Synthesis and evaluation of a conformationally constrained pyridazinone PNA-monomer for recognition of thymine in triple-helix structures

Article abstract of DOI:10.1016/j.bmcl.2003.12.093

A novel conformationally constrained pyridazinone E^ag-base PNA-monomer 2 capable of binding thymine in a triplex motif was designed and synthesised. A bis-PNA with the E^ag-base incorporated in the Hoogsteen strand was hybridised with a complementary DNA. Thermal stability studies revealed an increase in T_m (4.3°C per mod.) compared to a no-base unit, but showed no improvement over a previously described unconstrained analogue (E, 1). Surprisingly, no significant difference was found in the thermodynamic parameters (ΔH°, ΔS° and ΔG°) for PNA-DNA triplex formation involving 2 or the unconstrained analogue 1.

Full text of DOI:10.1016/j.bmcl.2003.12.093

Bioorganic & Medicinal Chemistry Letters 14 (2004) 1551–1554
Synthesis and evaluation of a conformationally constrained
pyridazinone PNA-monomer for recognition of thymine in
triple-helix structures
Anne Goldbech Olsen,a Otto Dahla and Peter E. Nielsenb,*
aDepartment of Chemistry, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen ø, Denmark
bDepartment of Medical Biochemistry and Genetics, The Panum Institute, University of Copenhagen, Blegdamsvej 3,
DK-2200 Copenhagen N, Denmark
Received 5 November 2003; revised 23 December 2003; accepted 23 December 2003
Abstract—A novel conformationally constrained pyridazinone Eag-base PNA-monomer 2 capable of binding thymine in a triplex
motif was designed and synthesised. A bis-PNA with the Eag-base incorporated in the Hoogsteen strand was hybridised with a
complementary DNA. Thermal stability studies revealed an increase in Tm (4.3 ꢀC per mod.) compared to a no-base unit, but
showed no improvement over a previously described unconstrained analogue (E, 1). Surprisingly, no significant difference was
found in the thermodynamic parameters (ꢀHꢀ, ꢀSꢀ and ꢀGꢀ) for PNA–DNA triplex formation involving 2 or the unconstrained
analogue 1.
# 2004 Elsevier Ltd. All rights reserved.
Sequence specific triplex targeting of double stranded
DNA in mixed purine/pyrimidine targets remains a
challenge in bioorganic chemistry. Due to the asym-
metric nature of the triple helix only a purine strand is
recognised through Hoogsteen base pairing.1 To expand
the triplex recognition repertoire of nucleic acids for
gene targeting purposes, novel nucleobases that recog-
nise the pyrimidine part of the Watson–Crick base pairs
C-(G) and T-(A) are required. Nucleobases that to some
extent recognise cytosine have been described.1,2 Recent
approaches to the recognition of thymine relies on
hydrogen bond contacts to both bases of the A–T
pair.3,4 The previously described E-base PNA-monomer
15 (Fig. 1) is able to recognise T when incorporated into
the Hoogsteen strand of a bis-PNA. To reach the O4 of
thymine in the major groove this monomer contains a
linker that is two atoms longer compared to conven-
tional PNA-monomers. Also, an extra nitrogen was
build into the base in order to minimise steric clash with
and/or steric clash with the 5-methyl groupof thymine.
The latter possibility is based on the observation that
E-base PNAs bind somewhat stronger to uracil- than to
thymine-containing targets.5 In order to address the issue
of flexibility, we have now synthesised a new analogue 2
(Fig. 1). This analogue contains a double bond that
should significantly restrict the flexibility in the linker,
and thereby reduce the added loss of entropy associated
with triplex formation.
Furthermore, the double bond might yield enthalpic
stabilisation through stacking interactions with neigh-
bouring base pairs.
5
the 5-methyl groupof thymine. However, the binding
affinity is far from optimal which might partly be due to
excessive flexibility of the backbone-‘nucleobase’ linker
Keywords: DNA recognitition; Triplex; Nucleobase; Peptide nucleic
acid (PNA).
* Corresponding author. Tel.: +45-353-27762; fax: +45-353-96042;
e-mail: pen@imbg.ku.dk
Figure 1. Structures of E-base PNA monomer (1) and Eag-base PNA
monomer (2).
0960-894X/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.12.093

1552
A. G. Olsen et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1551–1554
Figure 2. Proposed structure of the Eag/E T-A triplet.
.
