Three-pronged probes: High-affinity DNA binding with cap, β-alanines and oligopyrrolamides

Article abstract of DOI:10.1002/chem.201302972

TPP oligonucleotides: Hybridization probes that interrogate target sequences through base pairing, stacking on the terminus, and binding in the minor groove are presented (see figure). All subunits of the probes contribute to the target affinity, leading

Full text of DOI:10.1002/chem.201302972

DOI: 10.1002/chem.201302972
Three-Pronged Probes: High-Affinity DNA Binding with Cap,
b-Alanines and Oligopyrrolamides
Rꢀdiger Haug, Markus Kramer, and Clemens Richert*[a]
Sequence-specific recognition of DNA strands is most
minor groove) remained a challenge. Here we report three-
pronged probes (TPPs) that bind A/T-rich target strands
from three sides. The three-pronged approach results in ex-
ceptionally stable duplexes (DTm of up to +44.88C) and
high selectivity, even at the very terminus (DTm of À6.48C
for a terminal wobble base pair). Control experiments sug-
gest that all portions of the TPP can engage in target bind-
ing, including the linker.
A high-resolution structure of a duplex with the Uaq
cap[14] provided the starting point for the design of the
three-pronged probes (Figure 1). The terminus is tightly
capped by the anthraquinone moiety, and the uracil residue
of the linker is found in the minor groove. In fact, the O4
atom of U points straight into the groove, making it an ideal
attachment point for a minor groove binder (MGB). Since
the carbonyl oxygen of the pyrimidine cannot be reacted
itself without creating an undesirable lactim ether structure,
N4 of the corresponding cytidine derivative was chosen for
appending the MGB. A DNA sequence was selected for the
probes that matches the preference of the oligopyrrolamides
to bind A/T-rich duplexes.[15] Since individual b-alanine resi-
dues have been shown to be tolerated in minor groove bind-
ers,[16] we chose b-Ala-based linkers, hoping that they would
provide the necessary flexibility.
readily achieved through hybridization with a complementa-
ry probe strand, designed to bind following the Watson–
Crick base pairing rules.[1] But, this binding mode interrog-
ates only a very small fraction of the surface of a target
strand. A limited sequence fidelity results, with a small pen-
alty for mismatches in the duplex.[2,3] In contrast, active sites
of enzymes typically bind a large fraction of the surface of
substrates, and so do key proteins that act on DNA during
replication or repair.[4,5] Given the importance of sequence-
specific recognition of DNA, it is desirable to develop strat-
egies for more thoroughly interrogating target sequences
with synthetic probes, using a combination of base pairing,
interactions based on shape complementarity, and hydrogen
bonding in the grooves. The need for high-affinity probes is
particularly urgent for sequences rich in weakly pairing
bases (A and T) that are difficult to bind strongly with
DNA alone.
Substituents that increase the affinity of oligonucleotides
for target strands through additional stacking interactions
and/or preorganizing effects have been described.[6–8] Fur-
ther, elegant work exists on cyclic probes that use a combina-
tion of Watson–Crick and Hoogsteen base pairing[9] and that
target triplex-forming sequences. Two-pronged probes can
be constructed from oligonucleotides and a minor groove
binder.[10] Another approach combines base pairing with
stacking on the terminus of the duplex with the target. For
example, we have previously shown that non-nucleosidic
substituents at the termini of oligonucleotides can act as
“caps” that stabilize duplexes with canonical base pairs at
their termini.[11,12] Caps can give significant increases in af-
finity and base-pairing fidelity, but their effect fades with
the distance from the terminus. A probe featuring a stilbene
cap and a minor groove binder was prepared previously, but
it gave disappointing mismatch discrimination at the termi-
nus,[13] suggesting that cap and MGB did not act with proper
synergy in binding the target. So, a seamless recognition of
a target from three sides (Watson–Crick face, terminus, and
Scheme 1 shows key intermediates of the synthesis of
three-pronged probes 1–7. Oligoamides 8–10, featuring four
pyrroles and from one to four b-alanine residues, were syn-
[a] Dipl.-Chem. R. Haug, Dipl.-Chem. M. Kramer, Prof. C. Richert
Institut fꢀr Organische Chemie
Universitꢁt Stuttgart
70569 Stuttgart (Germany)
Fax : (+49)711-685-64321
E-mail: lehrstuhl-2@oc.uni-stuttgart.de
Figure 1. a) Principle of binding a DNA target strand with a three-
pronged probe through molecular interrogation from three sides. b)
Three-dimensional structure of the DNA duplex (ACGCG-Uaq)2 with
[14]
O4 of the uridine residue highlighted by an arrow. The anthraquinone
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201302972.
moiety acting as cap and the uridine residue acting as linker are shown in
red.
