A Janus-Wedge DNA triplex with A-W1-T and G-W2-C base triplets

Article abstract of DOI:10.1021/ja804607v

A new type of double-stranded DNA targeting format by formation of a Janus-Wedge (J-W) triple helix is described. The wedge residue W₁ is used for A-T and T-A base pairs while W₂ is used for G-C and C-G base pairs. Both wedge residues are attached to a PNA backbone that is designed to insert the probe strand into double-stranded DNA and base pair with both Watson-Crick faces. To study the stability of such an assembly, we have examined the formation of the J-W triplex with various sequences. Copyright

Full text of DOI:10.1021/ja804607v

Published on Web 09/11/2008
A Janus-Wedge DNA Triplex with A-W1-T and G-W2-C Base Triplets
Han Chen, Meena, and Larry W. McLaughlin*
Department of Chemistry, Merkert Chemistry Center, Boston College, 2609 Beacon Street,
Chestnut Hill, Massachusetts 02467
Received June 17, 2008; E-mail: mclaughl@be.edu
We describe here the development of a Janus-Wedge (J-W) triple
helix involving target A-T and G-C base pairs. Each base triplet is
formed from a target base pair and a third residue (a wedge residue)
capable of hydrogen bonding with the Watson-Crick faces of the
base pairing partners.
Triplexes as described by Dervan and Helene are formed from a
(third) DNA strand capable of binding to the Hoogsteen face of purine
residues of purine-pyrimidine base pair.1,2 With that design the third
strand is bound in the major groove of duplex DNA. The Dervan/
Helene approach can be very useful but is generally limited to the
targeting of specific polypurine sequences1,2 (with the formation of
T-A-T and C+-G-C base triplets). To generalize duplex targeting, a
mode of recognition must be developed whereby all four possible base
pairs can be targeted. Many improvements over the Dervan/Helen
targeting design have been reported using derivatives that employ only
a single hydrogen bond,3 recognition of each base pair as a unit,4 or
Figure 1. Janus-Wedge base triplets: (a) the third-strand residue W1
being able to bind purines in either strand,5 but improvements in
targeting methods6-8 have achieved only moderate successes.
The Janus-Wedge (J-W) triple helix is based upon a recognition
motif first suggested by Lehn9 in his work with heterocycles; it involves
the ability of the incoming third strand to hydrogen bond with the
Watson-Crick (W-C) faces of the two target strands (see Figure 1).
In the present study, we have used W1 to target A-T (or by simple
rotation of the heterocycle, T-A) base pairs as we described earlier,10
and W2 for G-C (or C-G) base pairs. We have used a neutral PNA
backbone for the heterocycles but have incorporated one lysine (K)
residue at the C-terminus to aid in aqueous solubility and to provide
some complementary charge-charge interactions. The choice of a PNA
backbone follows the observation that certain PNA sequences are
known to bind DNA duplexes to form D-loop structures, a form of
strand invasion that could be an important intermediate in the pathway
to a J-W triplex.11
binds to the W-C faces of the target A and T or (b) W2 binds similarly
to G and C.
to generate the aminotriazenedione ring involved attack of a secondary
amine on a CBz protecting group.13 Fmoc deprotection unmasked a
primary amine; we reasoned this amine could attack the remaining
CBz protecting group (forming a 10-membered ring, see Supporting
Information). Mass spectral analysis confirmed this possibility. Replac-
ing that CBz protecting group with trimethyl-phenylsulfonyl (Scheme
1) eliminated the putative side reaction, and PNA sequences containing
W1 and W2 could be prepared.
We prepared a mixed 8-mer PNA sequence with a terminal lysine
residue: (W1)3W2(W1)4K, purified it (HPLC), and confirmed its
mass. In previous work10 we had also prepared the (W1)8K PNA
sequence and showed the binding orientation of the heterocycles
and the polarity of the J-W PNA relative to the target strands. We
We previously described10 the synthesis of the protected W1
monomer from the readily available 2,4-diaminopyridin-6(5H)-one.