The specific recognition of thymine (uracil) is expected
to be identical for both monomers, through an NH
hydrogen bond to the 4-oxo groupof thymine ( Fig. 2).
The structural differences of the linker in 1 and 2 lead to
the implementation of a distinct synthetic strategy. In
order to prepare the conformationally constrained
PNA-monomer 13 (a protected 2) commercially avail-
able dimethyl 2-oxoglutarate (3) was chosen as starting
material. Several approaches where examined before the
successful route depicted in Scheme 1 was found. Thus a
direct route via 4 to a suitable substituted pyridazinone
6 failed, the main product being the pyrazoline 56
instead. In the successful route 3 was first transformed
to the known dihydropyridazinone 77 using hydrazine
with acid catalysis in refluxing methanol. The next steps
were initially tried without a N-1 protecting group, but
due to very low solubility of later compounds, especially
the unprotected acid corresponding to compound 11,
this strategy was abandoned. To improve solubility,
compound 7 was reacted with 4-methoxybenzyl chloride
and sodium hydride in DMF to obtain the PMB protected
product 8. Reduction of 8 using sodium borohydride in a
refluxing mixture of THF and methanol gave the primary
alcohol 9. Treatment of 9 with activated manganese(IV)
oxide in refluxing toluene afforded a tandem reaction
where the primary alcohol was first oxidised to the
aldehyde and next an oxidative aromatisation took
place to give compound 10. This reduction–oxidation
strategy to obtain the aldehyde was chosen since an
earlier attempt with aromatisation using CuCl2 followed
by Dibal-H reduction of the ester was unsuccessful. The
next step, a Wittig reaction between aldehyde 10 and
2-carboxyethyltriphenylphosphonium bromide,8 took
place in a 1:1 mixture of THF and DMSO in the pre-
sence of sodium hydride. Only the desired E-isomer 11
was obtained, as verified by a typical trans CH-CH
coupling constant of 16.3 Hz. Condensation of 11 with
N-(2-Bocaminoethyl)-glycine methyl ester9 using DCC
and DHbtOH afforded the ester 12. Finally, hydrolysis
of the ester with NaOH in methanol resulted in the
monomer 13.10,11
Scheme 1. Reagents, conditions and yields: (i) ref 12 (2 steps); (ii)
NH2NH2, AcOH, MeOH, reflux, 12 h (42%, 3 steps); (iii) NH2NH2,
AcOH, MeOH, reflux, 12 h (92%); (iv) 4-methoxybenzyl chloride,
NaH, DMF, 0 ꢀC, 1 h then rt, 12 h (80%); (v) NaBH4, THF, MeOH,
reflux, 1 h (58%); (vi) activated MnO2, toluene, reflux, 24 h (30%);
(vii) Ph3P+(CH2)2COOH Brꢁ, NaH, THF/DMSO, 0 ꢀC, 1 h then rt, 4
h (42%); (viii) methyl N-(2-Boc-aminoethyl) glycinate, DhbtOH,
DCC, DMF, 0 ꢀC, 1.5 h then rt, 12 h (50%); (ix) 2M NaOH, MeOH,
0 ꢀC, 2.5 h (46%).
the Watson–Crick strand (Table 1) and therefore facing
thymine in the DNA target. The remaining positions
.
.
consisted of conventional T A-T and J G-C triplets. The
ability of the Eag-base to recognise thymine was eval-
uated by incorporation of a no-base PNA monomer5 or
the original E-base monomer in the same position for
comparison. The thermal stability of the complexes
formed between these bis-PNAs and complementary
DNA was analysed by thermal denaturation. The
temperature versus absorbance profiles showed mono-
phasic, well defined transitions, from which the Tm were
obtained. The results summarised in Table 1 show sig-
nificant stabilisation when the Eag-base rather than a
no-base position faces thymine (ꢀTm=+4.3 ꢀC p er
mod.), but the triplex is less stable than when using the
original E-base (ꢀTm=ꢁ1.5 ꢀC per mod.).
The monomer 13 could be incorporated into PNA oli-
gomers under standard solid phase conditions,13 and
removal of the p-methoxybenzyl protecting group took
place under these conditions. The Eag base was eval-
uated in a bis-PNA14 recognition system designed to
recognise a single stranded DNA oligonucleotide target
by Watson–Crick and Hoogsteen base pairing (Fig. 3).