15822
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15822 – 15826

COMMUNICATION
thesized in solution, starting from 11 and active ester 12 or
methyl esters 11, 13 or 14 as building blocks, and a protocol
typical for Boc-based approaches.[17,18] Intermediate 15 was
also used directly for coupling to the probe. Additionally,
UV-Melting experiments were performed to study the
molecular recognition of target strand 20 (Table 1). With
aminoethyl-terminated C*Aq cap alone (5), the melting
point (Tm) increased by 11.58C, a value close to the 138C
measured earlier for the UAq cap on a decamer.[14] Upon
addition of the tetrapyrrole unit, the DTm values shot up to
38.3–44.88C, with the Tm increase levelling off at four b-Ala
residues (4:21). At this length, the minor groove binder con-
tributed 34.38C to the melting point increase, beyond the
11.58C provided by the cap alone. The DTm of 44.88C ob-
served for 4:21 is the largest melting point increase for any
terminal substituent of an oligonucleotide that we are aware
of. Further, the melting transition for the duplex of three-
pronged probe 4 with its target is highly cooperative
(Figure 2), a desirable feature for high fidelity hybridiza-
tion.[19]
terminally N-acetylated tetramer Ac
ACHTUNGTERN(NUGN b-Ala)4 as well as octa-
mer Ac(b-Ala)8 were prepared as controls (see Supporting
ACHTUNGTRENNUNG
Information for details of those and all other syntheses).
Next, we prepared the N4-aminoethylcytidine building
block. Phosphoramidite 16 was elaborated from uridine 17
in 10 steps and 22% overall yield via azidotriazolide 18 as
key intermediate. Automated chain assembly using commer-
cial 5’-phosphoramidites on a DNA synthesizer produced
support 19, which was smoothly extended with 16, followed
by deprotection of the aliphatic amine, and coupling of car-
boxylic acids 8–10, 15, or Alloc
removal of the Alloc moiety and acetylation. Two consecu-
tive coupling rounds with Alloc(b-Ala)4OH and acetylation
produced the probe with the Ac(b-Ala)8 chain (7). A two-
ACHTUNGTNER(NUGN b-Ala)4OH with subsequent
R
We then studied the components of 4 more closely. First,
probe strands containing only b-alanine residues were
tested. Four b-Ala residues (6:21) increased the Tm by 48C
over that of the cap-only version (5:21), and eight such resi-
dues by 8.38C (7:21). This increase of ~18C per b-alanine
AHCTUNGTRENNUNG
stage deprotection, with TBAF followed by NH4OH/
MeNH2, was used to cleave all protecting groups and release
the oligonucleotides from the support.
Scheme 1. Synthesis of three-pronged probes and control compounds. i) Synthesis cycle (repeated four times): 1) 12, DIPEA, 2) MeOH, HCl; ii) 1)
Ac2O, pyridine, 2) K2CO3 aq.; iii) 1) HBTU, DIPEA, 11, 13 or 14, 2) K2CO3 aq.; iv) 1) diphenyl carbonate, NaHCO3, 2) DMTCl, DMAP, pyridine, 3)
NaN3, [15]crown-5, 4) TBDMSCl, imidazole; v) POCl3, triazole, DIPEA; vi) AllocNHC2H4NH2, DIPEA; vii) 1) PPh3, H2O, 2) anthraquinone carboxylic
acid, HBTU, DIPEA; viii) 1) Cl3CCO2H, MeOH, 2) NCC2H4OP(N
8, 9, 10, 15, or Alloc-(bAla)4-OH, HBTU, DIPEA, (deprotection and coupling with Alloc-(bAla)4-OH twice for compound 7); 5) for compounds 5–7:
[Pd
(PPh3)4], Et2NH2+HCO3À; and then Ac2O, 2,6-lutidine, N-methylimidazole, 6) 1) TBAF, 2) NH4OH, MeNH2. (Compound 5 was formed simultane-
(PPh3)4], Et2NH2+HCO3À, 4)
ACHTUTGNRENGU(N iPr)2)2, DIPAT; ix) 1) 19, tetrazole, 2) I2, py, H2O, 3) [PdACHUTNGTRENNGUN
ACHTUNGTRENNUNG
ously under these conditions.) Abbreviations: Alloc=allyloxycarbonyl, Boc=tert-butyloxycarbonyl, Bt=benzotriazolyl, cpg=controlled pore glass,
DMT=4,4’-dimethoxytrityl, TBDMS=tert-butyldimethylsilyl.