A similar approach to the W2 monomer (Figure 1b) from the readily
available 6-amino-1,3,5-triazine-2,4(1H,3H)-dione failed, largely owing
to the very poor solubility of the starting heterocycle. The solution to
this synthesis was to use a cyclization reaction previously reported
for this class of compounds.12 The pathway involves the preparation
of an acyclic amidinourea,13 but that product is unstable and cyclizes
to form the aminotriazenedione with the amino group protected as the
CBz carbamate. We could use this same reaction using the amine-
containing linker (that bridging the triazene and PNA backbone), and
after cyclization couple the resulting product heterocycle to the PNA
backbone to generate the desired protected monomer (Scheme 1).
In principle this derivative should be ideal for the synthesis of PNA
sequences containing W2 residues, and in fact an efficient coupling
by the CBz-protected W2 monomer to a solid-phase bound W1 residue
was observed. However, upon piperidine deprotection of the bound
W2 residue we could not detect the presence of the expected primary
amine necessary to continue elongation. The initial cyclization reaction
Scheme 1a
a Steps: (i) benzyloxycarbonyl isocyanate; (ii) HCl·H2NCH2COOtBu,
Et3N f 3 (tBu ester of 4); (iii) TFA (iv) [2-(9H-fluoren-9-ylmethoxycar-
bonylamino)ethyl-amino]-acetic acid tBu ester, HCl/EDCI/DIPEA,f 5 (tBu
ester of 6); (v) TFA.
9
13190 J. AM. CHEM. SOC. 2008, 130, 13190–13191
10.1021/ja804607v CCC: $40.75
2008 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Thermal Stabilities of J-W Triplexes
Table 1. Thermal Stabilities of J-W Triplexes
-X X X X X X X X-
TM
X-Y
wedge strand
TM
-C C C C C C C C-
-C A C C C C C C-
-C A C C C A C C-
-C A C C C A C A-
-A A A A C A A A-
46.9, 69.8 ((0.3) °C
44.3, 69.1 ((0.4)
34.2, 67.6 ((0.7)
19.9, 67.2 ((0.7)
T-C
T-A
T-C
G-C
W1W1W1W1W1W1W1W1K
W1W1W1W1W1W1W1W1K
W1W1W1W1W2W1W1W1K
W1W1W1W1W2W1W1W1K
45.8, 69.2 ((0.5) °C
44.9, 69.1 ((0.5)
36.4, 70.1 ((0.3)
46.9, 69.8 ((0.3)
-
78.7
also prepared two DNA sequences in which T-C was used for the
target for W1 and G-C was used for the target of W2 (Table 1).14
The observed TM values for the (W1)8K probe and the T-C target
was 45.8 and 69.2 °C. The latter transition corresponds to the
duplex-random coil equilibrium and the former characterizes the
triplex-duplex equilibrium. Introduction of one W2 residue into
the probe strand should represent a mismatch condition for the all
T-C target, and in fact the TM characterizing the triplex-duplex
equilibrium in this case decreased by roughly 9 to 36.4 °C (Table
1). When the T-C target at the W2 site was changed to the canonical
G-C base pair, the target for W2, the TM value characterizing the
triplex-duplex equilibrium rose by over 10 to 46.9 °C (Table 1).
It is noteworthy that placing a single W2 residue into a sequence
that targets T-C residues results in a significant decrease in TM value
and that thermal stability is recovered when the G-C target is place
to interact with the W2 residue. An examination of potential
hydrogen bonding interactions for the T-W2-C base triplet suggests
the formation of an essentially isomorphic base triplet (Figure 2),
but the existence of such a base triplet does not contradict the data.
The measured TM values simply indicate a loss in complex thermal
stability when T-C is the target for W2. A more destabilizing
mismatch situation would likely result in complete dissociation of
the triplex. The reduction in thermal stability likely arises from
effects including (i) the tridentate interaction with G versus a
bidentate interaction with T, (ii) reduced base stacking effects for
a pyrimidine versus a purine, and (iii) unfavorable dipole-dipole
interactions (Figure 2).