The Eag base was incorporated in the Hoogsteen strand
of a bis-PNA at two positions opposite an adenine in
The nucleobase discrimination of Eag was studied by
introduction of a double C-mismatch as presented in

A. G. Olsen et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1551–1554
1553
.
Table 1. Thermal stability (Tm) of PNA DNA–PNA complexes using
Table 4. Thermodynamic data for the bis-PNA–DNA complexes
with Eag and E respectively. The data are based on three independent
experiments each including measurement of Tm at ten different com-
plex concentrations (SEM is indicated in parenthesis). ꢀG0 was cal-
culated from the ꢀH0 and ꢀS0 values obtained from the 1/Tm versus
lnCtot plot
a bis-PNA containing two no-bases, E-bases or Eag-bases respectively
hybridised to a DNA complement 50-dCGCAGATAGTAAACGC-30
containing the 10-mer target
Bis-PNA sequencea
Tm
(ꢀC)c
H-TJTEagTJEagTTTeg1eg1
eg1TTTACTATCT-NH2
H-TJTETJETTTeg1eg1
eg1TTTACTATCT-NH2
b
H-TJTaeg(Ac)TJaeg(Ac)TTTeg1eg1eg1TTTACTATCT-NH2
H-TJTEagTJEagTTTeg1eg1eg1TTTACTATCT-NH2
H-TJTETJE-TTTeg1eg1eg1TTTACTATCT-NH2
47.5
56.1
59.1
ꢀHo
ꢀSo
ꢀGo
ꢁ706 (Æ34) kJ/mol
ꢁ2037 (Æ102) J/mol
ꢁ74.0 (Æ2.0) kJ/mol
ꢁ722 (Æ96) kJ/mol
ꢁ2058 (Æ288) J/mol
ꢁ83.4 (Æ6.6) kJ/mol
a Three units of 8-amino-3,6-dioxaoctanoic acid (eg1) connect the two
antiparallel 10-mer PNA strands. Pseudoisocytosine (J) was used
instead of protonated cytosine in the Hoogsteen strand of the
bis-PNA.
b aeg (Ac) is N-acetyl-N-(2-aminoethyl) glycine, a no-base PNA unit.
c All Tm measurements in this paper were performed in 100 mM
NaCl, 0.1 mM EDTA, 10 mM Na2HPO4, pH 7.0 on a Cary 300Bio
spectrophotometer at a heating rate of 0.5 ꢀC/min.
As expected from the Tm data, these results (Table 4)
showed slightly lower free energy for the bis-PNA com-
plex with the E-base, but most surprisingly no significant
differences between the two monomers in the relative
values of the thermodynamics parameters (ꢀH0, ꢀS0 or
ꢀG0) could be detected. In particular the entropy loss
was identical. Therefore, introduction of the more rigid
Eag-base did not result in an entropic advantage.
Figure 3. Structure of the bis-PNA–DNA complex.
The results reported herein could be used as guidelines
for design of future analogues. Contrary to expectation
PNA clamps containing the new monomer 2 performed
poorer in DNA hybridisation assays than the fully
flexible analogue, and as concluded from thermo-
dynamic measurements, it is especially noteworthy that
the entropy loss upon triplex formation is virtually
identical for the two PNAs. Therefore, the higher flex-
ibility of the E-base cannot be the reason for the rela-
tively weak binding to thymine. This could indicate that
local flexibility has only little influence on overall binding
thermodynamics of oligonucleotide hybridisation, and
that the entropy term is predominantly determined by
overall flexibility of the oligomer in combination with
contributions from changes in solvation.
Tables 2 and 3. These results demonstrate a similar ꢀTm
discrimination against cytosine as the original E base
(15 and 14 ꢀC, respectively). However, the recognition
of U was different. Whereas the E-base preferred bind-
ing to U over T (ꢀTm=+3.5 ꢀC per mod.), the new
Eag-base showed a minor preference of binding to T
compared to U (ꢀTm=+0.5 ꢀC). This suggests the
absence of significant steric clash between the Eag-base
and the 5-methyl groupof thymine.
To gain further insight into the binding behaviour, the
thermal stabilities were measured at different con-
centrations in order to obtain thermodynamic data for
the bis-PNA–DNA complexes according to the follow-
ing equation.15
Acknowledgements
1=Tm ¼ ðR=DH0ÞlnCtot þ ðDS0 ꢀ Rln4Þ=DH0
ð1Þ
This work was supported by The Danish Research
Agency (Center for Organic Chemical Biotechnology)
and the European commission (contract no QLK3-CT-
2000-001658). Ms. Jolanta Ludvigsen and Ms. Annette
Jørgensen are thanked for expert technical assistance.