Chem. Eur. J. 2013, 19, 15822 – 15826
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15823

C. Richert et al.
Table 1. UV-Melting points of duplexes of ATTATTAAAA (21) and the
probe strands given.
Probe
Melting point [8C][a]
DTm [8C][b]
Hyperchromicity [%][c]
20
1
2
3
4
5
6
7
19.9Æ1.1
58.2Æ0.8
62.3Æ0.4
64.6Æ0.2
64.7Æ0.4
31.4Æ0.8
35.7Æ0.6
39.2Æ0.7
–
37
33
31
36
39
24
30
27
38.3
42.4
44.7
44.8
11.5
15.5
19.8
[a] Average of 4–6 melting points at strand concentrations of 1.5 mm each,
10 mm phosphate buffer, pH 7, and 1m NaCl, with a heating or cooling
rate of 18C minÀ1. [b] Difference to melting point of unmodified duplex.
[c] Hyperchromicity at 260 nm.
Figure 3. Binding of a b-alanine oligomer alone to DNA. 1H NMR spec-
trum of DNA hairpin 5’-CCAG-HEG-CTG-3’ in the absence (a) or in
the presence of one equivalent of (bAla)6 (b), in H2O/D2O (9:1) at 2 mm
strand concentration 500 MHz and 208C. Arrows show signals most af-
fected by the binding of (bAla)6.
The G:T base combination resulting from this change
forms rather stable mismatched base pairs (wobble base
pairs)[20] that are difficult to suppress. The results from this
A/G scan of the duplex (Figure 4 and Table S1, Supporting
Information) showed strong mismatch discrimination at po-
sitions 1, 4, and 7 of the target strand of 4, but little at the
terminal residue (position 10) that is outside the range of
the polyamides, even in fully extended conformation. When
a wobble pair was introduced at the center of the sequence
(G4), the loss in affinity of three-pronged probe 4 for the
target was so strong that there was a noticeable break-down
of cooperativity (Figure 4a). This effect was not observed
for the control duplex (Figure 4b).
In conclusion, we note that the combination of Uaq cap,
b-alanine oligomer and oligopyrrolamide leads to very high
affinity for an all-A/T-target strand. The combination pro-
pels the melting point into a region usually reserved for se-
quences with a significant G/C content. This was unexpect-
ed. Individual residues of b-alanine, when incorporated in
polyamide minor groove binders, had been reported to pro-
mote binding in an extended conformation,[16b] but were not
known to be DNA binders in their own right. Neither the
cap alone, nor the minor groove binder with a minimal
linker, a b-alanine oligomer alone, or a combination of the
latter without the linker to the cap lead to the same level of
cooperativity and selectivity in binding. As with every
probe, a hybridization experiment requires proper choice of
incubation temperature. Without this, either false positives
(temperature too low) or false negatives (temperature too
high) will result. Three-pronged probes are more selective
in their sequence recognition and they give a temperature
optimum for weakly pairing sequences that is more similar
to that of typical mixed sequences, so that massively parallel
high fidelity detection of different DNA sequences may be
more readily achieved, using the concept of isostable du-
plexes and universally stringent conditions.[21]
Figure 2. UV-Melting curves of duplexes of strands 4, 5, 7, or 20 with
target strand 21. The cartoons highlight the components of each probe in
simplified form.
residue is less than the 2–48C increase per b-alanine ob-
served for the duplexes of 1/2/3/4 with 21, confirming how
well the combination of b-alanines and pyrroles cooperate
in stabilizing the duplex with the target. Next, we tested the
DNA-binding properties of a b-alanine oligomer in free
form, without a covalent link to a probe. An NMR titration
experiment with free (b-Ala)6 and the mixed-sequence
DNA hairpin 5’-CCAG-HEG-CTG-3’ (where HEG denotes
a hexaethyleneglycol linker) showed that the oligoamide
does indeed bind to DNA (Figure 3). Binding affects pro-
tons located at terminal residues as well as H2 of the central
A residue, located in the minor groove. The signal broaden-
ing suggests that there is no static complex, though, but
rather a dynamic binding situation when the oligoamide is
in its free form.
Finally, we performed exploratory experiments with
three-pronged probe 4 to study the level of mismatch dis-
crimination and to gain insights into the range of nucleo-
tides covered by the oligoamide portion. For this, every
fourth base of the target strand was “mutated” from an A
to a G (mismatched target strands 22–25).
