic showing
the product (4 + 5) formation as a function of time in the absence
of DNA (x) or in the presence of 1 equiv of dsDNA A (O), dsDNA
C (b), dsDNA D (0), or dsDNA E (9). (c) Sequences of the
oligonucleotides used for the study.
Figure 1. HPLC traces of the reaction between equimolar mixtures
of 2 and 3 (0.65 µM each, 12 mM phosphate buffer, 120 mM NaCl,
pH 7.0, 5 mM TCEP, 25 °C): (1) initial time; (2) in the absence
of DNA after 40 min; (3) in the presence of 1 equiv of
ds-oligonucleotide A after 40 min. Peaks were assigned by LC-
MS.
extent that, in the presence of 5 equiv, the reaction stops
altogether (see the Supporting Information). This result is
consistent with a model in which the presence of excess DNA
lowers the amount of the productive heterodimeric complex.10
Using 0.5 equiv of hairpin A, the reaction is slightly slower
but leads to an approximate yield of 70% after 24 h, which
suggests the existence of some turnover, even considering
the background reaction. Interestingly, addition of 1.5 equiv
of 5 to the standard reaction mixture leads only to a weak
decrease in the reaction rate and efficiency (curve +, Figure
2a), which is in accordance with a relatively modest affinity
of this product for dsDNA A.11
also Figure 2a).
The reaction in the presence of this dsDNA proceeds near
40 times faster than in its absence (considering initial rates),
leading to an approximate conversion of 74% after 4 h
(Figure 2a). A 24bp ds-oligonucleotide B, which lacks the
T-rich hairpin unit, led to the same results as those with ds-
oligo A. The acceleration is certainly a consequence of the
presence of the double helical DNA because a single-stranded
DNA containing the same sequence (either chain of duplex
B) did not affect the reaction rate.
Moreover, in contrast to the reaction in the absence of
DNA, in which the thioester intermediate 4 was not detected,
in the DNA-promoted couplings this intermediate was clearly
observed, a result consistent with a reduced conformational
freedom of 4 when sited under the constraints of the DNA
groove.9
Other dsDNAs that might allow the formation of 2:1
complexes,6 such as C, also accelerate the coupling process,
although in this case the reaction was only 20 times faster
than that in the absence of DNA. dsDNAs with sequences
(9) A similar effect has been observed previously in native chemical
ligations promoted by peptides: (a) Severin, K.; Lee, D. H.; Kennan, A.
J.; Ghadiri, M. R. Nature 1997, 389, 706-709. (b) Kennan, A. J.; Haridas,
V.; Severin, K.; Lee, D. H.; Ghadiri, M. R. J. Am. Chem. Soc. 2001, 123,
1797-1803.
(10) Huc, I.; Pieters, R. J.; Rebek, J. J. Am. Chem. Soc. 1994, 116,
10296-10297.
(11) Using circular dichroism titrations we have calculated an ap-
proximate Kd of 3 µM (see the Supporting Information).
Analysis of the influence of the amount of dsDNA catalyst
A on the reaction showed that the rate increases on raising
the DNA concentration from 0.2 to 1 equiv; however, further
addition of the oligonucleotide induces a rate decrease to an
Org. Lett., Vol. 8, No. 20, 2006
4435

conversions when carried out in the presence of dsDNA G,
which features a couple of mismatched base pairs with
respect to its parent duplex F.13 Remarkably, addition of
excess urea after 30 min from the start restored the coupling
progress (Figure 3a).14 Circular dichroism titration experi-
ments revealed that although either tripyrrole (2 or 3) binds
dsDNA F or G as homodimers in a 2:1 mode the stability
of the complexes with the mismatched oligo G is higher
(∼400 times higher for 3 and 260 times higher for 2; see
the Supporting Information), which suggests that the reaction
retardation might be related to the capability of this
mismatched DNA to trap the reagents in unproductive
homodimeric complexes.15
Figure 3. (a) Graphic showing the product (4 + 5) formation as
a function of time in the reaction between 2 and 3 (10 µM each,
20 mM phosphate buffer, 200 mM NaCl, pH 7.0, 5 mM TCEP, 25
°C) in the absence of DNA (x) or in the presence of 2 equiv of
dsDNA F (4), mismatched dsDNA G (2), or single-stranded H
(b). The discontinuous curve represents the reaction progress in
the presence of G after addition of urea (8 M, 33 min from the
beginning) (]). (b) Sequences of the ds-oligonucleotides used.
In conclusion, we have demonstrated that the rate and
efficiency of a native chemical ligation reaction between two
tripyrrole units can be altered by using short dsDNAs with
specific sequences. The sequence-dependent rate changes are
most probably a consequence of the different complexation
abilities of the dsDNAs, albeit a definitive explanation must
wait for further studies. The fact that the dsDNA-accelerated
reaction can also be achieved in the presence of oversto-
ichiometric amounts of the product suggests that an ap-
propriate refinement of the system might lead to the
discovery of truly catalytic dsDNA-promoted reactions.
known to favor a 1:1 interaction mode with distamycin,6e
such as D, accelerate the reaction by less than 5 times
compared to the background and lead to low conversions
(10% after 4 h, Figure 2b), as does a noncognate DNA
template such as E. Therefore, these results confirm that the
rate and efficiency of the coupling reaction depend on the
dsDNA sequence. The observed differences might arise from
the relative amount of productive DNA-heterodimer com-
plexes present in the reaction mixture, but a definitive
explanation must wait for further studies.
We also considered the viability of retarding the reaction
by using dsDNAs that could particularly favor the formation
of unproductive complexes. To this end, we studied the effect
of adding 2 equiv of different dsDNAs to a reaction mixture
containing 10 µM 2 and 3 in 20 mM phosphate buffer and
200 mM NaCl, conditions in which the intrinsic coupling
proceeds relatively fast.12
Acknowledgment. This work was supported by the
Spanish Ministry of Education and Science (SAF2004-
01044) and the Xunta de Galicia (PGIDIT04TF209006PR).
V.I.D. thanks the Xunta de Galicia for a visiting professor-
ship, and J.B.B. thanks the Spanish Ministry for a predoctoral
fellowship.
Supporting Information Available: Synthetic proce-
dures, characterization data, experimental data on the dif-
ferent factors that influence the coupling reaction, and
circular dichroism titrations. This material is available free
of charge via the Internet at http://pubs.acs.org.
OL061551I
(13) Circular dichroism analysis indicates that mismatched DNA G keeps
the double helical structure (see the Supporting Information).
(14) A control experiment demonstrated that the excess of urea does
not alter the intrinsic coupling reaction.
Screening of several ds-oligonucleotides allowed us to
discover that the reaction is slower and leads to poorer
(15) UV-damaged dsDNAs with widened minor grooves show high
affinity for 2:1 complexes with distamycin A: Inase, A.; Kodama, T. S.;
Sharif, J.; Xu, Y.; Ayame, H.; Sugiyama, H.; Iwai, S. J. Am. Chem. Soc.
2004, 126, 11017-11023.
(12) In the absence of dsDNA, under these reaction conditions, thioester
2 does also participate in secondary processes and thereby it is almost
completely consumed after 1 h.
4436
Org. Lett., Vol. 8, No. 20, 2006