.
Table 2. Effect of target base changes on PNA DNA–PNA
complexes using the following bis PNA: H-TJTEagTJEagTTT-eg1e-
g1eg1-TTTACTATCT-NH2
Oligonucleotide target sequence
Tm (ꢀC)
ꢀTm/mod. (ꢀC)
References and notes
50-dCGCAGATAGTAAACGC-30
50-dCGCAGAUAGUAAACGC-30
50-dCGCAGACAGCAAACGC-30
56.1
55.0
26.0
ꢁ0.5
1. Fox, K. R. Curr. Med. Chem. 2000, 7, 17.
2. Prevot-Halter, I.; Leumann, C. J. Bioorg. Med. Chem.
Lett. 1999, 9, 2657.
ꢁ15.0
3. Guianvarc’h, D.; Benhida, R.; Fourrey, J.-L.; Maurisse,
R.; Sun, J.-S. Chem. Commun. 2001, 1814.
4. Guianvarc’h, D.; Fourrey, J.-L.; Maurisse, R.; Sun, J.-S.;
Benhida, R. Org. Lett. 2002, 4, 4209.
5. Eldrup, A. B.; Dahl, O.; Nielsen, P. E. J. Am. Chem. Soc.
1997, 119, 11116.
6. Brey, W. S.; Valencia, C. M. Can. J. Chem. 1968, 46, 810.
7. Teotino, U. M.; Cignarella, G. Gazz. Chim. Ital. 1959, 89,
1200.
8. Wakita, H.; Yoshiwara, H.; Nishiyama, H.; Nagase, H.
Heterocycles 2000, 53, 1085.
.
Table 3. Effect of target base changes on PNA DNA–PNA
complexes using the following bis PNA: H-TJTETJETTT-eg1eg1eg1-
TTTACTATCT-NH2
Oligonucleotide target sequence
Tm (ꢀC)
ꢀTm/mod. (ꢀC)
50-dCGCAGATAGTAAACGC-30
50-dCGCAGAUAGUAAACGC-30
50-dCGCAGACAGCAAACGC-30
59.1
66.2
31.1
+3.5
ꢁ14.0

1554
A. G. Olsen et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1551–1554
9. Dueholm, K. L.; Egholm, M.; Buchardt, O. Org. Prep.
Proc. Int. 1993, 25, 457.
(4-methoxybenzyl)-6-oxo-1,6-dihydropyridazin-3-yl]-but-3-
enoyl} glycine methyl ester (12): This compound exist as
two rotamers; chemical shifts for the minor rotamer are
given in brackets. 1H NMR (CD3OD) d 7.73 (7.72) (d,
1H, J 9.7 Hz, arom), 7.30 (d, 2H, J=6.8 Hz, PMB), 6.94
(d, 1H, J=9.5 Hz, arom), 6.85 (d, 2H, J=6.8 Hz, PMB),
6.50 (m, 2H, 2ÂCH¼), 5.22 (s, 2H, CH2-benzyl), (4.29),
4.11 (s, 2H, CH2), 3.76 (s, 3H, CH3O), (3.71), (s, 3H,
CH3) 3.52 (t, 2H, J=6.2 Hz, CH2), 3.47 (d, 2H, J=5.1
Hz, CH2), 3.25 (3.19), (t, 2H, J=6.2 Hz, CH2), 1.42,
(1.40) (s, 9H, Boc). TOF-MS-ES+ m/z 515.3 (M+H+).
N-(2-Bocaminoethyl)-N-{4-[1-(4-methoxybenzyl)-6-oxo-1,6-
dihydropyridazin-3-yl]-but-3-enoyl} glycine (13): This
compound exists as two rotamers; chemical shifts for the
10. Part of this work has been presented as a poster at the XV
International Round Table conference on ‘Nucleosides,
Nucleotides and Nucleic Acids’, Leuven, Belgium, Sep-
tember 2002: Olsen, A. G.; Dahl, O.; Nielsen, P. E.
Nucleosides, Nucleotides and Nucleic Acids 2003, 22, 1331.