Replacement of various C residues in the target sequences with
A residues results in binding to an increased number of A-T base
pairs by the W1 residues (Table 2).14 The temperature for first
transitions for these complexes decreased with increasing A content,
likely reflecting the formation of base triplets with a total of five
hydrogen bonds (see Figure 1a) in place of the T-W1-C base triplets
containing six hydrogen bonds. When the target sequence contained
only A-T and G-C base pairs, the PNA 8-mer was unable to form
a stable Janus-Wedge triplex. Presumably the eight-residue probe
sequence is simply not long enough to effectively invade the target
duplex and form the stable triplex when competing duplex formation
is also possible.
triplex as -15.3 kcal/mol (50 mM NaCl, 10 mM MgCl2). While
this value reflects effective binding interactions through the J-W
format, it will be moderated when strand invasion is required.
Binding of the PNA sequence containing the W2 residue to the
target DNA containing a G-C base pair (Table 1) was also
confirmed by nondenaturing PAGE analysis, which resulted in a
KD of 12 nM (Figure 3).
Figure 3. Gel shift assay for (W1)4W2(W1)3K and DNA (G-C, Table 1).
Initial studies with a longer (W1)14K sequence targeting an A14-
T14 duplex have indicated the formation of a stable complex on
the basis of thermal melting analysis and a fluorescence-based assay
in which an A-T selective minor groove-binding fluorophore is
displaced from the target duplex upon complex formation with the
J-W strand (see Supporting Information).
To target all four base pairs we need to design derivatives that
can discriminate all four base pairs. We expect that these residues
will exhibit both base-pair selectivity and enhanced stability in part
as the result of better base-stacking between the J-W residues.
Acknowledgment. This work was supported by the NSF (MCB
0451448).
Supporting Information Available: Synthetic schemes, procedures,
and thermal analyses. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) Moser, H. E.; Dervan, P. B. Science 1987, 238, 645–50.
(2) Francois, J. C.; Saison-Behmoaras, T.; Helene, C. Nucleic Acids Res. 1988,
16, 11431–40.
(3) Griffin, L. C.; Dervan, P. B. Science 1989, 245, 967–71.
(4) Gianolio, D. A.; McLaughlin, L. W. Nucleosides Nucleotides 1999, 18,
1751–69.
(5) Li, J.-S.; Fan, Y.-H.; Zhang, Y.; Marky, L. A.; Gold, B. J. Am. Chem. Soc.
2003, 125, 2084–93.
The stability of the Dervan/Helene (T-A-T)10 triplex is reported
to be -6.6 kcal/mol under optimal conditions (2 M NaCl, 50 mM
MgCl2).15 We have reported10 the stability of the (T-W1-C)8 J-W
(6) Horne, D. A.; Dervan, P. B. J. Am. Chem. Soc. 1990, 112, 2435–37.
(7) Jayasena, S. D.; Johnston, B. H. Nucleic Acids Res. 1992, 20, 5279–88.
(8) Gowers, D. M.; Fox, K. R. Nucleic Acids Res. 1999, 27, 1569–77.
(9) Branda, N.; Kurz, G.; Lehn, J. M. Chem. Commun. 1996, 2443–44.
(10) Chen, D.; Fraley, A. W.; Meena; Sharma, S. K.; McLaughlin, L. W. J. Am.
Chem. Soc. 2004, 126, 70–71.
(11) Nielsen, P. E.; Egholm, M.; Berg, R. H.; Buchardt, O. Science 1991, 254,
1497–1500.
(12) Yuan, C.; Williams, R. M. J. Am. Chem. Soc. 1997, 119, 11777–84.
(13) Yuan, C.; Williams, R. M. Tetrahedron Lett. 1996, 37, 1945–48.
(14) Errors in Tables 1 and 2 are reported for the first transitions.
(15) Pilch, D. S.; Brousseau, R.; Shafer, R. H. Nucleic Acids Res. 1990, 18,
5743–50.
Figure 2. Proposed structure of a T-W2-C base triplet.
JA804607V
9
J. AM. CHEM. SOC. VOL. 130, NO. 40, 2008 13191