We believe that the TPP probes are a step towards more
thorough interrogation of DNA target strands with synthetic
15824
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15822 – 15826

High-Affinity DNA Binding
COMMUNICATION
Acknowledgements
The authors are grateful to Drs. B. Claasen and E. Kervio for the acquisi-
tion and help with the interpretation of NMR results, H. Griesser for
a review of the manuscript and Dr. P. Grꢀnefeld for sharing experimental
results and discussions. This work was supported by DFG (grants No. RI
1063/9-1 and RI 1063/13-1).
Keywords: DNA · duplexes · hybridization · minor groove
binder · oligonucleotides
[1] J. D. Watson, F. H. Crick, Nature 1953, 171, 964–967.
[2] a) J. Gralla, D. M. Crothers, J. Mol. Biol. 1973, 78, 301–306; b) J.
SantaLucia, Jr., D. Hicks, Annu. Rev. Biophys. Biomol. Struct. 2004,
33, 415–440; c) W. W. Hadiwikarta, J. C. Walter, J. Hooyberghs, E.
Carlon, Nucleic Acids Res. 2012, 40, e138.
[3] a) K. Siegmund, U. E. Steiner, C. Richert, J. Chem. Inf. Comput. Sci.
2003, 43, 2153–2162; b) U. Plutowski, C. Richert, Angew. Chem.
2005, 117, 627–631; Angew. Chem. Int. Ed. 2005, 44, 621–625.
[4] A. Kornberg, T. A. Baker, in DNA Replication, 2nd ed., University
Science Books, Mill Valley, 2005, pp. 118–129.
[5] a) K. Hitomi, S. Iwai, J. A. Tainer, DNA Repair 2007, 6, 410–428;
b) J. J. P. Perry, E. Cotner-Gohara, T. Ellenberger, J. A. Tainer, Curr.
Opin. Struct. Biol. 2010, 20, 283–294.
[6] a) B. Vester, J. Wengel, Biochemistry 2004, 43, 13233–13241.
[7] a) K.-Y. Lin, M. D. Matteucci, J. Am. Chem. Soc. 1998, 120, 8531–
8532; b) F. Seela, G. Becher, Nucleic Acids Res. 2001, 29, 2069–
2078; c) H. Liu, J. Gao, S. R. Lynch, Y. D. Saito, L. Maynard, E. T.
Kool, Science 2003, 302, 868–871.
[8] See e.g.: a) R. W. Wagner, M. D. Matteucci, J. G. Lewis, A. J. Gutier-
rez, C. Moulds, B. C. Froehler, Science 1993, 260, 1510–1513; b) J. I.
Gyi, D. Gao, G. L. Conn, J. O. Trent, T. Brown, A. N. Lane, Nucleic
Acids Res. 2003, 31, 2683–2693; c) T. Kottysch, C. Ahlborn, F. Brot-
zel, C. Richert, Chem. Eur. J. 2004, 10, 4017–4028.
Figure 4. a) “G-Scan” analysis of the binding site of the oligoamide por-
tion of 4. a) UV-Melting curves of the duplex of 4 with 21, or 22–25 with
one mismatched G residue. Conditions: 1.5 mm strands, 10 mm phosphate
buffer, pH 7, and 1m NaCl. b) UV-melting curves of duplexes of control
strand 20 with target strands 21–24 under the same conditions, except
that the strand concentration was 18 mm in order to shift the curves into
the observable temperature range. Symbols used: * perfectly matched
^
sequence, & G1, ~ G4, G7, and + G10.
[9] E. T. Kool, Acc. Chem. Res. 1998, 31, 502–510.
[10] a) I. V. Kutyavin, I. A. Afonina, A. Mills, V. V. Gorn, E. A. Lukhta-
nov, E. S. Belousov, M. J. Singer, D. K. Walburger, S. G. Lokhov,
A. A. Gall, R. Dempcy, M. W. Reed, R. B. Meyer, J. Hedgpeth, Nu-
cleic Acids Res. 2000, 28, 655–661; b) J. Robles, S. B. Rajur, L. W.
McLaughlin, J. Am. Chem. Soc. 1996, 118, 5820–5821; c) A. N. Si-
nyakov, S. G. Lokhov, I. V. Kutyavin, H. B. Gamper, R. B. Meyer, J.
Am. Chem. Soc. 1995, 117, 4995–4996; d) E. Kawashima, Y. Naka-
nishi, Y. Terui, H. Tomitori, K. Kashiwagi, Y. Ohba, K. Kamaike,
Nucleosides Nucleotides Nucleic Acids 2013, 32, 196–205; e) S. B.