11. Selected data for new compounds: 1-(4-methoxybenzyl)-6-
oxo-1,4,5,6,-tetrahydropyridazine-3-carboxylic
acid
1
methyl ester (8): H NMR (CDCl3) d 7.22 (d, 2H, J=6.6
Hz, PMB), 6.75 (d, 2H, J=6.6 Hz, PMB), 4.86 (s, 2H,
CH2-benzyl), 3.78 (s, 3H), 3.69 (s, 3H) (2ÂCH3O), 2.77 (t,
2H, J=8.8 Hz, CH2), 2.45 (t, 2H, J=8.8 Hz, CH2). FAB-
MS m/z 277 (M+H+). 1-(4-methoxybenzyl)-6-oxo-
1,4,5,6,-tetrahydropyridazine-3-methanol (9): 1H NMR
(CD3OD) d 7.27 (d, 2H, J=6.6 Hz, PMB), 6.87 (d, 2H,
J=6.6 Hz, PMB), 4.86 (s, 2H, CH2-benzyl), 4.20 (s, 2H,
CH2OH), 3.79 (s, 3H, CH3O), 2.63 (m, 2H, CH2), 2.53 (m,
2H, CH2). TOF-HRMS-ES+ m/z calcd for C13H17N2O3,
249.1239, found: 249.1232. 1-(4-Methoxybenzyl)-6-oxo-
1,6-dihydropyridazine-3-carbaldehyde (10): 1H NMR
(CDCl3) d 9.66 (s, 1H, CHO), 7.62 (d, 1H, J=9.5 Hz,
arom), 7.35 (d, 2H, J=6.6 Hz, PMB), 6.87 (d, 1H, J=9.5
Hz, arom), 6.78 (d, 2H, J=6.6 Hz, PMB), 5.26 (s, 2H,
CH2-benzyl), 3.70 (s, 3H, CH3O). FAB-MS m/z 245
(M+H+). E-4-[1-(4-Methoxybenzyl)-6-oxo-1,6-dihydro-
pyridazin-3-yl]-but-3-enoic acid (11): 1H NMR
(CD3OD) d 7.72 (d, 1H, J=9.7 Hz, arom), 7.34 (d, 2H,
J=6.8 Hz, PMB), 6.95 (d, 1H, J=9.7 Hz, arom), 6.88 (d,
2H, J=6.8 Hz, PMB), 6.56 (dt, 1H, J=16.3, 6.8 Hz,
CH¼), 6.47 (d, 1H, J=16.3 Hz, CH¼), 5.24 (s, 2H, CH2-
benzyl), 3.77 (s, 3H, CH3O), 3.31 (d, 2H, J=6.8, 1.1 Hz,
CH2). TOF-HRMS-ES+ m/z calcd for C16H17N2O4:
301.1188, found: 301.1201. N-(2-Bocaminoethyl)-N-{4-[1-
1
minor rotamer are given in brackets. H NMR (DMSO-
d6) d (7.77), 7.76 (d, 1H, J=9.7 Hz, arom), 7.27–7.18 (m,
1H, BocNH), 7.24 (d, 2H, J=8.8 Hz, PMB), (6.93), 6.92
(d, 1H, J=9.7 Hz, arom), 6.88 (d, 2H, J=8.8 Hz, PMB),
6.53 (m, 1H, CH=), (6.34) 6.25 (d, 1H, J=16.3 Hz,
CH=), 5.12 (s, 2H, CH2-benzyl), 3.71 (s, 3H, CH3O),
(3.61), 3.60 (s, 2H, CH2), 3.30 (m, 2H, CH2), 3.18 (d, 2H,
J=6.8 Hz, CH2), 3.04 (m, 2H, CH2), (1.36) 1.34 (s, 9H,
Boc), TOF-HRMS-ES+ m/z calcd for C25H33N4O7:
501.2349, found: 501.2348.
12. Corey, L.; Tramontano, A. J. Am. Chem. Soc. 1981, 103,
5599.
13. Christensen, L.; Fitzpatrick, R.; Gildea, B.; Petersen,
K. H.; Hansen, H. F.; Koch, T.; Egholm, M.; Buchardt,
O.; Nielsen, P. E.; Coull, J.; Berg, R. H. J. Peptide Sci.
1995, 1, 175.
14. Egholm, M.; Christensen, L.; Dueholm, K. L.; Buchardt,
O.; Coull, J.; Nielsen, P. E. Nucleic Acids Res. 1995, 23,
217.
15. Marky, L. A.; Breslauer, K. J. Biopolymers 1987, 26, 1601.