Rajur, J. Robles, K. Wiederholt, R. G. Kuimelis, L. W. McLaughlin,
J. Org. Chem. 1997, 62, 523–529; f) J. W. Szewczyk, E. E. Baird,
P. D. Dervan, Angew. Chem. 1996, 108, 1596–1598; Angew. Chem.
Int. Ed. Engl. 1996, 35, 1487–1489.
molecules. Enzymes that catalyze pivotal genetic processes,
such as polymerases, frequently bind DNA substrates from
several sides, including interactions in the minor groove and
from above the “terminus” where the incoming dNTP forms
a terminal base pair.[22] Further, proteins at a pivotal point
of genetic control, such as p53, show contacts in the major
and the minor groove in their complex with DNA.[23] In
mimicking proteins, the design of the TPPs builds on the
successful optimization of molecular caps for the termini
and the exceptionally well developed chemistry of minor
groove binders. The extension of our current work to TPPs
with MGBs that are specific for other target sequences[15] is
a natural next step.
[11] a) C. F. Bleczinski, C. Richert, J. Am. Chem. Soc. 1999, 121, 10889–
10894; b) J. Tuma, C. Richert, Biochemistry 2003, 42, 8957–8965;
c) S. Narayanan, J. Gall, C. Richert, Nucleic Acids Res. 2004, 32,
2901–2911; d) S. Egetenmeyer, C. Richert, Chem. Eur. J. 2011, 17,
11813–11827.
[12] a) R. Lucas, E. Vengut-Climent, I. Gꢃmez-Pinto, A. AviÇꢃ, R.
Eritja, C. Gonzꢄlez, J. C. Morales, Chem. Commun. 2012, 48, 2991–
2993; b) J. C. Morales, J. J. Reina, I. Dꢅaz, A. AviÇꢃ, P. M. Nieto, R.
Eritja, Chem. Eur. J. 2008, 14, 7828–7835; c) R. Lucas, I. Gꢃmez-
Pinto, A. AviÇꢃ, J. J. Reina, R. Eritja, C. Gonzꢄlez, J. C. Morales, J.
Am. Chem. Soc. 2011, 133, 1909–1916.
Experimental Section
[13] Z. Dogan, R. Paulini, J. A. Rojas Stꢀtz, S. Narayanan, C. Richert, J.
Am. Chem. Soc. 2004, 126, 4762–4763.
Melting curves of oligonucleotide duplexes were acquired at 260 nm with
a temperature gradient of 18C minÀ1. Synthetic protocols, NMR and
mass spectra, additional melting points and data on NMR experiments
with free (b-Ala)6 can be found in the Supporting Information.
[14] A. Patra, C. Richert, J. Am. Chem. Soc. 2009, 131, 12671–12681.
[15] P. B. Dervan, Bioorg. Med. Chem. 2001, 9, 2215–2235.
[16] a) J. M. Turner, S. E. Swalley, E. E. Baird, P. B. Dervan, J. Am.
Chem. Soc. 1998, 120, 6219–6226; b) A. R. Urbach, J. J. Love, S. A.
Chem. Eur. J. 2013, 19, 15822 – 15826
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15825

C. Richert et al.
Ross, P. B. Dervan, J. Mol. Biol. 2002, 320, 55–71; c) V. C. Rucker,
C. Melander, P. B. Dervan, Helv. Chim. Acta 2003, 86, 1839–1851.
[17] D. M. Chenoweth, D. A. Harki, P. B. Dervan, J. Am. Chem. Soc.
2009, 131, 7175–7181.
[21] C. Ahlborn, K. Siegmund, C. Richert, J. Am. Chem. Soc. 2007, 129,
15218–15232.
[22] S. Doublie, S. Tabor, A. M. Long, C. C. Richardson, T. Ellenberger,
Nature 1998, 391, 251–258.
[18] T. Hintermann, C. Mathes, D. Seebach, Eur. J. Org. Chem. 1998,
2379–2387.
[23] Y. Cho, S. Gorina, P. D. Jeffrey, N. P. Pavletich, Science 1994, 265,
346–355.
[19] K. Siegmund, C. Ahlborn, C. Richert, Nucleosides Nucleotides Nu-
cleic Acids 2008, 27, 376–388.
Received: July 28, 2013
[20] F. H. C. Crick, J. Mol. Biol. 1966, 19, 548–555.
Published online: October 29, 2013
15826
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15822 – 15826